On January 15, the Swiss National Bank (SNB) abandoned its minimum exchange-rate policy with the euro, triggering the most volatile day of trading in the FX markets for decades.
While a number of FX brokers suffered terrible losses alongside their clients, among the most high-profile casualties was retail broker FXCM, which was forced to take out a $300 million loan from Leucadia National Corporation to ensure compliance with capital rules.
Euromoney speaks with Drew Niv, CEO of FXCM, to discover what went wrong, how the firm hopes to navigate the next crisis and broader lessons for the FX market.
Post-SNB policy shift, there are many concerns surrounding FX retail brokers, including pure agency models, with some arguing brokers in effect bet against their customers’ positions – other than indulging in legitimate hedging – and their income is effectively derived from client losses, combined with commissions on trades. Could you respond to this claim?
It is true that many retail FX brokers do benefit from customer losses as a business model. This is why FXCM launched its agency business model in 2007 to shift the paradigm in the industry. In FXCM’s no-dealing-desk execution system, trades are matched one for one with a liquidity provider.
For example, when a client enters a EUR/CHF trade with FXCM, FXCM has an identical trade with one of our liquidity providers. When a client profits in the trade, FXCM gives the profits to the customer. However, when the client is not profitable on that trade, FXCM ends up having to pay an equivalent amount of the loss to the liquidity provider.
In the case of the SNB event, while clients could not cover their margin call with FXCM, we still had to cover the same margin call with our liquidity providers. The failure of the banks to provide pricing during the event ultimately resulted in some clients having negative balances with FXCM.
Could you explain if and how you manage client positioning and balance books internally, either through crossing and internationalization?
We do not cross or internalize retail FX orders. Each and every trade is passed on to the best price shown by a third-party liquidity provider. FXCM makes its money from a commission or mark-up that is fixed and attached to the trade. We make the same amount of money whether the customer wins or loses.
How much turnover is required to make up the cost of capital?
We do not look at our business as balance-sheet intensive, like a bank, and we do not measure it that way. FXCM is a high fixed-cost business and so our ebitda margins have varied between 25% and 50% in the past few years, depending on FX trading volatility. The higher the volatility, the higher our margins.
Many levered retail investors placed bets, assuming the downside would be 100% backstopped by FXCM. Some market players suggest that aggressive marketing language adopted by retail FX brokers nurtured this impression. In your view, how did this assumption in the marketplace develop?
We believe FXCM’s system operated properly during the January 15 event. FXCM has sophisticated online trading technology that incorporates extraordinarily robust risk-management systems of our client’s positions which prevented highly levered investors from losing more money than they had in their account.
The software would mark to market the customer’s account on a real-time basis and if at any time the margin requirement was breached, the system would automatically liquidate the client’s positions. I cannot presently recall a specific instance of FXCM enforcing its right to collect negative balances on the rare cases where we had negative balances.
However, I also cannot recall an FX market event as extraordinary as what occurred on January 15.
|If the SNB can abruptly abandon its long-standing policy that |
it had just reaffirmed to the world, spitting in the face of what
a western country is supposed to embody, then what about
the other currencies that carry significant risk…?Drew Niv
Every FXCM client agrees when they open a trading account to be liable for negative balances in certain situations, particularly negative balances resulting from extraordinary market events. The client agreements clearly and expressly provide FXCM with such right and put clients on notice of such potential liability. Accordingly, given the extraordinary market events on January 15, FXCM believes it has the right to pursue debit balances of certain clients.
During this historic move, liquidity from banks became extremely scarce and shallow, which affected execution prices. Liquidating these trades in normal conditions would generally be quick, but instead took much longer because banks were not making markets. This was a flash crash. The move by the SNB affected the entire FX industry globally – both retail and institutional markets – and more than $10 billion was lost by financial institutions.
Unlike previous major market moving events where liquidity was not frozen, during the SNB event the global FX market ceased to function properly. The market saw pricing go from 1.20 to well below 0.8.
It should be noted that FXCM has systems in place to prevent clients from trading on off market pricing and it was these systems that allowed most of our clients to exit their trades between 1.02 and 1.04 and avoid extreme lows of the event. We understand that this was still a very large decline, but this was not an FXCM system failure. This was a global market function failure.
The Commodity Futures Trading Commission (CFTC) is considering tighter regulations on how FX brokers handle transactions that originate overseas, where retail leverage limits are less onerous than the US. Given the complex cross-border nature of FX flows, from the perspective of the end-user, broker and clearer, would efforts to impose rules on retail foreign-currency dealers be workable in practice?
At this time I have not seen a full proposal from the CFTC, so I am not able to comment.
Were the events of Black Thursday a vindication for the dealing-desk model?
I would not call it a vindication of the dealing-desk model. I think, if anything, it showed it is the absolute failure of the institutional FX market to function during a time of disaster. Dealing-desk-model brokers do not substantially rely on the market, if at all, so they were able to take profits and write the losses off very easily.
Remember, FX market data is not widely available so most people do not understand the epic failure and lack of liquidity that caused the market to malfunction. This was a failure far surpassing any stress put on the system during the financial crisis in 2008, where FX did better than most other asset classes.
With the lack of industry-wide standards regarding circuit-breakers or any kind of market-engagement rules, an event of this magnitude caused this flash crash. EUR/CHF bids were as low as 0.2 in the minutes that followed the event, spreads were 3000 pips wide, and market chaos was far from short or temporary. If the pair was EUR/USD, it would have been even more of a catastrophe.
Post-SNB policy shift, the dangers of volatile FX moves, imperiling leveraged positions, have been laid bare. What are some lessons with respect to FX market structure and retail FX flows?
The leverage debate is somewhat irrelevant. As you know, the EUR/CHF move was the largest move of a major currency since currencies started floating in 1971, reflecting a change of over 40 standard deviations. Even with lower leverage, the industry still would have seen a large hit.
The bigger questions are, is the risk of trading anything but G10 currencies too high? If the Swiss National Bank can abruptly abandon its long-standing policy that it had just reaffirmed to the world, spitting in the face of what a western country is supposed to embody, then what about the other currencies that carry significant risk due to overactive manipulation by their respective governments by a floor, ceiling, peg or band?
Given what happened with EUR/CHF, the industry is now looking very hard at potential similar issues – especially given the increased geopolitical risks in southern and eastern Europe. Typically, if a non-G10 country were to fix a currency rate, it is unrealistic and a black market is created. The black-market rate becomes the real rate at which the currency is trading.
|Swiss franc: special focus|
Given that this was Switzerland and not Iraq, the market presumed that the floor guaranteed by the SNB has some basis in reality and would be adjusted in an orderly way. Obviously this was proved false and the result showed markets that government manipulation of the free market has painful consequences.
Think about interest rates, which are another example of government manipulation that can cause significant financial consequences if moved abruptly.
Some currencies that FXCM will be removing include DKK, SEK, NOK, SGD, HKD, PLN and CZK. In the retail business outside of G10 currencies, we are only keeping TRY, ZAR and CNH, and increasing those margin requirements; the pairs we are removing from the platform are not material to our volume or our revenue.
Again, basic standards for protection in other asset classes like circuit-breakers do not exist in FX and would have helped mitigate the damage of the SNB disaster.
You have been diversifying into new businesses in recent years, but have now indicated you will sell non-core assets to repay the loan. Has this event fundamentally altered your longer-term strategy?
It is widely known and understood that FXCM’s core business has always been retail FX; it makes up the majority [77% of Q3 reporting] of FXCM’s revenue. So yes, FXCM will be changing course and focusing on our retail FX and CFD business.
We will, however, still offer some margin institutional business like our Prime of Prime offering. We announced that we anticipate that with the proceeds from the sale of certain non-core assets and continued earnings we can meet both near- and long-term obligations of our financing while preserving the strength of our franchise.
It makes sense to sell the assets that are not our core business. The institutional businesses are also easier to sell as they fit many other businesses. While FXCM will not be selling all institutional businesses, the company will be selling its two HFT assets, its stake in FastMatch, and a variety of other smaller institutional businesses while we discontinue the institutional department at FXCM, FXCM Pro.
We believe the most obvious pathway to restoring shareholder value will be paying off the Leucadia loan as quickly as possible, and we believe the sale of certain non-core assets is the easiest way to do that. We are looking to get the highest-possible valuations for these assets. FXCM is also considering closing or selling smaller, regulated entities that require large capital requirements, but that offers an increasingly low return on capital.
Are you interested in selling the business and, if not, how can you protect yourself from a takeover attempt, given your current low share price?
The founding partners, the board and I are not interested in selling the core retail FX and CFD business. Management and founders of FXCM still own a very large percentage of the company, I being the largest shareholder, and we are committed to restoring shareholder value in the company and have every incentive to do it.
We did put a poison pill in place to reduce the likelihood that any person or group would gain control of the company by open-market accumulation or other coercive takeover tactics.
In the future, will you have emergency, undrawn loan facilities in place to avoid having to arrange an emergency loan at such short notice?
FXCM had a $250 million credit facility in place, but unfortunately that line of credit has a material-adverse-effect clause. We would be unable to draw on a credit facility if faced with a similar situation, as we can only draw on the line if there has been no material change to the business.