Abstract
The corporate criminal offences of failure to prevent facilitation of tax evasion punish the unlucky, unpopular, and large more than genuine tax evaders or facilitators. This article argues the offences’ broad scope and the unpredictability of the ‘reasonable prevention procedures’ defence means that rather than operating as a deterrent, the provisions effectively assign guilt by lottery, impose fines that become just another cost of doing business, and punish British companies for doing business in the developing world where their investment is most needed.
Introduction
One of the many baffling features of Western politics is the fetishisation of small, as opposed to big, business. To say small business is the backbone of the economy is motherhood and apple pie; but anyone who says the same about big business is treated as a shill, an elitist or both. 1 These starkly different attitudes do not make much sense. Any cut-off between small and large businesses is arbitrary, and the differences between them matter less than their similarities. For the right, both create jobs and have real human beings relying on their success, whether the individual sole proprietor or the millions whose pensions and savings are invested in the stock exchange and loan capital. For the left, both maximise profit by the exploitative process of extracting surplus value from the labour of their employees. If anything, the prejudice would be more understandable if it were the other way around. Large businesses are not innately more kind-hearted or law abiding than small, but they have fewer opportunities for wrongdoing in that they are exposed to more public scrutiny and regulation, provide juicier targets for plaintiffs’ lawyers, and adopt more formal processes. It is the small business, conducting its arrangements without formal contracts and largely in cash, that can more easily cheat on its taxes 2 and steal its employees’ wages. 3
The tax offences in the Criminal Finances Act 2017 in part facilitate this prejudice through overcriminalisation that leaves too much discretion in the hands of prosecutors and juries. 4 Rather than targeting the businesses most likely to engage in or facilitate tax evasion, the offences effectively punish unlucky, unpopular, and above all big businesses, often for the dishonesty of small businesses they deal with.
First, the offences’ broad scope means that for any commercial enterprise over a certain size, prima facie liability is less a wrong than an uncontrollable statistical inevitability. Which businesses are charged becomes a matter of luck or political prejudice, rather than reflecting moral culpability.
Second, the ‘reasonable prevention procedures’ defence’s open texture makes it unpredictable and an invitation for jurors to apply their own prejudices against big business. The defence has become less an argument and more a bargaining chip with prosecutors, which in turn means limited case law can develop. It has never been relied on in court, precisely because it is so unreliable.
Third, the jurisdictional scope of the offences means that not only can foreign companies with only tangential connections to the UK can be punished. British businesses can also effectively be punished for doing business in the developing world, where informal tax arrangements are common and may indeed be tacitly permitted, while actual wrongdoers and facilitators face almost no consequences for their actions. The UK’s scarce investigative and prosecutorial resources are used to punish wrongdoing that foreign governments have made the difficult but ultimately legitimate choice to tolerate.
The offence: Less a wrong than a statistical inevitability
The offences of criminal failure to prevent facilitation of tax evasion are set out in sections 45 and 46 of the Criminal Finances Act 2017, which are based on the failure to prevent bribery offence in section 7 of the Bribery Act 2010. 5 The two tax offences differ in their jurisdictional scope (which we will come to in Part 3), but share two key elements. First, we look to the class of ‘associated persons’ — that is, people who are employees of, agents of or provide services to 6 the corporate defendant. 7 Second, we ask whether in that capacity, any associated person ‘facilitated’ — for example, aided and abetted, or was knowingly concerned with — any other person’s tax evasion. 8
That, rather terrifyingly, is it. There are two points to note about this definition.
The first is the different mens rea requirements for the different parties involved. Both the sections 45 and 46 offences are parasitic on the underlying offences committed by the tax evader and facilitator. By the nature of those underlying offences, the tax evader and facilitator must have acted in a subjectively dishonest way. Tax planning that is aggressive, optimistic or simply unsuccessful is not thereby converted into tax evasion: rather, the evader must have acted dishonestly, for example, by fraudulently misrepresenting their financial position to the tax authorities. 9 Similar, a facilitator must know, or at least be reckless to the possibility, that they are involved in tax evasion. 10 On the other hand, there is no mens rea requirement for the corporate defendant — the party actually being punished under the provisions — whatsoever. A company may pay its taxes honestly, and ask its associates and counterparties to do the same, but still be stung with criminal liability, charged an unlimited fine, 11 and branded a tax cheat in the media (which may not understand, or choose to ignore, the finer points of the statutory definition). 12
This approach to mens rea marks a considerable departure from the common law’s approach to other offences committed by corporate entities. These usually 13 require a mens rea of criminality to be located in the company’s directing mind and will — typically, its directors or senior management. 14 Thus, for a company to be held liable for facilitating another’s tax evasion, the requisite dishonest state of mind must be found in one of this class. 15 Two justifications are offered for lowering the standard for the new offences, although neither is very persuasive.
The first justification is that it may be difficult for prosecutors to prove mens rea to the directing mind and will standard, especially in large corporations where directors and senior management are a long way from facts on the ground 16 and delegate decisions to local managers. 17 But the emphasis on ‘proving’ mens rea is misleading. The problem is not evidential. It is not that directors have the requisite state of mind, and that evidence to prove it is somehow lacking — especially in large corporations that are more likely to document their decisions in formal procedures. Rather, the directors and senior management will usually not have that state of mind precisely because they are so far away from the facts of the tax evasion. Any crime will be easier to prove if the prosecution does not have to prove one of its elements, but this does nothing to show why a person without that element should be culpable. Indeed, the usual justification for subjective mens rea in white collar criminal offences is that punishments can only fairly apply to, and only effectively deter, people who are conscious of their wrongdoing and can choose to act differently. But liability for these offences does not require consciousness at all — and as we will see, gives large companies almost no way to avoid incurring at least prima facie liability.
The second justification is that there is less need to worry about unfairness to defendants in these offences, because they punish companies rather than individuals, through fines rather than imprisonment. 18 But this formalism evades the issue. Economic effects on companies have similar effects on real people, whether they be shareholders who receive less in dividends or employees and creditors whose wages are reduced. And any principled distinction between measures that punish people by taking their time and measures which punish people by taking their money is porous — for people convert time into money by working, and the effect of a fine is to render pointless the sacrifice of time for the money used to pay the fine.
The second point to note about the definition is the (deliberate) 19 breadth of the concepts involved. 20 Some commentators have characterised the provisions as primarily applying to tax planning advisers, 21 and this is the context stressed in the Explanatory Notes, 22 but in fact the class of associated persons is far broader. It is not limited to tax advisers, or people over whom the defendant has real control (that is, employees), or people who act on the defendant’s behalf and owe it corresponding duties (that is, agents). Rather, it includes anyone who provides services to the defendant 23 — potentially every single firm the defendant contracts with, and every single one of those firms’ workers who actually does work for the defendant. A defendant might well trust its employees not to facilitate tax evasion or trust that it can monitor those employees. But how can it know whether to trust the people who provide it with services, who it may not even know about, still less have any dealings with? The class of tax evaders is even broader, including any one of the people those associated persons deal with in their capacity as an associated person. But in the context of employment at least, that capacity has been very broadly defined, including actions against the defendant’s express instructions 24 and under the guise of the defendant’s authority but contrary to its interests. 25 Indeed, it seems that associated persons who act to enrich themselves or others rather than the defendant can still be caught by the legislation. 26
Now, almost everyone pays some kind of tax (even if only VAT), and most people are not thrilled about paying it. As a result, although minor tax evasion is not especially admirable, it is common enough — for example, purchasers paying tradesmen or suppliers in cash, often for a convenient discount that so happens to reflect the local VAT rate. The result is that, in a commercial enterprise of any significant size, the two classes of associated persons and evaders become very large very quickly, and the probability that at least one of each class engaged in misconduct in at least one transaction approaches statistical inevitability. And the absence of a limiting mechanism — of mens rea on the company’s part, or the requirement that the underlying offenders be under the defendant’s control — means the defendant and its directors can do almost nothing to prevent prima facie liability. HMRC will always be able to find at least one evader and facilitator if it looks hard enough. The purpose of the offences is to impose punishment as a means of deterrent — that is, in the hope that fining these entities will result in them changing their conduct in a way that makes it more difficult for primary offenders to engage in tax evasion. But since it is impossible to avoid liability, liability does not serve as a deterrent, and it becomes unfair to impose punishment as retribution for it. For how can you, and why should you, change your behaviour if you may well be punished either way?
The result is something perilously close to guilt by lottery. In an otherwise ideal world, which large businesses are charged with the offences depends on moral luck — if you are unlucky enough to have an associated person involved in tax evasion, and unlucky enough to be investigated that year, you are prima facie liable for the offence. The offence almost immediately loses much of its moral sting, and fines for it simply become another cost of doing business. On the other hand, in a less than ideal world, it means that prosecutors can target politically unpopular defendants (such as large businesses), or defendants who have the resources to pay significant fines (again, large businesses). The likelihood of prosecution depends less on a business’ moral character (if there is such a thing), and more on its size.
The defence: Less a justification than a bargaining chip
The only constraint on the breadth of the offence is the ‘reasonable prevention procedures’ defence. The defence theoretically has two alternative limbs: 27 (a) that the defendant had in place such prevention procedures as it was reasonable in all the circumstances to expect the defendant to have in place or (b) that it was not reasonable in all the circumstances to expect the defendant to have any prevention procedures in place.
Paragraph (b) we can disregard almost immediately. Linguistically, it adds nothing to paragraph (a): reasonable prevention procedures could include doing nothing at all in the right circumstances (for example, where there was a de minimis risk of facilitation). Practically, it is useless: if the defendant is charged and is relying on the defence, by definition facilitation and evasion have in fact occurred, making it difficult (albeit not impossible) to argue that there was a de minimis risk.
Paragraph (a) is more interesting, for three reasons.
First, the defence does not specify any causal connection between the prevention procedures and the underlying tax evasion. In other words, if the underlying tax evasion is (for example) specifically VAT fraud by a particular supplier, the defendant will not just be required to show its prevention procedures were reasonable in respect of that particular kind of tax evasion by that particular person, but reasonable generally, at preventing tax evasion of all kinds. A defendant relying on the defence opens itself up into a far-reaching enquiry into its relationships with associated persons and others more generally, which may also assist prosecutors to identify other underlying tax evasion.
The second reason is that the legislation does not set out any definition (exhaustive or otherwise) of what might constitute reasonable prevention procedures. 28 Rather, section 47(1) directs us to guidance about procedures formally approved by the Chancellor of the Exchequer, but in practice drafted by HMRC. 29 In other words, HMRC writes the rules, investigates the offence, and is intimately involved in the prosecution 30 (even if charges are formally brought by the Serious Fraud Office or Crown Prosecution Service). 31 Now, although it is difficult to draw a bright line in the murky area of tax evasion, 32 the guidelines HMRC has in fact produced vary between unhelpfully specific and unhelpfully vague. The examples it gives are too narrow to be of much use to most businesses, such as a multinational bank already bound by know-your-customer obligations and a mid-sized car dealership. 33 The principles it gives are open textured to the point of simply restating the underlying provisions — inviting potential defendants to engage in ‘due diligence’ and consider the ‘proportionality of risk-based prevention procedures’, 34 which sound suspiciously like synonyms for the statutory ‘reasonable in all the circumstances’. Some commentators characterise the guidance as a ‘damp squib’ which reveals the authorities’ lack of commitment to effective enforcement. 35 But vagueness in fact makes the offence easier for the authorities to prosecute. There is no safe harbour, and it is impossible for a defendant to know what HMRC will make of its arrangements in its own particular circumstances. (On the other hand, it is clear that HMRC expects more of large businesses than small because of the former’s greater ability to pay for precautionary measures. 36 But some would say that, if a small business cannot viably run its business ethically and legally, it should not be in business at all.)
The third reason is how the reasonableness of prevention procedures is to be assessed. The nature of any reasonableness requirement is that it will be assessed in hindsight, after the procedures have failed to prevent the harm they were supposed to 37 — for otherwise there would be no prima facie case of liability, and no need to rely on the defence. But because reasonableness is a question of fact, and the offences are criminal, reasonableness will be assessed by a jury — which means that the prejudices of 12 ordinary people (likely against large businesses) 38 have room to operate with extremely limited scrutiny and without the requirement of giving reasons. It is likely that because a jury’s decisions are so unpredictable, and the reputational consequences of losing in court are so grave, defendants will choose to reach deferred prosecution agreements (DPAs) with the authorities rather than contesting their liability 39 — even if they have an award winning compliance program and detailed due diligence procedures, as Airbus did before it reached a DPA with the Serious Fraud Office. 40
In each individual case, this means that the provision becomes less a defence than a bargaining chip with prosecutors, meaning more power is vested in the lottery of detection and the prejudices of the particular prosecutor doing the bargaining. And for the system as a whole, the fewer reasoned decisions there are on the meaning of reasonable prevention procedures, the more unpredictable the defence becomes — leading to a self-perpetuating cycle of defendants choosing to settle rather than try their luck, and even greater reliance on the guidance HMRC both drafts and enforces. Notably, the analogous failure to prevent bribery offence in section 7 of the Bribery Act 2010 has been in force for a decade, but there are still no reasoned decisions on the point; 41 similarly, the tax offences have been in force since 30 September 2017, but have not yet been subject to any judicial consideration.
The offences’ jurisdictional scope: Punishing those who do business in the developing world
We said in Part 1 that the Criminal Finances Act 2017 created two offences which differ in their jurisdictional scope. The section 45 offence applies where the underlying tax evasion is cheating the (UK) public revenue, or any offence under UK law consisting of being knowingly concerned in, or taking steps with a view to, the fraudulent evasion of (UK) tax. 42 This is unobjectionable — every state has an interest in protecting the integrity of its own tax system.
Where things become more complicated is the section 46 offence, which applies where the underlying tax evasion is of foreign tax. The legislation requires that both the evasion and facilitation be doubly criminal — that is, an offence under the law of the foreign country, and conduct which would be regarded by the UK courts as an offence if it related to UK taxes. 43 Because the underlying tax has no nexus to the UK, the nexus is provided by the facilitator’s conduct (if part of the facilitation takes place in the UK) 44 or the defendant’s connections to the UK (if it is incorporated or formed under UK law or carries on part of a business in the UK). 45
The usual concern with such broad jurisdiction provisions is that it is easy for foreign defendants with only tangential connections to the UK to find themselves facing charges here. 46 Foreign defendants may be caught if they carry on ‘part’ (an amount not quantified by reference to substantiality) of ‘a business’ (not necessarily their main business, or the one in respect of which the tax evasion occurred), which seems like a vanishingly low threshold. 47 Guidance on the Bribery Act suggests that the jurisdiction provisions will be read down to situations where the defendant has a ‘demonstrable business presence’ in the UK — but there is nothing like this in the statute, and the statute says the guidance is relevant to consideration of the defence rather than jurisdiction or the elements of liability. 48 Similarly, the nexus provided by part of the facilitator’s conduct may be very limited, and beyond the defendant’s control for the reasons given above — if the defendant cannot control the occurrence of the facilitation, how are they meant to control its location? This is not just unfair to foreign defendants, who may not have been able to predict that they would have to comply with UK law as well as their own and therefore may not be fairly held accountable or deterred by UK law. It is also a waste of scarce UK prosecutorial and court resources, which should rightly be focused on protecting the UK’s revenue. (Indeed, as a matter of private international law, English courts applying the common law would not enforce a foreign penal or revenue law, which was seen as an intrusion on British sovereignty.) 49
An equally pressing concern is for British companies which carry on business overseas. The nature of doing business in some parts of the developing world is that tax evasion, like bribery, 50 is formally illegal but in practice tolerated by the authorities. 51 It is a lazy stereotype to say that governments which tolerate a degree of corruption must themselves be corrupt. Every investigating and prosecuting authority has limited resources and must exercise its discretion in choosing cases to pursue. There are legitimate reasons for a government to choose to de-prioritise aggressive enforcement of tax law: for example, because it wishes to allocate resources to other sectors of government activity like healthcare, or prioritise the prosecution of violent crime, or benefit from economic growth driven by the informal economy. But the breadth of section 46 means that British prosecuting authorities can effectively step in to override this choice. And because section 46 focuses on the law on the books, rather than the law in practice, a British company operating overseas can find itself punished for a foreign tax evader’s conduct while the actual tax evader and local facilitator go free, and local competitors engage in similar conduct with impunity. The only way a British company can know with any degree of certainty that it has not engaged in a section 46 offence is not to invest in these jurisdictions at all. 52 But these include some of the countries most in need of foreign investment to lift their people out of poverty. Although the offences are designed to deter facilitation of tax evasion, effectively, British businesses are punished for — and deterred from — doing business in the developing world. 53
Conclusion
Inheritance tax is rightly said to favour the healthy, wealthy and well-advised. 54 In a similar way, the corporate criminal offences of failure to prevent facilitation of tax evasion punish the unlucky, unpopular and large more than genuine evaders or facilitators. The offences’ broad scope and the unpredictability of the ‘reasonable prevention procedures’ defence means that rather than operating as a deterrent, the provisions effectively assign guilt by lottery, impose fines that become just another cost of doing business, and punish British companies for doing business in the developing world.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.