A friend said to me recently, when discussing the activities in chat rooms such as the Cartel, that their institutions shied away from employing anyone in the more well-known rooms because “it didn’t pass the sniff test” – in other words, while what was going on may not have been technically or legally wrong at the time, it was unethical…it smelt.

This is, to me, a good description of how many in the industry think – including me. A court of law has found the three members of the Cartel innocent, but as I noted in this column last week, even the acquitted know that activities such as those carried out back then are unacceptable now and will be rightly punished in some shape or form.

There’s nothing like a long flight to give one the opportunity to go into matters in more depth and living in Australia I can assure you there is no such thing as a short flight! So I spent a chunk of time yesterday on a plane re-reading the documents filed by the latest class action over market manipulation. My conclusion is that the FX industry is going to have to argue the case for pre-hedging in yet another court room (actually court rooms – as we reported almost two weeks ago when the news broke, a European lawsuit is also planned).

A great deal of the evidence presented is nothing new to those of us who have followed these cases, a good 150 pages is taken up raking over the ashes of previous cases. The Plaintiffs have done additional research, however, which they claim shows that the manipulation, collusion and other alleged poor behaviour had been going on for at least five years longer than the previous class action found.

While 2008 to 2013 were the “poster” years for what they term “suspicious” price spikes around the Fix, the Plaintiffs’ research claims that in 2006 and 2007 the percentage of days likely to have been manipulated was over 40%, and in 2003 when the research starts it was 37%.

This tells me that the $2.3 billion paid by most of the banks to settle the previous class action may be dwarfed by any potential settlement of this action, should they choose to roll over and pay up again for actions they have effectively already admitted. Of course this will be great news for the lawyers, I am sure most of you saw the news that the firms in the previous class action were awarded $300 million – nice work if you can get it!

Looking at the analysis in this class action, one sentence grabbed me in my second read through – “No matter how the pricing data is looked at, it is undeniable a large, statistically significant swing in prices occurred right during the WM/Reuters Closing Rate fixing window.” 

First up, I would observe that this is what happens when you try and stick through a huge amount in any given currency pair in one minute at the tail end of the European day when liquidity is at a premium. It’s irrational and demonstrates ignorance of how the foreign exchange market operates to think that liquidity magically appears whenever it is needed – it’s tantamount to asking a trader to stand in front of a runaway steamroller.

Secondly, I would be interested to know whether there was a correlation between the “price spikes” claimed around the Fix and the net balance actually being traded. I am not for one minute condoning attempts to collude – they were crude and, to my mind, wrong – but it could be that on many of the “manipulation days” there was simply more volume going through and a heavy imbalance. In those circumstances there will be a price spike both in the window and leading up to it (because without pre-hedging the spike in the one minute window would have looked like Mount Everest alongside the Plaintiffs’ calculations, which would then resemble central Canada).

The Plantiffs argue that price clustering indicates the defendants’ traders were colluding and use data from AUD/USD to illustrate their point – which is that the price spike in the Fix window was observably higher than the average either side. Again, I have no truck with attempts to collude, but the Financial Stability Board’s analysis of the WM Fix indicated that in the one-minute window at 4pm volume was 10 times higher than an average minute. I’m sorry, but that is going to cause a price spike.

The claim also states, “Plaintiffs also considered whether the price spikes around the WM/Reuters Closing Rate fixing window were caused by an increase in liquidity in the FX market. As an initial matter, this hypothesis is implausible because, while the FX market was generally liquid around the WM/Reuters Closing Rate fixing window, it was not uniquely so. Rather, the data shows that the FX market was even more liquid at other times of the day, when the prices spikes were not seen.” (original italics) 

This kind of ignores the order book imbalance that often existed at the 4Pm Fix (and still does of course) and would appear to support a re-phrasing of my original question in Monday’s column (which continues to generate considerable feedback) which was why do you continue to use these fixes if you know they are susceptible to manipulation? to, If the FX market is even more liquid at other times of the day and does not exhibit the price spikes, why not execute your hedge then?

Whichever way you look at it, there is more than enough evidence here to support the case for a different way of doing business. Yes, the market is very liquid at 4pm London, but is it the right sort of liquidity? A market with a heavy order imbalance has high volume, but it’s not necessarily good liquidity – remember the oldest adage in FX, “The only definition of liquidity that passes muster is that you can get as much done as you like when you’re wrong.”

This lawsuit offers plenty of evidence of behaviours that we all, through a modern prism, find unacceptable, but what worries me, as I noted earlier, is that this could result in yet another attack on pre-hedging as a practice. 

Mark Johnson’s appeal has attracted some support from the industry around the appropriateness of pre-hedging and this case may end up going the same way. 

What would be interesting, now though, is how the GFXC approaches the question of pre-hedging in the Global Code if the employers of many of its members roll over and admit liability in this latest case? Are the banks, by potentially doing this, admitting that pre-hedging has no place in foreign exchange markets? If they are it will have significant consequences – not for the banks and other LPs, but for the customers.

In a world without pre-hedging, banks simply have to refuse to accept orders of a certain magnitude (and it’s not that high by the way) unless they act as an agent with the resulting slippage passed onto the customer (plus their fee of course).

In such circumstances a lot of the firms bringing this class action will be left to their own devices when executing larger orders and they will have a choice – either ask for a risk transfer price (and I am sure the relationship after a court case will be extremely robust…) or execute at the Fix or in the market themselves. In the latter instance they are open to signalling risk and a bunch of traders seeing the flow and (legally) jumping in front of it – what price transparency of order?

I mentioned the GFXC’s stance on pre-hedging if these “price spikes” before the Fix are seen to be the result of unethical behaviour (pre-hedging). Obviously the Committee’s first defence is its stance on pre-hedging states that is can only be used for the benefit of the client – but what if the client has effectively sued, in part, over the use of the practice? How can that be aligned with the Code’s principles?

Another interesting question is where do the trade associations stand on this. I remain appalled at the lack of intervention by some trade associations in the Johnson case, which, as I have long argued, is critical to the FX market’s future functioning, so where will they stand on this? After all associations like AFME are institutional-based so this would not be ignoring one individual’s plight, it would be to ignore the challenge facing most of the association’s senior members. I hope I am wrong, but their inaction over Mark Johnson may come back to hurt them, or rather their members, here.

I thought I would end on a light note by getting all “Oliver Stone” about things (so yes, this is very tongue in cheek and I am not for a minute suggesting it has, or will happen). It could just be that these associations have been directed to do nothing by their members in order that pre-hedging as a practice is eradicated. 

That seems a strange thing to write, but if you think about it, a world without pre-hedging looks remarkably similar to that of 20-plus years ago when traders “worked” an order, which meant they only needed to demonstrate best efforts. Customers appreciated skilful execution with limited slippage and a trusting relationship was built.

Pre-hedging, thanks to the use of small windows for huge trades, helped destroyed that relationship and emasculated the traders – getting rid of it would allow a good FX business to deliver value again rather than being either a fee-generating conduit in a world in which fees are inevitably squeezed, or the provider of a free option to certain customers.

There is nothing wrong with modernisation and disruption, however we should never ignore that some things just work well and don’t need radical change. The FX market without an overwhelming focus on certain windows was a healthy, vibrant place, with plenty of participants providing good, solid liquidity. A return to that would not necessarily be the worst thing to happen for everybody concerned (except the lawyers).

So perhaps this latest action should be the catalyst for a radical re-think of how the FX market operates, not just around benchmarks as I suggested on Monday. Perhaps we need to look at the great technology and market structure change it has empowered and ask how we change our behaviour to suit that?

I wrote recently, when discussing risk warehousing, that we are, as an industry, letting the technology dictate how we act. When it comes to executing certain trades perhaps we need to embrace the technology – especially around execution quality analysis – and change how we act accordingly.

Of course what is needed to kick start this is an acceptance by a significant proportion of the client base that they need to change how they operate – and it is at this stage that I sigh sadly and exit stage left… 


Twitter @lamboPnL

Twitter @Profit_and_Loss

Colin Lambert

Share This

Share on facebook
Share on google
Share on twitter
Share on linkedin
Share on reddit

Related Posts in