Comments from ACCA to HM Revenue & Customs
Dealing with tax evasion is a matter of balancing two conflicting models of taxpayer behaviour which together account for the levels of compliance experienced by the administration – on the one hand, the “rational economic taxpayer” model, which attributes compliance purely to the analysis by the taxpayer of the relative costs and likelihoods of paying tax voluntarily as against evading and the risk of discovery/penalty, and on the other hand the “political animal” model, which acknowledges the moral decision by taxpayers to comply with the law voluntarily, without any consideration of the potential benefits of failing to comply.
The current set of proposals seem directed squarely at the ‘rational’ model, geared as they are entirely at increasing the penalties for those who have been found to have broken the rules. Given that the deliberate evaders targeted by these measures are likely to align more with the “rational” than the “political” model, such focus seems sensible. However, the measures which are ultimately adopted should therefore also operate so as to deter deliberate actions, and care must be taken that any deterrent effect is genuine and not structured so as to inadvertently catch those who would intend to comply but due to genuine mistake or reliance on others find themselves in breach of a rule which they in fact support.
Strict liability offences are, by their nature, an indiscriminate weapon. The only practical control over them is simply not to prosecute in the first place. So, if concerns about inappropriate application of a criminal penalty are to be addressed, then it will have to be through the mechanism of discretionary prosecution policy on the part of HMRC.
The experience from the Liechtenstein Disclosure Facility is that HMRC can operate a successful discretionary regime, and get to outcomes which most commentators would view as “fair”. So it is possible that HMRC could operate any strict liability offence on the basis of a similar pragmatic approach, recognising that even in the post Common Reporting Standards (“CRS”) world there could be individuals who have through genuine mistake managed to fail to disclose. If their conduct is clearly not criminal, it would be unfair (in the broadest sense) to then attach a criminal record to them. However, this would involve considerable application of resource which HMRC may not be able to commit, and which, if available, could almost certainly return better results for the exchequer and society in general if deployed elsewhere.
That is one possibility. However, there is also the consideration of the individuals at whom this is really aimed – those who are deliberately and consciously failing to declare income or gains held offshore when they are in fact liable to UK tax, and with a view to evading that tax. These people need to be discouraged, and it is encouraging to see government looking at all the possible options to create such incentives within the framework of the consultation process.
However, it is important that before additional laws are added to the statute book it is absolutely clear what they are for and why they are needed, in line with the first stage of the consultation framework. Criminalising the failure to properly complete the parts of a tax return which relate to some jurisdictions but not others draws a line in the public mind between operating in one place rather than another. To the extent that taxpayers have a choice which jurisdictions to operate in, this is appropriate. For high net worth individuals, while they may have a natural affinity to one (or more) offshore jurisdictions for family reasons, choice of where to invest assets is more likely to be driven by purely financial considerations.
However, with increased economic migration around the world there are ever greater numbers of individuals falling within the UK tax net who have offshore connections but very little in the way of taxable wealth or income. For these lower paid and less well off taxpayers there is unlikely to be any choice as to location of assets or investments; all their ties will be with one country and there will be no scope for them to change that link country. Allowing the categorisation of such taxpayers’ behaviour to be treated as criminal, or not, purely on the basis of decisions made by their governments over whether to sign up to the CRS runs the risk of introducing an unnecessarily arbitrary element to the operation of the criminal justice system in the UK.
The indications are that the offence is intended to be perceived as a deterrent. But its only real deterrent value is in the fact that it is a criminal, rather than a civil, offence. There are already substantial financial penalties available under the existing regime, and of course the burdens of proof are that much lower for HMRC in civil cases. And if HMRC have enough evidence to support a criminal charge of failing to disclose then they will clearly be in a position to mount a successful civil action. Perhaps more importantly, HMRC will only be in a position to bring the charge once it has found out about the offshore funds. If the funds can be kept secret, there will be no prosecutions, civil or criminal.
Of course HMRC is not able to gather much in the way of detail as to the motivation and mindset of determined offshore evaders. However, there does seem to be literature which indicates that evaders believe themselves to be less prone to discovery than average; a sense of invulnerability is common.
A deterrent functions by influencing deliberate behaviour. The purpose is to make the outcomes from a particular course worse than they otherwise would have been, and thereby influence the behaviour of actors away from that course.
Strict liability offences exist for a different reason. They reflect the concern of society that certain acts or omissions have so damaging an impact that they should be classed as criminal, regardless of the intentions of those involved. They are typically the “malum in se” class of offences, that is, so abhorrent by their very nature that regardless of any external or artificial framework. The majority of such offences relate to health and safety, road traffic offences and possession of weapons.
It is of course a part of the role of the tax authority to tackle antisocial behaviour. However that should not extend to relying upon administrative traps to do both the things that need to be done and also things that should not be done. Evasion is properly and rightfully criminal. Absent-mindedly failing to declare something, or missing it through simple ignorance of what may be highly complex rules and regulations is not necessarily worthy of criminalisation. If this offence is to be created at all there should therefore be robust and comprehensive safeguards and statutory defences in order to prevent the unnecessary criminalisation of individuals.
3.7 Do you agree that the applicability of the offence should be limited to income tax and capital gains tax?
Yes. The complexities of implementing the offence sensibly in the context of lifetime transfers would make design difficult. In the longer term, the imposition of strict liability offences on executors has the potential to be problematic. While they may be able to use the funds of the estate to settle any civil penalty, a criminal record would inevitably attach to the individual in what may not be appropriate circumstances.
3.11 Do you agree that the offence should be restricted to taxable income and gains which arise offshore?
Yes. As is noted, the position when dealing with sums which arise in the UK but are subsequently moved offshore would be problematic. While HMRC would need only to prove the act of moving non-compliant funds offshore in order to meet the likely scope of the offence, establishing the non-compliant nature of the funds would, as noted in the civil penalties consultation paper, be potentially very difficult. Establishing it to the necessary balance of probabilities to satisfy a criminal burden of proof is likely to be extremely difficult.
No matter how long the civil regime has to bed down, these evidential burdens will remain and the difficulty of establishing beyond reasonable doubt, rather than just on balance of probabilities, that the funds moved are non-compliant funds is likely to make the establishment of a strict liability criminal offence inappropriate.
3.19 In your opinion, which option (to apply the offence only to investment income or gains, or to apply the offence to all offshore income and gains) would best deliver the policy intention?
Restricting the offence to certain categories of offshore activities could be seen to defeat the policy objective of the measure, and in addition introduce distortions into the treatment of offshore non-compliance. The creation of a comparative advantage to certain forms of non-compliant behaviour could be unhelpful.
That having been said, it is of course more likely that an individual could inadvertently misrepresent their tax affairs in respect of other assets than in respect of investment income and gains, which will almost invariably involve the participation of banks or other financial intermediaries who will almost as a matter of course remind investors of their obligations to report taxable income. Likewise, it is currently the case that financial products and investments are typically more attractive to evaders than other less moveable and more traceable forms of property, so the likelihood of deliberate evasion is greater in that asset class (although it must of course be acknowledged that a differential penalty regime would inevitably shift the relative attractiveness of different classes to the conscious evader).
With that in mind, a restriction of the offence to investment income and gains only would be preferable, as it would restrict the scope for unfair application to unintentional non-reporters while retaining the high profile impact of introducing a strict liability offence, and having the same over impact on the likelihood of the determined evader attempting to misrepresent their affairs to HRMC.
3.25 Do you think that the offence should apply to income and gains which are reported under the Common Reporting Standard?
Restricting the application of the offence by reference to factors which act as a proxy for intention might be considered counter-intuitive in the context of a strict liability offence. If the intention of the taxpayer (to make life difficult for HMRC) is relevant to the offence then this should be built into the offence itself rather than introduced indirectly, and thus imperfectly, through other aspects of the design of the offence.
3.30 Should all income and gains in CRS jurisdictions be exempted from the offence, or should the offence apply to any income and gains which are not automatically reported to HMRC?
This would potentially result in certain classes of non-compliance effectively benefitting from taking place in CRS territories. The aim of the new offence is to dissuade individuals from doing things to make it more difficult for HMRC to identify their non-compliance, that is, to dissuade individuals from structuring their affairs in a way which makes it harder for HMRC to identify particular specific elements of their affairs. Excluding non-financial income and gains in CRS countries from the regime would significantly dilute that policy intention.
However, as is acknowledged, individuals would be unclear whether they might be potentially exposed to a criminal liability in a CRS territory. Such lack of clarity would however be apparent only to those who had turned their mind to the offence, and its likely interaction with their own circumstances. Any individual sufficiently determined as to investigate both the functioning of CRS and of HMRC’s range of sanctions is perhaps unlikely to be swayed by whether a particular sanction is strict liability or otherwise.
For individuals who are ignorant both of their obligations and the offence, it is by no means clear that in every case a criminal sanction should attach to the circumstances leading to that ignorance. As such, restricting the scope of taxpayers to whom the offence could apply will function as a safeguard for the unintentional non-reporter, while having little impact on the likely behaviour of the deliberate evader. The symbolic nature of the existence of the criminal offence would of course remain.
4.4 Do you agree that a de minimis threshold is appropriate?
Yes, without a shadow of a doubt. The exclusion of IHT from the offence will remove perhaps the most likely avenue for those of otherwise limited means to fall within its scope. Nevertheless, this crime will in principle apply not just to the wealthy and determined tax evader but also to those less wealthy economic migrants who have neither the education to understand the full complexities of the UK system nor the funds to engage professional advice. Criminalising their behaviour, especially on the arbitrary ground of whether the authorities in their home country have signed up to the CRS or not, is in nobodies’ interests. While the defence of “reasonable care” may well operate in such cases, getting that far along the administrative road would be a waste of valuable HMRC time and potentially distressing for the taxpayers. A simple filter which excludes such cases from the very beginning is essential.
4.6 Should the de minimis be set by reference to the potential lost revenue arising from the failure/inaccuracy, or some other measure? If so, should the potential lost revenue be calculated in the same way as it is for the purposes of determining civil penalties?
As is discussed in the detail of the threshold calculations, the level of tax undeclared may vary considerably depending upon the nature of the underlying amounts not declared. However, in the context of a tax offence, measuring seriousness by reference to the amount of tax at stake has an undeniable logic.
4.10 Should the threshold be incorporated in statute or guidance?
Secondary legislation. HMRC is not geared up to operate as a discretionary criminal prosecutor on the sort of scale which is likely to be required for an administrative offence of strict liability, and creating the additional volume of necessary expertise would be a less than optimal use of funds.
4.18 Are there any further options (for setting the threshold)?
4.19 Which approach to setting the threshold do you favour?
Setting the threshold by reference to historical yields is probably the fairest measure. Attempting to reverse-engineer the threshold from deposit balances and interest rates would be too sensitive to subsequent rate movements, and would in any event be less transparent. Ironically, that lack of transparency would actually increase the effectiveness of the measure – individuals would be less able to work out whether their transgressions might breach the threshold, and would thus have a greater incentive to remove the risk of prosecution altogether by ‘coming clean’.
4.20 The Government's view is that the threshold should apply for each tax year, rather than in respect of a cumulative amount of potential lost revenue, as a new offence would be committed for each tax period – e.g. each time an incorrect return is filed. Do you agree?
Yes. Allowing the de minimis to operate as a function of time as well as value brings a second dynamic into play and would result in significant potential for inconsistent outcomes.
4.23 Do you agree with the principle that the available criminal sanction for offshore non-compliance should not be seen as more lenient than the available civil sanction?
As a principle, yes. However it must be borne in mind that in practice the desire to implement a criminal offence here is precisely because it is criminal, not civil.
4.30 Should an unlimited financial penalty be available to the courts?
Yes. In reality, the penalty is not of course absolutely unlimited. It would not in any event be impossible to insert a cap on liability under the offence, even if set at a comparatively high level – say £100,000.
4.37 Is the harm which could be caused by a failure to declare offshore income and gains sufficient that a custodial sentence could be justified in the most serious cases?
Yes – although whether it should follow as a consequence of the strict liability offence where normal offences are also available for prosecution is perhaps more open to debate. There must be a suspicion that a custodial sentence would in fact only be pursued in those cases where HMRC were convinced of an individual’s guilt in respect of more serious offences, but unable to provide the relevant evidence of intent to the court.
4.39 If a custodial element is appropriate, should the maximum sentence be six months?
Given the administrative nature of the offence 6 months as a maximum seems appropriate.
5.8 Should it be a defence for (i) a person to demonstrate that they had taken reasonable care in conducting their tax affairs, or (ii) a person to demonstrate that they had sought and followed appropriate professional advice? What would be the impact on the likelihood of successful prosecutions if statutory defences are included?
There must be safeguards to protect individuals in cases of genuine mistake or reasonable reliance on others. While in practice there may be scope for taxpayers to rely upon HMRC’s discretion in bringing prosecutions, this will not and cannot be a proper procedural safeguard. Accordingly the ‘reasonable excuse’ type defences should still be available.
5.9 Should any other statutory defences be introduced?
5.16 Are further safeguards appropriate? What should these be?
Provided that the offence is sufficiently clearly defined, the defences of reasonable care and reasonable reliance on others properly framed and the application of the de minimis sufficiently robust there should be no need for additional safeguards or defences beyond those already discussed above.