It is a well-known and undisputed fact that the SARs regime under which the NCA receives countless reports either seeking consent to proceed with transactions, as defences to potential allegations of money laundering or terrorist financing or otherwise as required by AML or CFT legislation is broken and not fit for purpose. Equally though, when the Government considered getting rid of the regime, industry agreed we need it in some form. But what exactly do we need? How can it help to detect and prevent money laundering and punish criminals without causing unnecessary delays and costs to business, or result in devastating consequences for those about whom a suspicion has been raised but who turn out to be innocent?
The Law Commission has given this matter significant thought, and has now published a 200 page consultation paper setting out its thoughts and proposals and seeking views.
In this article, Emma Radmore of Womble Bond Dickinson (UK) LLP examines the key proposals.
The starting point
The Law Commission paper follows an agreement between it and the Home Office to review the reporting of suspicious activity under POCA and the Terrorism Act. The paper looks mainly at the consent regime, although many of the observations apply equally to SARs that request consent to carry on with a transaction and to SARs that are made in compliance with a different legal requirement. Consent SARs are separately processed, because of the timing issues that surround them.
Law Commission research shows that the NCA receives around 2000 SARs each day or which around 100 are consent SARs.
The key problems identified and agreed to date include:
- The huge volume of SARs
- The low intelligence value and poor quality of many disclosures received, often due to defensive reporting
- Misunderstanding of the use of the exemption from liability for "authorised disclosures" (and abuse of the exemption by a small number of people)
- The overall compliance burden on entities who are under a duty to report suspicious activity; and
- The impact of suspension of transactions on those who make the SAR and those who are the subject of it – which can only get worse with the new CFA powers to extend the reporting moratorium while law enforcement seek further information.
There has been criticism of the "all crimes" approach to money laundering offences – that is, that the money laundering offences apply in relation to property derived from any crime, regardless of its seriousness. On the one hand, this has its advantages because the reporter does not have to identify a particular underlying crime. But this means that minor offences and regulatory breaches that carry a criminal consequence are caught. For example, a target company has breached a piece of environmental legislation, which is a criminal offence. Technically therefore the, probably minimal, savings made are criminal property and, as a result, the would be acquirer and their legal advisers will need to make a SAR and seek consent to continue with the transaction. Although the Law Commission's research shows SARs of this kind make up a tiny proportion of SARs received, the view is that these SARs made for technical compliance with no substantive value impose disproportionate burdens in terms of the time they take to investigate and make.
Some of these technical issues could be helped if the law moved to a "serious crimes" rather than "all crimes" approach. Defining a "serious crime" could be done either by specifying a list of crimes or by reference to the possible penalties – for example, including only those that carry a maximum penalty of more than one year's imprisonment.
The Law Commission favours the "maximum penalty" approach which would be significantly easier to apply and would not need constant review, but is still not without problem. Not only would an appropriate threshold need to be agreed, but the Law Commission asks whether setting a threshold risks making arbitrary distinctions between offences. There is also a worry that to exclude say, environmental and regulatory offences would diminish their seriousness (as examples, the two "failure to prevent" offences, which do not carry custodial penalties, would not be "serious crimes" under this formulation). Also, even a "technical" offence may actually result in significant profit.
Overall, the Law Commission concludes that a "serious crimes" approach would be problematic but welcomes views, including on whether, perhaps, the requirement to make a SAR could be restricted only to serious crimes.
Any change? On balance, the Law Commission seems to tend towards maintaining the status quo.
What is suspicion?
At a basic level, the Law Commission states suspicion can range from:
- Imagining something without evidence
- A possibility, which is more than fanciful, that the relevant facts exist
- Suspicion on some verifiable or articulable grounds; and
- Having a strong or settled suspicion that is firmly grounded and targeted on specific facts.
Many pieces of industry guidance over the years have grappled with how to help institutions define "suspicion". POCA and the Terrorism Act do not define it, nor does any EU AML legislation or the FATF. Lack of clarity is a double whammy for would-be reporters, in that they risk either committing offences by failing to report (or actually laundering) because they do not recognise a suspicion, or will make unnecessary SARs because they are overly cautious.
International standard setters are not blameless. The concept of "knowing or having reasonable grounds to suspect" is set out in FATF Recommendations and MLD4. The Law Commission points out that, in some ways, this could be helpful to the would-be reporter, as it requires less investigation to report on the grounds of a suspicion than it would to require actual knowledge.
Here, as in many other areas, there is a dichotomy – which many forget. In order for there to be a crime in the first place, any preconditions necessary to that crime must have been fulfilled. The would-be reporter does not have the luxury of knowing whether, had the suspected crime been investigated on its own merits, it would have met the necessary conditions for prosecution. Many reports have queried why the thresholds are set at the level of suspicion and sought clarification on what the Government intended this to cover.
There has been much case law over the years over what suspicion means. The Law Commission considers what it describes as a "hierarchy of states of mind", which is:
- Knowledge – that is, actual knowledge, based on what is known to be true and the person's state of mind
- "blind-eye" knowledge – deliberately not asking questions because one would not want to hear the results. There is also a subset of "constructive knowledge" – where a person had the means of knowledge but failed to make reasonable enquiries. There is a key distinction between this and wilfully shutting one's eyes
- Belief – case law has noted that suspicion and belief are not the same because belief is a lesser state of mind than knowledge but nevertheless requires acceptance of relevant facts, whereas suspicion does not
- Reasonable grounds/cause to believe – case law has addressed whether there should be an objective foundation for a belief. The generally accepted view is that this analysis should stem from the existence of a fact or facts and not just a belief that the facts existed. Case law has stated that the "presence of uncertainties does not prevent there being reasonable cause to believe, but the judge must still be satisfied that there is reasonable cause to believe"
- Reasonable ground/cause for suspicion – the House of Lords said that "reasonable grounds to suspect" means a subjective suspicion supported by objective grounds – hence imposing on a prosecutor a greater obligation than mere suspicion would. In relation to the Terrorism Act, a case considered whether "reasonable grounds to suspect" means the accused must actually suspect, or whether it was enough that the facts known to the accused, if assessed objectively, would be enough. The Court held that there was no difference in meaning between "grounds" and "cause" in this context and said one should not take the view that there had to be actual suspicion. But in other contexts, courts have expected both a subjective and objective element, and that the quality and reliability of material that is alleged to provide the basis for the reasonable suspicion is a key factor; and
- Suspicion – once defined as "a state of conjecture or surmise where proof is lacking". In the POCA context, the leading case stated that "the defendant must think that there was a possibility, which was more than fanciful, that the relevant fact existed".
Suspicion in the POCA context
So, all that is a long-winded way of saying that different people would define suspicion (or cause for it) in different ways and in different circumstances. The Law Commission paper discusses how the courts have interpreted the concept in the context of money laundering offences.
The main case (Da Silva), already mentioned above ("… a possibility, more than fanciful....") stated that a vague feeling of unease is not enough for there to be a suspicion, but on the other hand the feeling need not be "clear", "firmly grounded and targeted on specific facts" or based upon "reasonable grounds". The better view seems merely to be that there must be something more than an "inkling".
The paper turns to the question of whether the fault element of "reasonable grounds to suspect" could in fact be purely objective. Case law appears to have concluded that this is not the case – that the application of the objective test is predicated on the accused having formed a subjective suspicion in the first place. Followed through, this would mean a person did not commit a relevant offence if a person should have suspected because there were reasonable grounds to do so, but in fact did not suspect.
It is no wonder that firms have cried out for better guidance. Sadly, they have not received it. On a pragmatic level, employees of firms that have a "nominated officer" can absolve themselves of their liability by reporting to that officer – which leaves the officer then needing to decide whether there is a suspicion or reasonable grounds for one, that should be reported onwards. Industry guidance varies widely:
- JMLSG has tried its best, referring to judgements and suggesting that something between speculation and belief would be a suspicion
- The Law Society has taken a different tack, suggesting that suspicion may arise from something unusual or unexpected, and then, after making enquiries, when the facts do not seem "normal or make common sense"
- But the Consultative Committee of Accountancy Bodies refers merely to a state of mind that is more definite than speculation but falls short of evidence-based knowledge and would include a slight opinion without sufficient evidence; and
- Meanwhile, the EU law makers have also suggested an evidence-based approach to suspicion is appropriate, while the Canadian view of "reasonable grounds to suspect" appears closer to the British one of requiring a subjective suspicion based on objective grounds.
The confusion and challenges resulting from these inconsistent pieces of guidance, not surprisingly, include:
- Lack of understanding - application of different standards of suspicion and therefore inconsistent reporting, poor quality disclosures because the grounds for suspicion are not clearly articulated and general confusion about what the law is or requires;
- Risks to those in the reporting sector – who have to cope with both the administrative burden of policing the low threshold with the risk of liability if they make the wrong decision in failing to report; and
- The position of the "suspect" – who may find their account frozen and therefore be seriously disadvantaged and suffer severe damage for which they will not be compensated if innocent.
Suspicion and the disclosure offences
The paper points out that the disclosure requirement places a very onerous obligation on reporters as it is suspicion based, yet requires only minimal effort from reporters once they have identified a suspicion.
The paper reveals that there were 158 cases of a potential breach of s331 POCA referred for investigation between 2013 and 2016, of which 12 proceeded to trial. These cases arise where a nominated officer receives a report from an employee who has formed a suspicion, but has not shared the suspicion and has not reported onwards to NCA.
The Law Commission points out that the failure to disclose offences for the regulated sector create four different ways of committing the offence – knowing, suspecting, having reasonable grounds to know and having reasonable grounds to suspect. The Law Commission suggests that if the law is to be interpreted in line with the case of R v Saik, then the term "suspect" is redundant as subjective suspicion would come within "reasonable grounds to suspect". In relation to a case under the similar provisions of the Terrorism Act, however, it was stated "it is plain beyond argument that the expression "has reasonable grounds for suspicion" cannot mean "actually suspects"". If that is right, says the Law Commission, then "reasonable cause to suspect" is a wholly objective test within the context of the disclosure offence. The paper suggests that this would be the appropriate construction for the disclosure offences and so, within the regulated sector, a person would commit the offence if they had reasonable grounds for suspecting notwithstanding that they did not in fact hold the suspicion. The justification for this construction is that employees in the regulated sector who are trained to spot behaviour indicative of money laundering should be blameworthy for their failure to report.
The paper puts forward several reasons why the element of objectivity is appropriate, such as:
- Legislation needs to deter those in the regulated sector from failing to act competently and responsibly where information comes to them that ought to make them suspect money laundering (which of itself could provide a further protection – that the prosecution would have to prove the accused did actually have the relevant information);
- It applies only to the regulated sector, and to information that comes to the relevant person in the course of their employment in that sector – and such people should be better prepared to deal with it than those outside the sector;
- There are defences – both the "reasonable excuse" defence and a defence if the employer has not provided appropriate training; and
- The Court has to have regard to approved industry guidance.
Against this, of course, are the dangers of increased volume and decreased quality of reports, which arguably helps no-one and could detract for the benefits for enforcement agencies of actually receiving the reports.
All in all the Law Commission concludes, based on the factors above and the understanding that there is no requirement for the prosecution to prove that money laundering did actually take place in order to secure conviction, that the objective test sets too low a threshold for liability.
What to do?
The Law Commission tends to the view that it would be both undesirable and impractical to formulate a precise and workable definition of suspicion that would add anything to its ordinary, natural meaning. However, it seeks views on this, and on whether, if respondents believe there should be a definition, there is anything better than the Da Silva formulation of "a possibility, which is more than fanciful, that the relevant facts exist".
Whether there should be guidance on how reporters apply the suspicion threshold, though, it another matter. The Law Commission suggests that one value of guidance could be to allow NCA to search reports on the basis of key words suggested in the guidance, while also allowing supervisors better to educate and advise their members. The Law Commission suggests there should be Government-approved guidance which sets out factors capable of founding a suspicion or reasonable grounds for it, with a non-exhaustive list, and factors that should be excluded. It does not recommend a particular way in which the guidance should be issued, but mentions the models provided under the Police and Criminal Evidence Act, the Bribery Act and the Criminal Finances Act as possibilities for consideration.
Change the form?
Alongside the guidance, the Law Commission proposes a prescribed SAR form, or set of forms. This would not only give better direction to a would-be-reporter on what is required by way of suspicion, but would also ensure consistency of format and presentation, should lead ultimately to a reduction in requests for further information and should make it more difficult for those who try to abuse the system by withholding information to do so.
The Law Commission usefully summarises which offences under POCA work on the basis of "just" knowledge or suspicion, and which extend also to reasonable grounds for knowledge or suspicion. In principle, the Law Commission agrees that a cumulative test that requires proof of substantive suspicion bolstered by objectively reasonable grounds is appropriate for the reporting offences. But it may be less appropriate for the other offences. The Law Commission thinks that, if a reformulated "reasonable grounds to suspect" test applies to the required disclosures under s330 and 331 it would address some of the current problems, and that this objective approach would be in line with the Home Office and Treasury's views that reporters should apply a risk-based approach to reporting rather than a tick box one. The Law Commission also thinks this test should apply to the s332 reporting requirement on nominated officers outside the regulated sector. But it does not think it appropriate to make any change to the current requirements or expectations under the Terrorism Act, not least because it is arguably right that there is a lower threshold for such crimes (plus there are significantly fewer SARs under the TA anyway).
How would this affect the predicate offences in sections 327- 329 POCA? The Law Commission thinks that, because the "authorised" disclosures triggered under these sections are generated in a different way, the threshold for the underlying criminality would need to change in order for the reporting thresholds to change. On the whole, the Law Commission thinks there are good arguments for keeping the authorised disclosure level at mere suspicion as raising the threshold could make prosecution harder.
That said, the suspicion based trigger does create the problems of poor quality SARs and risks severe consequences on the accused, which cannot be undone if the report proves groundless. So there is an argument for requiring something more than subjective suspicion. The Law Commission has proposed changes that will, somehow, manage to retain the low level of suspicion threshold yet lead to fewer and more focused SARs that impact more proportionately on customers. In particular, those made as a defence against money laundering should be evidence based and therefore more likely to be of use to law enforcement agencies.
What are the proposals?
In summary, the Law Commission proposes:
- That the current "knows or suspects" test for the predicate offences in ss 327- 329 POCA remain unchanged
- That the thresholds for the disclosure offences in ss 330 – 332 POCA change from "knows or suspects; or has reasonable grounds for knowing or suspecting" (or just "knows or suspects" in the current s332) change, in all cases to "knows or has reasonable grounds to suspect"
- That the thresholds for the reporting obligation for required disclosures under ss 330, 331 and 332 ("knows or suspects; or has reasonable grounds for knowing or suspecting") change to "knows or has reasonable grounds to suspect; and the authorised disclosure changes from "knows or suspects" again to "knows or has reasonable grounds to suspect"; and
- That, therefore, there would be a defence to the offences in ss327-9 if an individual in the regulated sector has no reasonable grounds to suspect property is criminal property.
The Law Commission has assessed whether these changes would cause any problems under EU or international standards, and has concluded that neither FATF nor MLD4 require a threshold as low as mere suspicion, although it does note some ambiguity in Article 33 of MLD4.
Regardless of whether, or how, the threshold is amended, the Law Commission sees benefits in issuing statutory guidance. The guidance should also cover what may be considered a "reasonable excuse" for not reporting – and one "reasonable excuse" could be that a report had been lodged with another enforcement agency in relation to the same matter. The guidance could also address several typical scenarios in which reports are currently made but are felt to be of little value.
Freezing funds of a suspect
The Law Commission has carefully considered the significant difficulties and detriment that the current practice of banks of freezing entire accounts where money laundering is suspected can cause. The banks feel they have no option but to freeze the entire account, as money is fungible and the law does not provide a mechanism for them either to freeze only the suspicious amount, or allow to be used any legitimate amount (such as a regular salary payment). The Law Commission proposes now that banks should be permitted to ringfence funds to the amount of any suspected criminal property where the amount is clear and readily ascertainable. It also proposes that there should be no offence under s327-329 POCA for an employee of a credit institution who suspects (or, in the new proposed world, has reasonable grounds to suspect) the funds are the benefit from criminal conduct but where that suspicion relates only to a portion of those funds, if funds are transferred, but only if (i) they are transferred to an account within the same institution or, (ii) the account does not fall below the level of the suspected funds, provided the transaction is conducted in the course of business in the regulated sector and is done to preserve criminal property. The Law Commission believes this will be of help to banks, and will limit the number of SARs proffered as a defence against money laundering. It also mentions that the definition of criminal property in POCA could perhaps be amended, but that this creates wider issues outside its remit.
The word "consent"
The final substantive issue the paper covers is whether the word "consent" is the most appropriate word to use in relation to a SAR, or whether an expression like "immunity", "waiver", "exemption" or similar would be better. On the whole, the Law Commission seeks little benefit in a change like this, but suggests there should be official guidance on what "appropriate consent" under the relevant legislative provisions actually means.
The paper covers a number of related areas which are outside the scope of this article, including information sharing and data protection. At the end of the paper is also looks at whether there is a possible alternative to the consent regime. But it seems common ground that removing the authorised disclosure exemption would leave those who come into contact with criminal property exposed. So there is the possibility of amending the threshold for the actual offences to a higher degree of fault while retaining the suspicion threshold for reporting. This could lessen the exposure while still ensuring a flow of information to the authorities. If this meant moving to a "non-consent" regime it would allow minimal disruption to legitimate activity unless the enforcement agencies took further action. Of course, there are disadvantages also to this – not least the pressure it would put on the enforcement agencies, quite apart from the fact the industry has not suggested it.
Of course, the paper would not be complete without considering the introduction of a corporate offence that would bite where an employee or associate failed to report. The Law Commission does not express a strong opinion on this, merely suggesting it would also need to have a "due diligence" defence and may help to create a corporate incentive to put in place good procedures.
This article has merely highlighted some of the background to, and key proposals from, the Law Commission paper. It has been a long time in the making and contains a wealth of useful, if inconclusive, information on case law and the practices of other jurisdictions.
For now, the consultation period closes on 5 October. The Law Commission will then need to evaluate the responses (and there are likely to be many), before presenting its final recommendations to the Government – which will then need to decide what action to take. All this means that, although change is almost certainly on the way, it will not happen quickly.
This article was written for Financial Regulation International, an Informa Group publication.