Could you provide a timeline of how the Swiss franc price action unfolded on what the industry now calls Black Thursday?
It all happened so quickly that I don’t think it would be a very long timeline at all. At 11:30 on January 15th, the SNB removed its 1.20 floor on EUR/CHF. Around 47 seconds later the pair dipped below 1.20 and was trading just above 1.17. Ten seconds after that it tumbled harder and faster than any major pair in the history of FX. Due to the fact that most positions in the market were long, the artificial floor having represented an easy money trade for so long, a lot of people got burned when it moved the other way.
Where did client stop losses get executed?
As an agency broker all of our flow necessarily went to the banks. All of our clients’ stops were triggered and executed but due to the illiquidity the banks were experiencing they found it increasingly difficult to handle this flow. Our clients’ orders were all executed from 1.19 down to 1. Despite the exceptionally difficult market conditions we managed to get our trades covered at an average price of 1.11. All this at a time when many other brokers were filling their clients at prices below 1.
What caused the big loss at FxPro, what could have been done better?
Our losses were of course entirely down to us being an agency broker. As to what could be done to avoid such events in the future; the big discussion in the wake of Black Thursday has centred on leverage, quite erroneously in my opinion. I personally don’t see this as a situation that could’ve been managed any better with lower gearing. For example, even with a book composed solely of 1:100 orders, with 100,000,000 exposure (i.e. 1,000,000 margin) a 30% move would’ve caused a loss of 29,000,000. With 1:50 leverage the loss would’ve just been 28,000,000 rather than 29. The only thing that could’ve really been done differently is to force clients to close their positions, which is completely against FxPro’s philosophy of non-intervention. In future, if forced to choose between interfering in clients’ trades and hybridising the order book (taking on some of the flow internally) I’d say that we would be more inclined to opt for the second option.
Looking at the book of the EURCHF on the positioning graph on FxPro’s website, there has been an overwhelmingly one-sided exposure. Do you think that using out of the money options could protect brokers from massive losses at a reasonable price?
Of course hindsight is always 20/20, isn’t it? I don’t believe it’s possible to ever be completely protected from such unforeseen events. I don’t see how out of the money options would have been of much use without having a rough time window on when the SNB was planning to remove the floor. Those in the know could certainly have placed some ludicrously profitable Puts. For us it would have been a lucky bet, and we’re not in the business of bets, lucky or otherwise. We were hedged to a point, but as I say, there are no foolproof solutions when it comes to eliminating risk.
What could be the long term implications of the Swiss franc debacle on the industry as a whole?
One of the implications may be that regulators start placing an undue amount of attention on leverage as a way of protecting clients. In my view this would be a mistake as it runs counter to the demands that traders themselves have made on the industry. They vote with their trading account balances and what they seem to be choosing is regulated brokers who offer high leverage but also protect them from negative balances. I can’t stress enough that leverage was not the real issue here; even with 1:30 leverage the loss in the previous example would’ve been 27,000,000 rather than 29. Not to mention the fact that even though we do offer high leverage ratios to our clients, this in no way means that the majority of EUR/CHF positions were utilizing maximum leverage, they simply were not.
One thing I would personally like to see is for regulators to start looking a little closer at the way market makers execute orders, especially during such volatile conditions. According to our calculations there was more than enough time for them to fill their clients’ stops at or around 1.18, which means that a significant amount of unnecessary negative slippage was passed on. If the move had gone the other way would they have passed on the positive slippage? Or would they have just filled orders at the declared price? I think we all know the answer to that question.
Finally, and perhaps most importantly, the events of January 15 have shown us that 100% agency model brokers are exceptionally vulnerable in certain market conditions. The move to agency execution was predominantly an attempt create unimpeachable business models where solidarity with the end client is a given. On Black Thursday the market exacted a huge toll for this luxury. As a result I think some sort of hybrid arrangement, where we can continue to guarantee no conflicts of interest while also mitigating the impact of such outlier events, is ideal.
Is the decentralized way in which the FX market is structured the right way, or do we need a centralized approach? Why?
If you’re talking about taking FX exclusively on-exchange, I think the notion of a centralized global market is a huge contradiction in terms. It would be impossible for FX as it currently functions to be completely centralized. If anything the overarching trend is for increasing decentralization, rather than the other way around. Decentralized systems have repeatedly proven themselves to be more efficient and transparent, as well as anti-fragile. In contrast, overly-centralized systems tend to be opaque, inefficient, and highly vulnerable to attack.
Do you think that client funds reporting mechanisms with the FCA are adequate?
I do think they are adequate, but it’s important to stress that company culture is probably the most important factor here. In this respect I think the regulators often find themselves in an uphill struggle with certain companies that are unwilling or unable to change. At FxPro we have strict rules and procedures, for us transparency has always been paramount, so we try to go over and above our regulatory requirements to ensure that clients know they are dealing with a broker they can trust.
How do you feel the industry should tackle any incoming regulatory backlash?
FxPro has always held that the only way forward for the industry is increased transparency, fair trading practices, adequate capitalization, negative balance protection and complete segregation of client funds. With these things in place and the appropriate company culture to support them, we feel regulatory backlashes wouldn’t have to occur in the first place.
What losses did FxPro sustain from the event?
I’ve seen wildly differing reports in the media. I’d like to take this opportunity to state that none have been accurate. The most important thing is that our operations have not been affected in the slightest. We are sufficiently well capitalized to be able to weather such storms and our clients can still enjoy the most professional conditions in the industry. In addition, we have honored our commitment to negative balance protection, fulfilled all of our regulatory requirements, both with CySEC and the FCA, and have rapidly moved to consolidate our position in the market.
Have you changed your risk management policies in the aftermath of the event? How?
A company should never take on more risk than it can afford to. This, of course, hasn’t changed, if anything the recent CHF debacle has demonstrated that many brokers were not sufficiently well capitalized and many will now have to rethink if they can compete at this level. At FxPro we have made several changes in the way we monitor currencies, particularly those that are either pegged to EUR or USD.
There is an ongoing debate about prime brokers using pre-trade risk management checks instead of post-trade in order to avoid a repeating of the debacle which caused so many industry wide losses. Do you think that would solve the bulk of the issues?
I think that just as risk management will change in the retail space, it will also change among the prime brokers. Having said that I feel that the only thing we are likely to see is increased margin requirements until a sense of risk-tolerance returns. It’s unlikely that we will see abnormal requirements continue for very long after the underlying market conditions have returned to normal. This is another reason why I think hybrid models will come further into play. In the current climate they make as much sense from a competitive standpoint as they do from one purely concerned with risk management.