ABOUT THE AUTHOR: Rani Sawaf is the CEO of Centroid Solutions Ltd. (Switzerland).
Centroid24 is performance and risk management software.
After writing about the crucial time when risk management suddenly matters, a commenter asked for a specific example. His question went as follows:
Dear Mr. Sawaf,
Can you please give me an example how exactly a broker should evaluate the risks?
Let’s suppose that a FX company has 40M capital
Their clients hold 100M long EURCHF – fully hedged
What needs to be done in terms of risk on 14-th of Jan in order to avoid a bankrupt on 15-th where some banks quoted below 0.05?
Before answering in detail, it should be said that I have now the vantage point of being more knowledgeable after the event that occurred. However, this case does teach an important lesson.
So How to evaluate the risk? This first leads us to an overview of the risk we are facing.
I mainly see the following direct and indirect financial risks relevant in this case:
1) Market Risk – bound to the artificial price of the CHF
2) “Clients defaulting” – going to negative equity
3) Contractual risk – asymmetric “ruling” between retail client contracts and liquidity provider contracts
What to Look for in a Liquidity ProviderGo to article >>
Market risk: Ideally, brokers would have identified positions in CHF (EUR/CHF, USD/CHF as major ones) as being at risk due to the support of the Swiss National Bank (SNB), and judged the EUR/CHF as an “artificial price”.
Having seen levels around parity just before the SNB intervention in 2011, it would be a possible scenario to immediately see such levels again without support of the SNB. Reflecting this in millions, we have an order of magnitude of 20M for the EUR/CHF only (1.20 to 1.00). Additional impact of CHF pairs would have to be considered as well (e.g. USD/CHF).
“Clients Defaulting”: The big losses that many brokerage companies suffered has been generated out of (hedged) positions that remained open with the counterparty (liquidity providers) while all long positions were closed due to margin calls, this as a result of the pricing gap of around 2,500 pips.
If we take your example, you do not give further details on client’s equity. Let’s assume that the 100M EUR/CHF long is being held by 500 clients with an average equity of 10,000 EUR. So we would dispose beyond the margin call of normally 2% (2M), a total of 5M of client’s equity. This still leaves us with a loss of 15M.
Contractual Risk: It will be very difficult for most brokers to claim the negative equity back from their clients, whereas they remain accountable toward their liquidity providers.
What could have been done in such a case to prevent heavy losses or even bankruptcy?
We can distinguish various measures (pre-event):
1) Have an adequate capitalization in accordance with the exposure taken, assessing what capital would be left in relation to scenarios.
2) Lower the allowed leverage for the CHF cross-currency pairs. Assuming a leverage of 100, we would lower the leverage to the height of 20% to around 80. This would trigger a reduction of the number of long contracts by the same amount (assuming clients to do not want to inject more liquidity) or if willing to stay would have to inject more equity. This is in my view the most effective measure if willing to stay with the positions.
3) Raise the call margin level for the CHF cross-currency pairs, e.g. from 2% up to 10%. This would just reduce the amount of the loss.
4) Work with options instead of hedging the same number of lots, buying put and writing calls on the EUR/CHF and USD/CHF in function of the equity and the direction of the specific instrument (providing contractual risks cannot be mitigated)
5) Pass the risk (pure STP) of CHF cross-currency pairs entirely and keep the commission.
6) A radical measure would be the avoidance of dealing with the CHF (as some brokers avoided/avoid dealing in Turkish lira or ruble).
There are aspects of the risk that can be computed/managed by a system (e.g. level of effective exposures, adequacy of capital risk, etc.) others however remain – at least for the time being – a human judgment and task. I once again point out the fact that the VaR95% is not the magic formula which solves problems and certainly not the event that occurs.
It is however an important indicator, as the altimeter is to an aircraft, in addition to speed and inclination, and of course it cannot be considered alone. Also similar to the airline industry, risk management is a learning process, where anticipation is possible, but in some cases unfortunately not.