Christoffer Sagild
As new regulations on trading kick in, Christoffer Sagild, Senior Risk Manager at Saxo Bank, argues that the industry needs to up its game when it comes to risk and margin calculation.
It already began in the wake of the 2007-2008 financial crisis. The counterparty credit risk on OTC derivatives was not accounted for properly, and it was exactly the absence of appropriate assessments of credit exposure and no consideration of default risk that were the main reasons for the severity of the crisis.
The crisis caused several regulatory changes that made OTC derivatives trading much more complex as it required careful consideration on all aspects of credit risk.
The same greater scrutiny was put on Central Counterparty Clearing Houses (CCPs) because of their key role in risk management. Historically, the margin calculations performed by CCPs had only covered market risk in normal market situations. Now, they also need to cover liquidity risk and credit risk. A CCP like CME uses an old in-house developed margin methodology called SPAN, adapted by many other CCPs as well, but the model is not very well suited for discretionary add-ons, and the industry is now seeing a move to VaR-methodologies, which incorporates these added risks better and more consistently.
One thing is the calculation model itself, another is a requirement of speed in the calculations. There is an increased demand for the ability to mitigate counterparty credit risk intraday, which means that there is an increased demand for the ability to calculate risk real-time. Historically, variation margin was calculated and posted once a day.
Real-time margin calculations obviously require computing power and proper computing techniques. Modern risk calculation models and techniques need to be in line with increased trading speeds and changing market conditions.
In the coming years, investment and development work in proper risk calculations models and computing techniques is going to be of utmost importance in the field of quantitative finance, and risk management in particular, where we will see an increasing amount of CCPs moving away from outdated risk and margin models.
From a Clearing Member’s (CM) point of view, or any other counterparty subject to margin requirements, the tradeoff between accuracy and stability/predictability in the requirements is key. With high accuracy follows an unstable margin requirement, since it constantly adjusts to market moves. The accuracy is important because it seeks to optimise the funding requirement. But stability and predictability are equally important, since it allows for CMs to be prepared to post additional collateral and have predictable funding schedules.
The higher accuracy of the VaR-methodologies introduces more unpredictability, because of the much more complex nature compared to e.g. the SPAN model or other simple scenario models. Stability measures should be a part of the model development as well.
In addition to this, procyclicality is another major issue that needs to be considered in margin methodologies. Procyclicality measures can also be considered as part of the tradeoff against accuracy to prevent having to raise margins when volatility increases, in effect creating even more volatility.
How are quant finance pioneers achieving more accurate results?
Watch our interviews with:
Lorenzo Bergomi >>
Blanka Horvath >>
Jessica James >>