27th October 2022

How ‘legal avoidance’ means getting away with tax fraud, and what needs to be done about it

A joint policy paper by the All Party Parliamentary Group on Anti-Corruption and Responsible Tax and TaxWatch

Introduction

Tax avoidance is legal, tax evasion is illegal. That clear and unambiguous dividing line has been the foundation on which an entire industry of tax avoidance has been built. The reality is however that this celebrated distinction is a myth, fostered by the tax industry that profits from it and left undisturbed by the public body that knows better: HMRC. In this report we demonstrate that much that is considered ‘legal’ tax avoidance (or at worst, a failed but lawful attempt to avoid tax) could in fact potentially be subject to a criminal prosecution. And we show that, in so-called ‘legal’ avoidance cases where in fact fraud is suspected, even if there are millions or perhaps billions of pounds in tax at stake, HMRC choose not to pursue criminal investigations. These points are illustrated by means of a series of case studies appended to the report.

Given the controversial nature of this report it is important to be clear about what we mean. The most serious tax fraud in the UK is prosecuted under the criminal offence of ‘cheating the public revenue’ and, at the heart of that offence, lies the question of whether the defendant was being dishonest in their actions. A determination as to whether or not any particular individual was dishonest, so as to be guilty of that offence, can only be arrived at by a jury after the defendant has had the opportunity to present their case. Accordingly, in this paper, we are not seeking to establish whether any particular individual has acted dishonestly. What we are saying, however, is that based on the extensive evidence in the public domain on these avoidance schemes, the conduct of those involved in them could appear dishonest to a jury, and it would therefore be open to HMRC to investigate the schemes with a view to criminal prosecution.

The cases we consider are not unknown to HMRC. In all cases they involve well-known companies or schemes. The facts we have based our conclusions on are largely drawn from legal documents, including cases litigated by HMRC under civil as opposed to criminal law. The question therefore arises, why has HMRC not pursued these schemes under their powers of criminal investigation? The answer is straightforward. It has been the long-standing policy of HMRC (and the Inland Revenue before it) to not pursue the vast majority of tax fraud cases under the criminal law, and instead use civil procedures to recover any tax lost.

There are practical reasons for this approach. HMRC’s primary role is as a tax collector and in most cases it will be cheaper and quicker to use civil procedures to reclaim tax lost from tax avoiders than to pursue criminal charges. This has the perverse effect, however, of setting standards of behaviour in the tax industry far below the standards imposed by existing criminal law. Certain types of fraud (usually involving the creation of false documentation, or non-disclosure of relevant information) are seen as unacceptable, whereas other forms of fraud – for example basing a purported tax saving on an economic fiction – are almost invariably treated as acceptable. This is despite the fact that under the criminal law, there is no distinction between these two different types of fraud. The distinction lies only in what HMRC acting in accordance with their policies will, and will not, investigate as a criminal matter.

It is entirely appropriate that policymakers and legislators should question whether HMRC’s policy to not enforce the criminal law in large numbers of cases of tax fraud is the right approach. The last time that Parliament took an interest in the enforcement powers of the revenue departments was in the early 1980s with the publication of the Keith Committee reports. In the interim, the issue of tax avoidance has become an issue of huge public concern and, to a large degree, the public debate on the issue has proceeded on a false premise: the idea that all tax avoidance is by definition a lawful activity. This premise, propagated by the media, by the tax industry, and by academics, all with the complicity of HMRC, has undermined the ability of parliament and through it the Government to grapple with the issue.

We hope that this report, by setting out clearly the facts and the law in relation to tax avoidance, and setting this against HMRC’s practice, can facilitate a proper debate on this important subject.

Recommendations

If we were to sum it up in a single sentence the recommendation of this report it would be this: HMRC should be enforcing the law of the land, not the ‘rules of the game’. To do this HMRC needs to take a far tougher approach when it comes to the promoters and enablers of tax avoidance schemes, using the powers it has under the criminal law to bring forward more prosecutions of the enablers of tax crime. Calls for HMRC to do more to tackle tax crime have been made before. In 2016, the Public Accounts Committee called on HMRC to do more to tackle tax fraud, and yet the number of prosecutions has since that time continued to decline. Now is the time for concrete action.

We recommend that HMRC officers should be required by law to consider for separate investigation and potential prosecution the promoters and enablers involved in tax avoidance arrangement. The case should then be referred for prosecution unless a determination is made that a successful prosecution would be unlikely or contrary to the public interest. Further, any civil settlement reached between HMRC and a taxpayer should be conditional on a requirement on the taxpayer to co-operate with any future criminal investigation into their advisers.

The legal framework

In assessing their tax own liabilities, it is possible that an individual or a company will take a different view of what their obligations should be than HMRC. If HMRC disagree with the position of the taxpayer, they can issue their own assessment of the amount of tax due. The taxpayer can then challenge this assessment at a tax tribunal. The purpose of a tribunal is to determine the right amount of tax that should be paid, in accordance with the applicable tax legislation. Crucially, this is a wholly civil mechanism. It is not the tax tribunal’s job to determine whether the taxpayer’s behaviour was dishonest, fraudulent, or otherwise criminal. This means that the question of whether tax fraud has taken place or not is very often not considered by a court, even where HMRC has had to take formal steps to recover unpaid tax.

The central feature of tax fraud is that it involves dishonest behaviour. The common law offence of cheating the public revenue is the primary offence that tax fraud comes under, although there are other offences where dishonest tax conduct is outlawed. Cheating the public revenue is an extremely broadly-defined offence. In short, anyone that engages in any form of dishonest conduct that risks prejudice to the public revenue can be tried on indictment with the potential sentence of life imprisonment. There is no need for the prosecution to prove that the defendant profited from their actions, or even that the revenue suffered a loss. As the leading textbook on criminal law, Smith, Hogan and Ormerod explains, the breadth of the offence means that often the only live issue at trial will be dishonesty.¹ The offence can also be charged as conspiracy to cheat the public revenue, further widening the scope of the offence to anyone involved in the creation or operation of a dishonest tax scheme e.g. professional advisers, promoters and so on. Any perception that the offence is somehow intended for or targeted specifically at the taxpayer is mistaken.

Crucially, under the criminal law, what constitutes dishonesty is to be determined by a jury applying the standards of ordinary decent people – not a group of tax experts. It is perfectly possible that a tax scheme which is dealt with through the civil process described above could be one which a jury in a criminal trial would consider to be dishonest. So-called ‘legal’ tax avoidance and tax fraud should therefore be treated as categories which substantially overlap. The question of which process is followed in which case is therefore often a discretionary one for HMRC, rather than being driven by whether or not a crime was committed.

That discretion is a broad one. While HMRC has an extensive range of investigatory and law enforcement powers, including powers of criminal investigation broadly similar to those of the police,² its primary role is to obtain ‘the highest net return that is practicable having regard to the staff available […] and the cost of collection’.³ Accordingly, HMRC’s function as the body responsible for investigating crimes committed in connection with the taxes under its charge is ‘ancillary to, supportive of and limited by’ that duty to maximise revenue.⁴ This means that there is no obligation on HMRC to investigate any incidence of fraud as a criminal matter. Instead it is open to the department to pursue an investigation through an entirely civil process. How it exercises that discretion is (aside from the exceptional circumstances where judicial review might be appropriate) entirely a matter of policy for HMRC.

HMRC policy

HMRC’s civil fraud investigations – COP 8 and COP 9

HMRC’s Fraud Investigation Service operates two civil fraud investigation procedures, referred to as Code of Practice 8 or Code of Practice 9 (‘COP 8’ and ‘COP 9’).⁵ Both COP 8 and COP 9 are civil processes where the goal is to reach a private agreement with the taxpayer to settle any taxes due and impose civil penalties where appropriate.

Under COP8, a taxpayer is invited to meet with HMRC and disclose all relevant facts relating to the issue which HMRC wish to enquire into. In a COP 8 investigation, there is no explicit allegation of fraud that is made against the taxpayer. The purpose of the COP 8 process is to allow HMRC gather as much information as possible in order to make a correct assessment of any tax due. Under COP9, by contrast, there is an explicit allegation of fraud, and in return for admitting to fraudulent conduct HMRC will agree not to pursue a criminal prosecution. Instead, a contractual settlement will be reached where the taxpayer agrees to provide a full disclosure of their assets and pay any taxes and penalties due. In these processes the possibility of criminal prosecution is explicitly used as a threat to encourage the taxpayer to make a full disclosure of all relevant information.

It is a serious (but widespread) error, to infer that HMRC’s classification as between these two procedural pathways serves to distinguish between ‘legal’ tax avoidance on the one hand and, on the other, unlawful tax fraud handled through civil enforcement mechanisms. In fact the determination of whether to follow COP 8 or 9 is a tactical one taken right at the outset of the investigation rather than a forensic one taken in full view of the evidence.

HMRC’s criminal investigations policy

Under the Police and Criminal Evidence Act, HMRC can conduct criminal investigations with a view to prosecution by the Crown Prosecution Service or the equivalent bodies in Scotland and Northern Ireland. HMRC’s Criminal Investigations Policy⁶ sets out when the department will consider using its powers of criminal investigation. It makes clear that in the majority of tax fraud cases, the preference will be to use HMRC’s civil fraud procedures outlined above, stating that it is ‘HMRC’s policy to deal with fraud by use of the cost-effective civil fraud investigation procedures under Code of Practice 9 wherever appropriate’. A criminal investigation, in contrast to those civil processes, is resource-intensive, and provides the taxpayer with little incentive to be cooperative with regard to the information HMRC requires in order to be able to recover all the tax that is due.

There are some cases of criminality where HMRC will ‘consider’ going straight to a criminal investigation. These include cases involving money laundering or organised crime, cases where the taxpayer is in a position of trust or responsibility, and cases which involve active deception. As HMRC characterise this latter category, they will tend to consider prosecution where there is ‘any deliberate omission, concealment or misinterpretation of information, or the false or deceptive presentation of information or circumstances in order to gain a tax advantage’. HMRC illustrate this category with examples of what they will consider to be tax fraud i.e. deliberately submitting false tax returns, falsely claiming repayments or reliefs, hiding income, gains or wealth offshore, and smuggling taxable goods. These forms of active deception are indeed, needless to say, examples of behaviour which a jury might consider to be dishonest. They are, however, by no means the only forms of dishonesty that occur in a tax context. We consider certain others below.

It should be noted that, in addition to those categories of cases, HMRC will occasionally use their powers of criminal prosecution to serve make an example out of somebody, to remind the public that those powers exist. It is for this reason that in the past HMRC pursued criminal convictions of celebrities, due to the high profile these cases would attract. However, this strategy also brought with it the risk of a high-profile acquittal, as happened in the prosecutions for tax offences of Ken Dodd and Harry Redknapp.⁷ Ordinarily, however, absent aggravating features such as evidence of active deception, a person whose tax behaviour might amount to a crime is highly likely to face no more serious sanction than (i) a requirement under threat of criminal investigation to make full disclosure, and (ii) civil penalties.

Criminal conduct in cases of ‘legal avoidance’

Criminal conduct on the part of the taxpayer

Tax avoidance typically begins with a tax adviser (or sometimes, in the case of a big business, an in-house tax function) creating a scheme. This is a kind of blueprint or plan setting out a number of steps which, when implemented, will make it appear that the taxpayer has a tax position which does not correspond to their economic reality. For example, a scheme may make it appear that a taxpayer has suffered a loss or incurred an expense that entitles them to a tax benefit, when in fact no such thing has happened. Alternatively, a series of contracts can be constructed to make it appear that income in the hands of an employee or contractor is something else – a loan from a third party, say – when in reality there is no intention on anyone’s part for the loan to be repaid. In the corporate world, contracts can be constructed that mean that a transaction between group companies can be significantly overvalued or undervalued, with the consequence that the real value ends up in a tax haven. These are the kinds of arrangements generally treated as ‘legal avoidance’ rather than fraud. These transactions are professionally designed to misrepresent to the tax authority the true economic reality of the taxpayer, leading to a loss of tax revenue.

Owing to HMRC’s policies, juries very rarely get to consider these kinds of schemes. Although these schemes generally rely on misrepresenting the economic substance of an arrangement, there is generally no active deception of the kind that HMRC’s criminal investigations policy focuses on – it is not necessary to (for example) falsify documents or hide sums of money in order to purport to obtain these tax advantages. Ordinary reasonable people, however (i.e. the members of a jury, who are not steeped in the self-justifications popular among tax professionals) are likely to think that it is dishonest to artificially create tax losses attributable to a business activity that isn’t a real business, whether or not the losses are inflated by false documentation.

Indeed, as the case studies appended to this report demonstrate, when juries do get to consider these schemes, they often do see dishonesty in them. These are scenarios which the tax industry would normally indignantly defend as ‘legal’ and which HMRC would normally choose to deal with through the processes of civil assessment to tax. But where HMRC has (exceptionally) deemed it fit to pursue the matter as a criminal one, the prosecution will positively encourage the jury to see those same kinds of fact patterns, normally treated as ‘legal’, as evidence of dishonesty.

On these exceptional occasions where ‘legal avoidance’ is prosecuted as tax fraud it is very likely to be advisers, enablers or promoters who are in the dock, as opposed to taxpayers, and with good reason – the taxpayer can say by way of defence that they were professionally advised that the scheme would be effective. This defence is not necessarily absolutely watertight, but in most cases of so-called ‘legal’ avoidance any criminal culpability will be with the advisers and enablers.

Criminal conduct on the part of professionals, advisers and enablers

The features of tax avoidance schemes discussed in the previous section – i.e. that they may be dishonest whether or not they involve ‘active deception’ – could all be valid grounds for a criminal investigation into the actions of the advisers and enablers, and those participants lack the excuse that they knew no better and were acting under advice – it was they who were giving the advice! But there is more to their potential liability than that.

For example, another key feature of avoidance schemes is that, when they are presented to the client (or to management, in the case of an in-house scheme), they are almost invariably accompanied by professional advice about the effectiveness of the scheme. This advice may be wildly optimistic because the scheme is obviously ineffective as means of reducing tax and would be highly likely to fail any scrutiny by HMRC or the tax tribunal. It may therefore be that a jury would consider the advice to be dishonest – the adviser has promoted a scheme which they know will not ultimately work. Again, this category of dishonesty is highly unlikely to make it as far as a jury. Indeed, despite the pageant of self-evidently preposterous arguments that have failed at tribunal as a matter of legal analysis in avoidance cases, resulting in tax being payable after all, it is almost unheard of for the adviser giving the original favourable opinion (often a QC) to have faced prosecution for that aspect of their role in the scheme. And this is despite the fact that QCs in this field of practice are notorious for giving opinions which do not withstand the scrutiny of the courts.⁸

In addition to the QC giving the opinion, there will generally be a variety of other professionals involved: solicitors and accountants for example. These professionals will generally be aware from the QC’s opinion that, in order for the scheme to be effective, certain steps have to be taken. For example, it may be that activities which are said to be taking place on paper must actually take place in the real world. Similarly, in order for the scheme to be effective, basic implementation must be effective – for example formal documents must be properly executed in the right order and so on. Very often these steps are not taken at all, or are taken negligently, and the scheme could fail by reason of inadequate implementation as much as by reason of the legal analysis being wrong. In these circumstances, continuing to advise that the taxpayer may file a tax return claiming the saving may well be a dishonest act. This is because the advice would be given in the knowledge that the tax benefit is not available in any event, whether or not the legal analysis is valid. A jury may well conclude that this is dishonest conduct, and yet in the overwhelming majority of such cases, even if the implementation is sloppy, HMRC will treat the matter as ‘legal avoidance’.

In view of all these potential routes to a conviction vis-à-vis the enablers and advisers, it is worth noting that a prosecuting authority is likely to have a wealth of evidence to support any such approach. Following a full disclosure by a taxpayer to HMRC under either COP 8 or COP 9, there should be ample evidence to open a criminal investigation into the professionals, enablers and advisers involved. Having collected any tax lost to the avoidance scheme from the taxpayer, however, there is little financial incentive for HMRC to pursue those other participants, and no statutory or policy requirement for them to do so either.

The tax fraud game

The rules of the game

HMRC’s highly selective enforcement of fraudulent tax conduct, and in particular the more-or-less exclusive focus on what we here describe as ‘active deception’, creates a wide arena of possibility for unscrupulous tax advisers. It is possible for them to design schemes that probably do not work, promote them to users, and see them through to implementation, without significant risk of criminal prosecution either of themselves or users, irrespective of the degree of dishonesty at play. To minimise risk of prosecution, all they need to do is comply with what we here describe as the ‘rules of the game’. To be clear, this is not a formal set of rules which HMRC operate as a parallel regime alongside the strict legal one they are meant to be enforcing – but that is the practical effect of their policy. Broadly-speaking the rules of this game are as follows:

1. A tax avoidance scheme should be created by qualified lawyers and accountants, executing genuine documents. The economic position being claimed may well be a fiction, but it must at least purport to exist on paper.
2. Secondly, there should be no ‘active deception’ in the implementation or reporting.
3. Finally, once the scheme is uncovered by HMRC or investigated, the taxpayer should disclose everything.

If these rules are followed, no matter how dishonest the scheme is, whether on the part of the taxpayer, or the advisers, or both, they are almost certainly safe from prosecution.

There are any number of ways that such a scheme can succeed by default even if the legal arguments are invalid. It is possible that HMRC will not detect the scheme,⁹ or, if they do, not fully understand it, resulting in the taxpayer’s position being accepted. Perhaps they will make a procedural error, allowing the taxpayer to challenge HMRC’s assessment on a technicality – there are lawyers that dedicate their careers to making procedural challenges to HMRC’s attempts to counter tax avoidance. Perhaps, faced with a barrage of legal firepower deployed by corporate opponents, HMRC will give up and settle for a significantly lower amount. Mistakes become more likely of course as HMRC’s resources become increasingly stretched. And there is always the possibility that the scheme, even if it ends up having the legal arguments tested at tribunal, will survive even that stage, deliver the tax advantage, and mark the taxpayer out as having achieved a ‘legal’ outcome, irrespective of any criminal dishonesty that may have taken place. In all of these ways, tax fraud may well have been committed, and HMRC may well know about it, but public money (in vast quantities) ends up in private hands nonetheless.

Accordingly, it adds insult to (fiscal) injury for there to be a widespread claim that, where HMRC choose not to enforce the criminal law, and challenge avoidance through civil means, or reach a settlement with the taxpayer, that means that what has been done is ‘legal’. This is a myth which has the effect of reinforcing the rules of the game at the expense of the actual law. A non-criminal outcome absolutely does not mean that in law these schemes are ‘legal’, much though the tax avoidance industry will insist that it does. On this point criminal law expert David Ormerod is authoritative:

The difficulty in distinguishing the shades of avoidance and evasion means that it is always possible for the Revenue to charge a defendant with cheating in respect of a scheme which is alleged to be dishonest evasion and which the (non)taxpayer believes to be, at worst, an ineffective avoidance scheme. Commentators on the decision in Charlton¹⁰ asked how what they perceived to be merely ineffective tax avoidance could be criminal. The criminal lawyer’s response to that is simple: the schemes might be classified as ineffective tax-avoidance in civil law, but that does not prevent them being criminal because the cheating offence is now so broad that it turns solely on the question of dishonesty.¹¹

The fact that tax crime is simply permitted to happen without enforcement measures being taken against it was recently set out in blunt terms by Janet Alexander, the head of HMRC’s Taxpayer Protection Task Force, the body set up to recover money stolen from HMRC-administered covid support schemes. ‘In the UK’, she said ‘we use civil powers to recover the monies, we don’t normally criminally prosecute – that is the way that we handle tax investigations in the UK. It doesn’t mean it’s not a fraud, it’s just not the way that we deal with it.’¹² The point we make in this report is that that needs to change.

Why this matters

Avoidance-based tax fraud continues to be endemic in the UK, as demonstrated by constant stream of avoidance cases before the tax tribunal. It is clear that there are significant numbers of practitioners that continue to design and market schemes, and senior lawyers willing to sign off on them. Civil processes for reclaiming the cash from their clients provide no disincentive for tax professionals to carry on playing this game. In fact, a recent analysis of HMRC’s published figures on tax avoidance schemes known as ‘disguised remuneration’ schemes shows that the numbers of people involved increased considerably after several legislative interventions were made to stop them.¹³ Furthermore, the prevalence of tax avoidance is not limited to individuals using schemes. As demonstrated by HMRC’s diverted profits tax compliance facility (an amnesty for large multinational companies involved in artificial schemes) big business continues to engage in avoidance.¹⁴ HMRC will say that the market has changed since the heyday of artificial tax structuring prior to certain legislative and cultural changes in the 2010s, and this is true to an extent, but the perverse incentives created by their policies remain.

Given HMRC’s primary focus on revenue collection, their strategy of extracting the full amount of revenue from a taxpayer as quickly and cheaply as possible through civil investigation procedures will always have an important role. However, it is our contention that HMRC’s strong preference for civil investigations creates several issues for the tax system that need to be addressed, and that the systematic failure to apply the criminal law to the enablers of tax fraud misses an opportunity to strike at the source of the problem. We break the issue down into four aspects. The first is deterrence, the second effectiveness, the third is fairness, and the fourth is justice and the rule of law.

(1) The issue of deterrence is clear. Tax advisers know that, provided they comply with the rules of the game, they can perpetrate criminal conspiracies to cheat the public revenue with effective impunity. HMRC’s deliberate removal of the deterrent effect of criminal prosecution for professional advisers has created a legal and moral vacuum that has allowed tax avoidance to flourish. Indeed the infrequency with which HMRC will seek to apply the criminal law to the enablers of tax avoidance has encouraged a culture in the tax industry and its supporters that maintains that tax avoidance is all perfectly ‘legal’, and that tax avoidance cases arise from a morally neutral dispute between the taxpayer and HMRC over the application of the law to the facts.

So powerful has this narrative become, that when tax campaigners draw attention to the moral outrage of tax avoidance, their protestations are dismissed by experts, and they are told that their outrage is down to their failure to understand how complicated tax rules work. In fact it is these so-called ‘experts’ who have failed to understand: specifically, they have mistaken the rules of the game for the law of the land. If a taxpayer and their advisers conduct themselves in a way that a jury would consider to be dishonest, then a fraud has most likely been committed, even if HMRC’s treatment of it and the tax industry’s claims about it are to the effect that what has happened is ‘legal’. Far from being the result of a failure to understand the law, the sense on the part of ordinary decent people that tax avoidance is a moral outrage might well accurately reflect the fact that a criminal offence has indeed been committed in the matter in question.

Apologists for the tax avoidance industry love to treat the question of whether ‘morality’ plays a role in tax as a debate, so that they can argue from their purported position of expertise that it does not. But in fact the pretence that this is even up for debate serves to reinforce a false narrative about how the law in this area works, thereby positively exacerbating the precise moral problem that they claim does not exist. By way of analogy, suppose a residential area where there is a 30 mph speed limit, but the speed cameras are set only to go off if people are driving at 67 mph. Suppose then that some drivers are speeding through the area at 58 mph and nothing is being done about it. Suppose further that the response of the motoring lobby, when people complain, is to say that there is no morality in the speed at which you drive through a residential area, and that the acceptability of driving at 58 mph is simply a legal question to be determined in motoring law enforcement proceedings. That is the situation we are in when people say that tax avoidance is perfectly legal. They are taking advantage of catastrophically weak enforcement to create a culture of impunity around criminal acts that adversely affect the rest of us.

(2) Secondly, it is not at all clear that the balance struck by HMRC between civil and criminal approaches is effective even on its own terms. There are cases where recourse to a criminal investigation early on would in fact have been more effective at collecting the tax. One powerful example of this is a case involving Rangers Football Club, which ended up dragging on through the civil courts for many years before the Supreme Court finally found in favour of HMRC. HMRC lost their case in the lower and upper tax tribunal, however, and those losses were used as marketing tool for accountants selling similar tax avoidance schemes, convincing many thousands of people to become involved in tax avoidance. An early criminal investigation of the football club officials found to have misled HMRC in that case could well have brought the matter to a conclusion sooner.

(3) Thirdly, there is the question of fairness. It might be thought that in the interests of fairness all acts of criminality of a similar type (for example acts exceeding a specified degree of seriousness) would be prosecuted, but in fact HMRC are under no obligation to enforce tax law in a manner which is fair as between taxpayers.¹⁵ This is an element of their operational independence and there are good reasons for it as a matter of principle; it means that HMRC can address tax abuse on a case-by-case basis without having to constantly second guess what a court might think.

But the fairness we are considering here is on a wider scale: it is about fairness as between the economic strata of society. Research by Taxwatch found that in the 11 years between 2009 and 2019, the government prosecuted 23 times more people for benefits crime (86,000 prosecutions) than tax crime (3,600 prosecutions). This vast disparity was not driven by a greater propensity of benefits claimants to commit fraud, but by government policy which refers all benefits fraud cases worth more than £5,000 to the prosecuting authorities.¹⁶ This is in marked (and, we argue, unacceptable) contrast to the billions pounds of tax which are treated as a matter for civil enforcement irrespective of the dishonesty involved, provided that the rules of the game are complied with.

(4) Finally, as regards justice and the rule of law, clearly if serious criminal conduct is systematically not addressed as such, then justice is not being served and the rule of law is being thwarted. While cheating the public revenue is a common law offence, it has a statutory basis insofar as it exists because Parliament chose to abolish the more general offence of cheating except in the case of cheating the public revenue.¹⁷ HMRC’s policy in this area is therefore thwarting the will of Parliament that dishonesty in a tax context should be treated as a criminal matter attracting serious sanction.

Recommendations

An immediate legislative fix

The current system of handling antisocial tax behaviour is in large part a matter of policy on the part of HMRC rather than the strict application of a set of rules. We nonetheless propose an immediate legislative fix which will address the most glaring defect in that system. That defect is the virtual immunity from prosecution enjoyed in practice by advisers, enablers and promoters, irrespective of the degree of dishonesty and bad faith they display, provided they comply with the ‘rules of the game’. HMRC’s own research confirms that criminal prosecution remains the biggest deterrent to tax crime among wealth managers, whilst HMRC’s preferred approach of using civil procedures to reclaim tax owed is the least effective deterrent.¹⁸ Yet most tax advisers will know that the risk that they or their clients run of ever facing a jury will be vanishingly small provided they take the precautions we have identified.

We therefore recommend legislation requiring that HMRC (a) consider for separate investigation and potential prosecution the promoters and enablers involved in any tax avoidance arrangement, (b) pursue that investigation and refer it for prosecution unless a determination is made that a successful prosecution would be unlikely or against the public interest, and (c) only offer the taxpayer in the case in question a civil pathway in accordance with COP 8 or COP 9 on the condition that the taxpayer agrees to give evidence for the Crown in respect of the arrangement.

There may be concerns that this measure would impede legitimate tax planning and should therefore be expressly reserved for only the most egregious or aggressive instances of tax avoidance. It should therefore be emphasised that from the perspective of this legislative intervention, the egregious or aggressive instances of tax avoidance will be self-selecting: legitimate tax planning has nothing to fear from a review with a view to potential criminal investigation. Indeed to expressly carve out legitimate tax planning based on a civil law understanding of tax avoidance would be to reinstate the precise problem that this intervention seeks to address: the decriminalisation of certain categories of cheating the public revenue on the basis of having the superficial characteristics of legitimate tax planning.

In addition, there may be concerns that this measure would create an additional burden for HMRC. In principle it should not create a substantial additional burden, since (as noted above) HMRC already reviews all the tax avoidance that comes before it for potential criminal investigation. The primary effect of this measure would be to take out of HMRC’s hands the discretion to wave it through civil channels even in circumstances where there is potentially fraudulent behaviour on the part of promoters and enablers. To emphasise, it should absolutely remain within HMRC’s discretion to offer the taxpayer themselves a civil pathway in accordance with COP 8 or COP 9; the purpose here is to remove the effective immunity from prosecution currently enjoyed by fraudulent promoters and enablers provided they play the game according to the rules.

It might further be observed that a legislative intervention is not necessary to modify HMRC policy in this area. While in theory this is the case, a core purpose of the intervention is as a deterrent to anti-social tax behaviour. A change in HMRC policy would be welcome but it would take time to establish and would be uncertain in its scope. This intervention by contrast, it is to be hoped, would stop certain forms of anti-social tax behaviour dead in their tracks more-or-less upon enactment.

As additional support for this recommendation, we note that the proposed measure would be in accordance with recent OECD recommendations specifically addressed to the UK regarding combatting tax fraud.¹⁹

Looking ahead

In several European countries, the collection of tax is seen as a separate activity from law enforcement, with the authority to investigate tax crime held by branches of the police specializing in economic crime. An additional recommendation, for the longer term, is therefore that the option be considered of separating the enforcement of tax law from the collection of tax altogether.

Thirty-five years ago the Roskill Committee, recognizing the challenges facing law enforcement in prosecuting serious fraud, recommended that the investigation and prosecution of serious fraud be brought together. This led to the creation of the Serious Fraud Office. A separate body could similarly be established to both investigate and prosecute tax crime, which can be equally as complex as other kinds of fraud. Such a body should be under the oversight of the Attorney General’s Office rather than HM Treasury.

The proposed department could also be revenue generating. When HMRC fails to prosecute a dishonest tax adviser, they may collect the tax due from that adviser’s client, but they will receive nothing from the adviser. By contrast an investigative and prosecuting authority in relation to tax crime would be able to deploy proceeds of crime legislation to make significant recoveries from the advisers in addition to any tax collected from their clients.

Raising additional revenues would not be the primary concern of the suggested body, however. Its primary role (in contrast to HMRC’s) should be to prosecute tax crimes. And with such a body in place HMRC would have no discretion as to whether to refer a matter for criminal investigation. It may of course still be expedient to pursue the tax through civil rather than criminal procedures. Any decision to not pursue criminal charges, however, would have to be authorised by an officer of the proposed enforcement authority.

Institutional arrangements such as these, elaborated upon by operational policies to be jointly determined by the suggested new authority and HMRC in establishing their working relationship, would have the consequence that cases of abusive tax behaviour will necessarily be addressed with both relevant objectives – that of raising revenue and that of upholding the law – in mind.

APPENDIX: Case studies

The case studies considered here fall into three categories. In the first subsection we look at tax avoidance schemes where there was no prosecution. This is, as already noted, the norm. We call it the ‘tax fraud game’ because in these cases, even though there might have been fraud, the activity was within the scope of what HMRC will generally treat as not warranting criminal investigation. In other words it was within the ‘rules of the game’ described above.

In the second subsection we look at some exceptional cases where, in contrast to the norm in these matters, there were prosecutions (and indeed convictions). We call this ‘breaking the rules of the game’ because the question on the part of HMRC does not appear to have been ‘was there conduct which could be characterised as dishonest?’ but ‘did the taxpayer and their advisers play the game?’ By the same token, judging by the reaction on the part of the tax professionals involved, the question on their side appears to be whether HMRC is itself breaking the rules of the game by actually prosecuting in these circumstances. In fact, it is of course wholly within HMRC’s discretion to enforce the law of the land if (exceptionally) they choose to do so, rather than merely the rules of the game.

In the third subsection we consider how the game is played in the big-money realm of international corporate tax avoidance, where HMRC seemingly take an even more generous approach than in cases of domestic avoidance, barely enforcing the rules of the game, let alone the law of the land.

The tax fraud game

a. Working wheels

The ‘Working Wheels’ tax scheme caught the attention of the public because it involved well-known radio personality Chris Moyles. It (alongside certain other more-or-less identical schemes) was defeated by HMRC at the tax tribunal in the case of Flanagan & Ors v Revenue & Customs [2014] UKFTT 175 (TC).²⁰ Like many tax avoidance schemes it relied on a complex structure involving funds going round in a circle and magically attracting a tax advantage for the scheme user en route. The advantage came in the form of artificially-inflated business losses, which were then used to reduce the tax liabilities of scheme users under a provision of the tax code called ‘sideways loss relief’. Sideways loss relief allows a taxpayer to offset losses made in relation to one source of income against another.

Schemes like this which involve money going round in a circle often fail at the tax tribunal by reference to anti-avoidance principles deployed by judges. The tribunal had no need to deploy those principles in this particular instance, however. In order for the scheme to deliver its purported tax saving, the scheme’s users had to be used car dealers, because that is the business the inflated losses were said to arise from. But on the facts they simply weren’t used car dealers. A key factual assertion on the basis of which the scheme’s users filed their tax returns was a falsehood.

Some low-value artificial transactions were entered into in respect of some used cars as part of the scheme’s implementation, but the tribunal found that the scheme’s users ‘took [no] interest whatever in the details of the purchases and sales, that they were indifferent to whether a profit or loss was made, and that they obtained the bare minimum of information solely in order that that information could be entered on their tax returns.’ That being the case the scheme’s users were not, on the facts, the used car dealers that they had claimed to be. This falsehood would not have been enough, it should be recognised, to establish fraud. There needs to be dishonesty.²¹ And so in a criminal prosecution the question would have arisen whether Chris Moyles and the other scheme users were being dishonest when they falsely claimed to be used car dealers.

It might be thought that the answer to this question depends on the advice they were given. If the advice had been that they had to genuinely start up a used car business and take it seriously as a commercial enterprise in order to be used car dealers for tax purposes, then it would be very hard for them to claim that they had been honest when they filed their tax returns, since they knew that they had not done these things. If, on the other hand (and this is the more likely scenario), they had been advised that they could lawfully claim to be used car dealers for tax purposes without the need for that claim to actually be true, then the allegation of dishonesty would sit most comfortably with the advisers who sold the scheme on that false basis.

This being an appeal against a mere civil assessment to tax, however, as opposed to the criminal proceedings that HMRC could have instituted on the same facts, Chris Moyles’s honesty and the honesty of the other scheme users was not in issue, and still less the honesty of the people who sold him the scheme. The falsehood about being a used car dealer had no legal consequence for anyone, aside from giving the civil tax tribunal an extra reason to deny the purported tax saving.

The case serves therefore to illustrate the practical consequences of HMRC’s early determinations as to whether they are going to treat a case as avoidance or fraud. In this instance there were the superficial signs of purportedly legal but artificial tax abuse (circular transactions with no commercial purpose &c) and no smoking gun showing dishonesty, and so the matter was treated as avoidance. It was not until the evidence was heard by the civil tax tribunal that the findings of fact were made – i.e. that the scheme users were not used car dealers – that made clear that the scheme users’ tax returns were filed on the basis of an outright lie. But by that stage HMRC were institutionally committed to the process of simply recovering the tax owed (albeit with civil penalties no doubt payable in addition).

The fact that cases along these lines play out like this arguably serves to encourage tax abuse of the most egregious kind. Taxpayers can be filing tax returns that contain outright lies, just as in a clear-cut case of fraudulent tax evasion, but because the advisers have dressed the scheme up to look like legal tax avoidance, they stand a chance of being able to defend the scheme at tribunal, without the slightest risk of ending up in prison instead. They are, as it were, playing by the rules of the game, even if they are breaking the law of the land.

b. Eclipse

Film schemes were a popular form of tax avoidance used by many high-net-worth individuals in the past. They made use of various tax incentives the government had put in place to promote the British film industry. The schemes were designed to generate fictitious film investment, which would generate losses, just as in the Working Wheels scheme discussed above. High net worth investors would become partners in a partnership which purported to carry on some film-related business. The partnership would make a loss which could then be offset against the taxable earnings the investors had made in their real jobs.

Typically with sideways loss schemes the losses would be inflated by some sort of external financing which would mean that the amount of tax relief claimed would be disproportionate to the amount invested. As with Working Wheels the borrowed money would go round in a circle, so the losses the scheme users offset against their other income were a fiction. The result was that investors ended up making money out of the tax system and not the film industry.

HMRC had several options for attacking these schemes. The preferred way of dealing with them was to deny the tax benefits to the partners (the investors) on the basis that the partnerships were not engaged in a trade that was set up with a view to making a profit (a condition of claiming tax relief). HMRC’s argument was that the schemes were designed to be loss making, with the only benefit to users coming from the reduction in their tax bill, and so the appearance of a profit-making business was (like the losses it was designed to generate) a fiction. However, it was also possible for HMRC to bring criminal charges on the basis that the fictitious nature of the losses constituted fraud. In this report we look at two contrasting film schemes; one of each was dealt with as a civil matter (i.e. Eclipse, considered in this section) and another of which (i.e. R v Walsh-Atkins, considered in the ‘breaking the rules of the game’ subsection) was dealt with as a criminal matter.

The Eclipse tax avoidance scheme was designed by HSBC for its high-net-worth customers by Neil Bowman, a former partner at EY and Director of Structured Tax Products at HSBC between 2003 and 2009. HSBC are believed to have earned £25m in fees for their part in the scheme. As this scheme was marketed to clients, the clients would be investing in a partnership which would buy the distribution rights to films produced by Disney. These partnerships would then seek to market the film rights for a profit. Invariably, this ‘marketing’ operation was a failure, and the partnerships ended up making a loss on their investment. These losses were then used to reduce the amount of taxable profit of the investors that had been made in their real employment.

The value of any investment in the partnerships was inflated by loans provided by HSBC and other major banks. However, the vast majority of the money raised by these loans was never applied to any commercial activities of the Eclipse partnerships. Instead, through a series of back-to-back transactions, the loans were simply returned to the lending bank. Their only purpose was make the value of the investment by partners in the film partnership larger than it really was practice.

In addition to the circular flow of funds, a subsequent HMRC enquiry into the scheme revealed that the trade in film rights that was the foundation of the scheme was an illusion. The Eclipse partnerships only held the rights for little more than a day, selling them straight back to Disney. Disney, not the partnerships, did all of the marketing of the films. Disney remained in real control over the film rights at all times. The Eclipse partnerships as a result never had any trade. The losses generated by the partnerships were completely artificial. As a result of the ruling of the court on this issue,²² the inflation of the loans worked the other way and left investors facing tax bills higher than their original investment; in some cases up to 20 times higher.

The scheme users are now taking action against HSBC, seeking to claim back losses of £1.4bn on the basis that they were defrauded by the scheme. According to claimants, HSBC should have known that the scheme was not a legally viable means of reducing a tax bill. Further, the claimants point out that Jonathan Peacock QC, whose legal opinion underpinned the scheme, had advised that in order to qualify for tax relief, the marketing agent established by the partnership had to ‘actually undertake the role it had been assigned’. The claimants argue that the scheme was not implemented in line with the original tax advice, noting that Mr Peacock was never consulted on the final implementation of the scheme.

Whether or not that is what happened in this case (the claim is not yet resolved) this is a widespread phenomenon in so-called ‘legal’ tax avoidance that turns out to be legally ineffective. The scheme relies, in some way, on squaring a circle. For example, one party needs to control something for commercial reasons but another party needs to control things for tax reasons. The scheme is nonetheless implemented as a paper exercise, with the professional advisers involved knowing that the square has not been circled – they are just hoping it winds up yielding the tax benefit anyway one way or another.

That knowledge that the scheme relies on an imaginary perfect implementation that does not reflect the reality being implemented could very well be understood as dishonesty, and yet generally HMRC simply comply with their policy and treat these matter as if it was just bad luck on the part of the taxpayers that their clever scheme did not actually work.

Breaking the rules of the game

c. R v Walsh-Atkins

R v Walsh-Atkins demonstrates how some of the same arguments deployed by HMRC when treating film schemes as a civil matter (as to which see Eclipse above) can equally be applied in a criminal prosecution for tax fraud. The case was a criminal trial which ended in the conviction (under the common law offence of cheating the public revenue) of a number of people involved in all the component parts of a film scheme, from film makers to scheme promoters and investors.

The tax avoidance scheme in this case was set up by Terence Potter, a former partner at EY and former senior member of the Chartered Institute of Taxation. In many ways the scheme he constructed was similar to other film tax schemes. High-net-worth individuals with large amounts of taxable income became investors in film partnerships. The partnerships made losses, which were inflated to increase the tax deduction. The losses were then offset against the taxable income of the investors.

There was one key difference, however. The losses in this case were not created through third party loan arrangements but by the inflation of the cost of production via false invoices. The inflation was tax driven, in that the level of losses was predetermined based on the amount of tax losses that were required. The inflation of invoices also allowed the film makers to over-claim film tax credits, a form of subsidy for the film industry which is based on production spend.

In view of HMRC policy in this regard, it is clear that this falsification of documents is what persuaded HMRC to pursue criminal prosecution in this particular case. However, this element of the fraud was not the only issue pursued at the trial. A significant part of the case against the investors and scheme operators involved how sideways loss relief was used to claim tax deductions.

In the case summary the Crown made the point that it would be ‘commercial nonsense’ to increase expenditure to inflate losses, yet that is exactly what happened in this case. As it was put by the Crown: ‘ultimately, if people are putting up their own money then unless they are stupid they want to make some money out of it.’ The fact that these partnerships were constructed to inflate costs and lose money was evidence in support of the overall case that the purpose of the scheme was a conspiracy to cheat the revenue. The prosecution sought to prove that the investors never really had any interest or participation in the trade beyond the opportunity to lower their tax bill, and as such they could not be considered active partners.

This element of the prosecution case is analogous to the findings of fact of the tax tribunal in Eclipse discussed above. In that case, there was no trade. The marketing business which the partnership was supposed to carry out was a fiction. Control over all aspects of the film’s marketing and distribution remained with Disney at all times. In Eclipse, however, rather than adding to an overall picture of criminality, the fictitious nature of the trade was a mere technical feature in a civil legal analysis with no more serious consequence than to deny the intended tax advantage.

Terence Potter received an 8-year prison sentence for the role he played in the scheme, whereas the mechanisms of law enforcement have completely bypassed the people behind the Eclipse scheme and others like it, even though they have sought, by means of similar fictions, to generate tax losses leading to billions in unlawful tax claims. Obviously there is a clear ‘smoking gun’ in the case of R v Walsh-Atkins, in the form of the falsified documents, and that piece of evidence was what HMRC considered sufficient to pass the file over to the Crown Prosecution Service. But once in the hands of the Crown that piece of evidence was treated as merely part of an overall picture of dishonesty that took in the fictional nature of the entire scheme. There is no reason to assume that schemes lacking the evidential ‘smoking gun’ of a falsified document must necessarily be less dishonest.

d. R v Charlton

The case of R v Charlton and others [1996] STC 1481 provides a rare illustration of a tax lawyer facing criminal prosecution for their role in a tax avoidance scheme. A barrister was successfully prosecuted, and sentenced to fifteen months in prison (reduced to nine on appeal), for facilitating what he very obviously thought was perfectly legal tax planning. Accountants involved in the transaction were prosecuted too, but it is the attitude displayed by the barrister in his appeal against conviction and sentence which is particularly revealing for present purposes. He clearly thought he was playing the game according to its rules, and was indignant that HMRC had elected to enforce the law instead.

The scheme involved UK companies buying inputs from third parties at market prices, but buying them through captive offshore companies which applied a mark-up, creating (i) increased deductions from taxable profits onshore and (ii) an accumulating bundle of untaxed income offshore, which the business owners dipped into for personal purposes.

The transactions were highly artificial, the purportedly arm’s length prices were significantly inflated to maximise the tax advantage, and the legal advice that the scheme was viable lacked credibility. These features of the matters were all (rightly) treated by the prosecution as contributing to the overall picture of dishonesty. But to a reader familiar with how HMRC treats abusive tax conduct what is surprising about the case is how ordinary these features are in the world of ‘legal’ tax avoidance. HMRC sees matters like this all the time and doesn’t categorise them as fraud for the purposes of internal procedure, still less actually prosecute.

As well as being highly artificial there were aspects of the scheme which were simply incompetently implemented, and the reason HMRC prosecuted in this instance appears to have been because these features of the implementation were not fully disclosed at the outset of the investigation. The defence of the barrister involved was that he only advised on the basis of what was in his instructions, but he was convicted on the finding that he knew more about the poor implementation as time went on, even past the point where HMRC was investigating and still no disclosure of the full details had taken place. It was seemingly this concealment which sent the handling of the matter down the criminal pathway, resulting in the barrister’s prosecution and imprisonment, notwithstanding that in essence the scheme barely differed from the kind of badly implemented and highly aggressive avoidance which HMRC is accustomed to defeating in civil tax litigation. The judgment of the Court of Appeal in the tax barrister’s appeal against his conviction and sentence notes that he

denied any dishonest involvement with the schemes. It was his case that he acted in the best traditions of the Bar by protecting his clients from any oppressive inquiries by the Revenue. It is apparent that [he] has a certain hostility to the Revenue and he conceives it to be his duty to ensure that the Revenue act within the limit of the powers entrusted to them by statute. He contends that all his actions in these cases were directed to that end and at no time was he acting dishonestly.

As anyone with experience of the tax industry will attest, this kind of attitude – an apparent belief that even the most tendentious avenues of tax avoidance are somehow in the spirit of (rather than running counter to) the core democratic principle of the rule of law – is widespread among tax professionals, and for the most part this barrister was just playing the usual game. In this instance he breached the rules of the game so far as HMRC were concerned, meaning that the fundamental dishonesty inherent in this kind of conduct ended up leading him down the extremely rare path of criminal prosecution.

The game being played on a global scale

e. General Electric

Fraudulent conduct dressed up as legal tax avoidance, and then treated as non-criminal by HMRC, is common throughout the world of international corporate tax planning, and this is something HMRC are quite frank about. In a recent document on the subject of non-compliance in this area they say this:

Our investigations into [international corporate profit shifting] have established that in a large number of cases the factual pattern outlined to HMRC at the start of an enquiry does not stand up to scrutiny once tested. That may be a result of a careless error (for example individuals within a group being unaware of what the actual facts are) but it may also be a result of a deliberate behaviour, that is a group knowingly submitting a [transfer pricing] methodology in a Corporation Tax Return based on a false set of facts. A common issue is an overstatement of functions performed, assets used and risks assumed in entities taxed at lower rates, and an understatement of the functions performed, assets used and risks assumed in the UK [emphasis added].²³

And of course transfer pricing based on a false set of facts is just one of the many kinds of dubious conduct on the part of multinationals. In this case study we consider a complex ‘hybrid arbitrage’ tax avoidance scheme.

In 2002 General Electric moved $5bn between the US, Luxembourg, the UK and Australia in just four days as part of a structured transaction. The scheme allegedly generated a tax benefit for GE in the UK of up to £760m over a period of 10 years. The principle behind a hybrid arbitrage scheme is relatively simple, although the schemes themselves can be very complex. Tax lawyers search for mismatches between the domestic tax legislation of different countries in order to try to make sure that their income falls through the gaps of the tax system and is not taxed anywhere. One of the most famous examples of a hybrid arbitrage scheme is the so called ‘Double Irish’ structure that was employed by many technology companies in the previous decade. These schemes used a company in Ireland owned by a US corporation that was not considered to be tax resident in either Ireland or the United States, leading to no tax being paid on any profits generated by the company.

The GE scheme exploited a mismatch between the tax treatment of Australian partnerships under UK and Australian tax law. Under UK law Australian partnerships were considered to be transparent, meaning that the partners were liable for any taxable profits or losses generated by the partnership. However, in Australia, these partnerships were not transparent, meaning that they had their own tax liability separate from any liabilities of the partners. In the GE scheme an Australian partnership had two UK-based companies acting as partners. The mismatch between UK and Australian law meant that expenses incurred by the partnership could be deducted from tax liabilities in Australia, as well as from the partners’ tax liabilities in the UK. GE then granted a multi-billion-dollar loan to the partnership from a third GE company which generated UK and Australian tax losses on the same transaction.

Hybrid mismatches are a well-known form of tax avoidance and tax authorities have been alive to the threat that that companies will seek to exploit them for many years. To counteract this threat the UK has implemented a number of anti-avoidance provisions in legislation. This includes the unallowable purpose rule, which allows HMRC to disregard the effect of any transaction that has been entered into solely in order to gain a tax advantage. In 2005 the government also introduced new anti-arbitrage rules. This allowed HMRC to disallow any tax deduction claimed by a company where there was no taxable receipt somewhere else, or where a company had claimed another deduction for the same expense. However, the rules only applied where the main purpose was to gain a UK tax benefit.

Under the rules companies could seek an agreement with HMRC that the tax authority would not apply the rules in the future to any particular transactions under a clearance process. In 2005 GE approached HMRC to gain clearance under the new anti-arbitrage rules for 107 loan transactions amounting to £21.2bn, including the transactions involving their Australian partnership. Although HMRC were initially highly sceptical that the Australian transactions were anything other than a tax avoidance scheme, they agreed a partial clearance which allowed GE to deduct most of the interest costs associated with the Australian transactions in the UK. HMRC say they did this after receiving assurances from GE that the transactions constituted a commercial investment and that any tax advantage gained from the structure would arise in Australia and not the UK.

In 2019 HMRC filed proceedings at the High Court seeking to void the clearance agreement they had reached with GE on the basis that GE had failed to disclose material facts that relating to the transactions. In particular, HMRC alleged that they were not aware that the creation of the Australian partnership was part of a larger set of transactions that saw money move between the US, Luxembourg, the UK, Australia and back to the US over a period of just four days. This suggested that there was little or no commercial purpose to the transactions, other than the exploitation of the arbitrage opportunity in the UK and Australia.

HMRC alleged that GE had withheld information, specifically by deleting key passages of documentation from the minutes of board meeting before sending it to HMRC. The full minutes would have shown that the amount of money going through the UK was far more than the amount needed to buy assets in Australia, suggesting that the transactions were tax driven rather than being commercially driven. The tax authority also alleges that GE stated that their view was that UK anti-avoidance legislation should not apply to the transactions because the main purpose of the transaction was not to avoid tax in the UK, whereas in 2013 GE told the Australian Tax Office that the purpose of the scheme was “to gain a tax advantage in the UK not Australia”.

HMRC’s original claim against GE was that the company had made an innocent mistake in failing to disclose material facts in the course of the clearance discussions. But they then later attempted to change their claim to allege fraud. HMRC’s change in strategy to openly allege fraud against a major multi-national company sent shockwaves through the tax profession when revealed in the press. It was simply unheard of for HMRC to allege fraud against a professionally-advised large company with regard to their international tax affairs (although such allegations are common in Europe and elsewhere). GE embarked on legal proceedings to challenged HMRC’s attempt to allege fraud, arguing that HMRC was out of time to do so. They won this argument at the Court of Appeal. HMRC applied for, and was granted permission to argue the point at the Supreme Court. However, before the matter was resolved, HMRC reached an out of court settlement with GE accepting a tax liability of just 10% of the original amount claimed.

This case has demonstrated the problems that can arise when HMRC do not raise allegations of fraud in a timely manner of fail to investigate evidence of fraud. The court papers show that HMRC’s Fraud Investigation Service twice turned down the opportunity to investigate the case after it was referred to them. Although clearly HMRC believed that the clearance agreement they entered into was obtained fraudulently, the time it took them to reach that conclusion meant that they encountered procedural difficulties in bringing the claim; difficulties that would not have been an issue if they had investigated the fraud (if that is what it is), as such, in a timely manner. None of the people who perpetrated it appear to be at any risk of criminal prosecution.

This is not an official publication of the House of Commons or the House of Lords. It has not been approved by either House or its committees. All-Party Parliamentary Groups are informal groups of Members of both Houses with a common interest in particular issues. The views expressed in this report are those of the group.