The FCA has fined inter-dealer broker Tullett Prebon for breaches of Principle 2 (acting with due skill care and diligence), Principle 3 (effective organisation and controls) and Principle 11 (open and co-operative dealings with the regulator) of the FCA Principles for Businesses. The issues related to failures in respect of firm systems and controls that ought to have detected and challenged endemic broker misconduct and also failures to deal appropriately with FCA information requirements.
Whilst not the focus of this article, the broker misconduct at Tullett primarily took the form of "wash trades", "three-party switch" transactions or transaction structures designed solely to generate brokerage, sometimes in return for entertainment or even promised assistance in the manipulation of LIBOR.
Breach of Principles 2 and 3
The misconduct was localised to "name passing" brokers in the Rates Division. Brokers of this type arrange transactions between counterparties and then to conclude the transaction, pass the names to those respective counterparties. In the Final Notice the FCA comment that the "name passing" business was heavily dependent on personal relationships between the Tullett brokers and the traders at the trading firms. Tullett brokers were encouraged to develop and maintain relationships with traders, including through corporate hospitality (described by the FCA as "sometimes lavish"). The brokers' bonuses were linked directly to the level of brokerage that they generated. The FCA determined that this combination of factors gave rise to a risk of broker misconduct, in particular the risk that brokers and traders might act together in order to generate unwarranted brokerage or in return for superior corporate hospitality.
Against the backdrop of this risk, the Final Notice describes Tullett's Principles 2 and 3 failures as relating to monitoring, compliance and controls. Dealing with each aspect briefly in turn:
- Monitoring. The FCA found that there was in effect no monitoring of the "name passing" brokers either by management or Compliance. Indeed a decision had been taken by management that Compliance should not monitor the "name passing" business on the basis that it presented a lower market abuse risk than other businesses. There was a system in place that could have been used for real-time monitoring of transactions. Compliance and the Group Executive Committee assumed that the system was being used for this purpose by Desk Heads and Divisional Directors. However, there was no process to ensure that this assumption was communicated as an instruction and accordingly, the system was only used to monitor revenue generation.
- Compliance. The Compliance Department and Group Executive Committee believed that Divisional and Managing Directors had central responsibility for compliance. In 2009 the Managing Director for Rates and Divisional Directors signed a statement confirming this responsibility. However, there was no specific training or understanding of what this meant in practice. There was also in practice no meaningful interaction between the Compliance Department and the Rates Division. Senior management in Rates (who were supposed to be responsible for compliance) even ignored/turned a blind eye to obvious red flags including abnormally large commissions and explanations from brokers when asked for explanations such as "you don’t want to know alright" (this in respect of a trade that was later deemed by the FCA to be a "wash trade"). As the FCA put it, the first line of defence was consequently "little more than a theoretical notion".
- Controls. Corporate entertainment controls did exist in the form of delegated authority for expenses sign-off and policies in the Compliance Manual and Handbook. However, these sources failed to provide guidance on key topics such as the level of appropriate entertaining and many brokers and managers were not even aware of them. They also failed to explain why disproportionate entertainment generated misconduct risk. Accordingly, brokers requesting sign off for expenses and managers signing off had no guidance as to what was expected and where the risks lay. There were some financial limits but these were set as a percentage of desk revenue, which discouraged scrutiny of individual claims, as did a generous limit before pre-authorisation was required (at one point set at £10,000). There was no Compliance oversight of expenditure on corporate entertainment and whilst Finance checked the claims, this was treated as an accounting rather than a compliance exercise.
Breach of Principle 11
During an FCA visit in 2011, Tullett was served with an information requirement relating to a Tullett broker. This requirement was made in respect of an unrelated LIBOR investigation. During the visit, Tullett Compliance and Legal Departments informed the FCA that Tullett operated a policy of deleting call recordings after one year. Whilst this was the policy, unbeknownst to Legal and Compliance, the Voice Communications Department in fact retained the majority of recordings dating back to 2006. A senior manager who spoke with the Communications Department that day, was told that calls involving the relevant broker were available from over one year prior, but the manager did not action this information or correct what the FCA had been told. The Communications Department were not otherwise consulted either during the FCA visit or during any of the subsequent interactions between Tullett and the FCA regarding this requirement.
The same senior manager was involved in a call with the FCA in 2012 to scope a second information requirement. During that call, the senior manager undertook to check whether calls from 2009 were available. The FCA also requested confirmation of the calls' unavailability (because of the 12 month deletion policy) in the written information requirement. However, neither the senior manager nor Compliance, who assisted in compiling the response to the requirement, checked the position with the Communications Department. If they had, they would have learned that relevant communications as far back as 2007 were available.
In 2013, in the course of exchanges between the Communications Department and the senior manager, it became apparent that communications dating back to 2007 were available. The FCA were not informed. Further, in correspondence with another law enforcement agency (also investigating the LIBOR issue) the manager inaccurately said that recordings dating back to 2006 had unexpectedly turned up in an office clear out. Despite various on-going LIBOR related interactions, the FCA was not given the accurate position in respect of the calls until April 2014. Even at that point FCA was inaccurately told (by a different manager) that the tapes had unexpectedly turned up. It was not until October 2014, in response to an information requirement aimed specifically at determining the circumstances in which the audio had come to light, that the true position emerged.
The FCA state that it was the audio eventually disclosed to them (that they had been seeking since 2011) that was the genesis of the investigation into improper trades by Tullett brokers.
It is of note (but not that surprising) that the FCA viewed the seriousness of Tullett's failures as being at level 4. It viewed the Principle 11 failures as amounting to recklessness, rather than deliberate concealment. Tullett settled the matter at stage 1 – had it not, the penalty would have been £22m.
All too many aspects of this case are startling for the reader. The blatant and endemic misconduct of the brokers (and the 'Nelsonian' blindness of their managers) is one of those aspects, the total failure of Compliance oversight is another. However, authorised firms and practitioners will likely find more meaningful guidance in the bigger picture. Tullet could have taken steps to reduce the risk of misconduct at any stage – it was neither particularly complex nor hidden. Fundamentally, though Tullett did not assess the combination of "name passing" brokers' interactions with traders and incentive structures as posing any enhanced risk (other than perhaps a market abuse risk). Had the risk been identified then it seems unlikely, for example, that guidance around expenses sign off would have been so vague or that the transactions monitoring system would have not been properly used to guard against misconduct (rather than simply watching revenue). Authorised firms, especially larger operations, must design controls and oversight systems bearing in mind the particular risk(s) that their business(es) generate.
In respect of the FCA's case under Principle 11, the obvious message must be that, especially in larger operations, real care must be taken to consult across all potentially relevant parts of the business before responses are provided to information requirements. It is also obviously the case that any errors in what the FCA is told, must be corrected promptly and accurately.