By Dale Stevens, business development director for Capital Markets at SAS UK
The case for real-time risk management was made clear in report published by Lepus earlier this week. The report is based on interviews with senior representatives from four global banks and two independent consultants with extensive past experience in the field of risk. It finds that banks have reacted strongly in the face of the financial crisis, but that there are a number of inadequacies with the current approach to risk management.
Firstly, value at risk (VaR) models are unreliable for rare but extreme market convulsions. VaR’s statistical approach is not without merit, but it’s insufficient on its own. Secondly, end of day VaR creates a disconnect between the risk profile as seen by the chief risk officer (CRO) and what is actually happening at the trader level. This makes it difficult to ensure that the front office is adhering to the risk appetite prescribed by the board.
Thirdly, while traders can manage their own positions and portfolios on a near real-time basis, it is the aggregation of the firm’s position that gives the greatest transparency, but this aggregated view is difficult to achieve because of the siloed structure of many firms. Finally, the report questions why the process of making real-time, or at least intraday, risk assessments can’t be applied across entire enterprises, when some executives have access to at least parts of the entire picture at such intervals.
It is this question that really needs addressing, because current trends require a solution where actions are consistent with business strategies. This can’t be achieved by working in the way that many organisations do at the moment. For example, widespread institutional reliance on VaR is problematic. If traders do not have the right technology solutions in place to analyse and break down VaR in near real-time, relying on it to provide an accurate indication of risk is nothing more than a gamble. Ultimately banks do have the risk appetite framework that is needed to move forward and meet the demands of the regulators. However, it is the technology that now needs attention – greater focus is needed in this area to bring the technology up to the speed necessary to meet the demands.
At present, most institutions are focusing on cementing their overnight, or at best, intraday, processes. With data coming in from around the world, bringing all systems together and aggregating data from a variety of different pricing functions into one complete data set is no mean feat.
In addition to this, CROs are coming under increasing pressure and are facing new regulation being put in place to prevent a repeat of the situation of autumn 2007. The UK Financial Services Authority (FSA) has imposed a firmer stress testing regime, the Basel committee is modifying its internal model approach and the implications of Dodd Frank are only just beginning to be felt. Not only that, but regulators are now ready to apply a premium to capital held by those institutions not deemed to be adequately controlling intraday risk. Pre-trade risk management is something that is on the horizon too.
Banks operating without near real-time information are at a serious disadvantage and these regulatory drivers are making the need for a ‘real-time’ solution to risk management more ‘real’ than ever.
However, it’s not only regulation that is driving the move towards real-time risk management. The ability to price risk more quickly than competitors also drives profitability. This is a significant opportunity and further illustrates how important the move towards real time is going to be in the next few months.
The benefits of this approach are clear. A near real-time (NRT) solution would facilitate the completion of critical stress testing scenarios. It would enable better planning for the impact of changes in the value of financial instruments held. As a result, there would be fewer losses. Overall, risk taking would be a far more managed process. Firm-wide stress testing and scenario analyses would enable firms to better prepare for market shocks and manage positions inside target risk levels. Such a solution would also minimise liquidity constraints caused by unplanned reductions in instruments and position values. It would reduce counterparty risks and also make the allocation of capital far more efficient.
There are of course challenges in achieving this risk management ‘nirvana’. These include data quality and achieving the required level of high volume data access and processing. However, the technology overhaul suggested in the report does not have to mean a replacement or upgrade to existing systems. Rather it should be a universal platform that connects throughout the bank, integrating all the necessary data on demand and using the latest high speed technology.
Imagine a world with real-time access to information and with the ability to provide a view across silos. One in which both continuous real-time ‘what if’ scenario testing and continuous real-time assumption testing can be carried out. There could be integrated scenario analysis with no manual intervention and real time automatic updates of all trading activities – and all with no need for replacement of existing systems. While to many this may seem more hope than reality, watch this space over the next few months because the importance of real-time risk management is set to grow and grow.