I am writing this sitting in a US airport awaiting my flight home – that airport is JFK in New York and, thanks to a short delay, I cannot help but think that just a few years ago it was here that Mark Johnson was arrested and commenced a period of hell that few, if any of us, can understand.
I understand there may be other avenues available should this fail, but Johnson’s challenge to his conviction is probably reaching its zenith with the en banc hearing request submitted last month. I wrote a few weeks ago of the feeling of helplessness that many had watching the case progress through the latter stages, but I also feel there should be a sense of anger at both the inaction of those bodies that purport to represent the FX industry and – it has to be said – the apparent acceptance by the US court system that a man’s freedom can be taken from him in spite of evidence based upon generalisations and inaccurate information, both backed up by a lack of understanding as to how markets work.
I accept that ACI – The Financial Markets Association submitted an amicus brief in support of Johnson’s appeal, but – and I see how this may appear demanding – surely there is more to be done to help deliver an en banc hearing? There were material misconceptions delivered by the appeal court bench that need rectifying, and surely an industry body is well placed to do that? I want to stress how highly I view ACI’s intervention in the first place because it stands in a very good light compared to the apparent absolute lack of interest, interaction or support from other bodies such as the Global FX Division at AFME. These bodies are funded by industry participants, it would be nice to know what they do for that money beyond a few glossy white papers.
More serious from my point of view sitting here at JFK (and undoubtedly from Mark Johnson’s viewpoint) is how the US legal system sees this latest move. The US prides itself on the fairness of its legal system and in general one has to say that it works well. There will always be the odd one that slips through the net in terms of wrongful conviction, but the percentages are very, very small. That should not mean, however, that the US becomes complacent over its record – it must maintain a vigilant stance to ensure that every defendant gets the best possible opportunity to highlight their case.
This then, brings me the Johnson case. In the latest submission the defence team highlights two errors of submission in the evidence and guidance given to the appeal bench – errors that are not disputed by the government. I accept that everyone makes mistakes (traders apparently aren’t allowed to, however) but this second admission of error by the government (the first, for which it was forced to apologise to, and was criticised by, the court, was during the bail hearing) means that at whatever level you want to put it, there is a chance that the judgement to strike down the appeal was based upon incomplete or inaccurate evidence and that should mean that a further appeal is granted.
The US rightly prides itself on the fairness of its legal system and the very, very low number of wrongful convictions, but that does not mean it can be complacent – if there have been errors, and there have, then another hearing is surely the appropriate response?
One error was the omission of the title “Cahill Buying Period” from a graph highlighting HSBC traders’ positions and the level of the market during the day of the Cairn order. The inference from the defence team is that the government strongly suggested to the appeal bench that Cahill started buying at 2.50pm (10 minutes before the fix) but in reality the chart highlighted (although did not label) that he was actually buying for an hour before that.
The second mistake is one of language. The appeal court upheld the appeal noting, “Johnson later told a colleague that his team had to ‘make sure the fix [was] below [1.5730] or [Cairn was] going to be sure [it had] been ripped off.” What Johnson actually did was stress the need to keep the fix below 1.5730 or Cairn was “gonna feel they’ve been ripped off”.
I understand why the DoJ argues that “the essential point that Johnson believed that Cairn would not complain so long as the price it ultimately paid was lower than that it would have paid under the alternate form of transaction, i.e., the full-risk transfer”, but that does not prove that the intent was to maximise the profit from this trade whilst keeping a lid on the market. It could just as easily suggest someone trying very hard to avoid a potential area of conflict.
I also find this latest statement from the government yet another example of its muddled thinking when it comes to the FX market – it argued initially that HSBC should have bought the GBP 2.25 billion in one minute, which would have destroyed the market, and then accepted that argument was nonsense (which should have been enough for the court to throw the case out – it was for the “Cartel”) and now it appears to think that one player can control the market to the degree that they can manipulate the rate to a few pips. That is nonsense – even allowing for the fact that the “Cartel” case was dismissed, the evidence there highlighted how three or four players allegedly colluding to manipulate the fix could not achieve that end on a regular basis. If three or more banks can’t do that, how can the DoJ think one can?
I have written so many times about the potentially negative impact on the FX industry from some of the judges’ decisions, but the bottom line is this case is really about the biggest obsession in FX right now – market impact. It boils down to how well this order was executed. It’s not about lying to the client as claimed by the US Government – Johnson clearly did not directly lie to Cairn (although I still have issues with some statements made by another HSBC staffer not on his team) – so it has to be about was this best execution?
At face value I would say no, it wasn’t, mainly because of what the original trial heard from HSBC sterling trader Frank Cahill that his intention, when executing the deal to, according to the original judge sentencing in New York, “was to increase the price of the British Pound, something which I believe likely had an impact on the spot price used to fill Cairn’s order”, but is this Johnson’s fault? The only evidence we appear to have is that Johnson instructed Stuart Scott, Cahill’s immediate manager, to make sure he didn’t push Cable above 1.5830 – does that prove intent? Again, it can just as easily prove he was trying to curb any intention Cahill may or may not have had to buy more aggressively. The government’s case is that Johnson should be responsible as Cahill’s ultimate manager but there were levels of management in between and he didn’t directly converse with him during the order itself so how can he be responsible? I understand chain of command logic and the buck has to stop somewhere, but it also involves senior managers trusting those below and around them.
It is here that culture comes into play and again is a reason why I think Johnson should not be facing jail – if there were mistakes or transgressions they took place more than one level below his, or on a different team. He could do something about the first – in fact he did by sending instructions to avoid buying to aggressively – but nothing about the second; another HSBC staffer talking about the Russian central bank. This is where HSBC’s culture should take over – it is the bank’s duty to check that all that took place was in accordance with its policies, not a middle manager on one desk of hundreds in the markets division. And, of course, the bank did hold an investigation and it gave the order a clean bill of health. This means then, that at the very worst, this is tantamount to a case of mistaken identity.
Going back to best execution and market impact, it would be a fascinating and, I suspect, revealing, exercise to replicate the Cairn order using modern TCA techniques. I have argued for such an approach before but even from when I did just two years ago, the data and analytics available have improved significantly – this improvement should be leveraged to achieve a fair outcome.
Data is, apparently, immutable, so therefore why have we not run the order through one of these tools to see what the expected outcome would be? It can be run several ways, take a liquidity profile of any early December day, run the simulation according to what happens if you buy in a one-minute window (mayhem); a 10-minute window (significant market impact); and a one hour window (the really interesting result).
Given how well modern TCA tools work we could probably also optimise the strategy and provide a TWAP window that would, in all probability, give us the appropriate time horizon for an order of that size at that time of year to be executed with minimal impact – I suspect the result would be longer than the notice Cairn gave HSBC.
It needs to be stressed that this will not be a guarantee of execution outcome – these things never are – but it would be better that the opinions of people unfamiliar with the business of foreign exchange (and that includes, I am sad to say, the government’s initial expert witnesses). Such an exercise would, however, give people an empirical understanding of how the market reacts to such events and help to explain what best execution would look like in this case and whether there was amplified market impact.
It is hard to escape the conclusion that I, and many others, have felt all along, that Mark Johnson is a trophy scalp for the DoJ, which was getting desperate to nail a banker for something – anything. That doesn’t help him of course, but surely the fact that there are doubts around how the evidence was presented to the appeals court means there should be another hearing?