Real-Time Risk Management is a Must

It seems that every day, the industry is becoming more obsessed with speed. And for good reason: The faster we can conduct various business processes—especially those pertaining to trading and risk management—the better, right?

The speed-in-trading issue has been adequately covered in the pages of Waters over the past few months, and this is set to continue, as latency of incoming feeds and outgoing execution instructions are incrementally eroded.

But real-time risk management is another issue altogether. We’ve covered it on a number of occasions over the years—Clive Davidson’s feature on page 23, which explores how banks address the issue of calculating risk exposures in as close to real time as possible, is our most recent iteration—even though my residual mood at the back end of each of these investigations could best be described as one of frustration. You see, I like happy endings and unambiguous resolutions. I like being able to draw a line under a subject, knowing that for the time being at least, we’ve got to the bottom of it so that I can move onto the next thing. But I’ve learned over the years that real-time risk management is not one of those issues you can neatly pigeon hole into a clearly defined and universally understood space in the financial services industry. Part of the problem is that the ‘real-time’ label, when attached to that of risk management, becomes something of a misnomer in the sense that extrapolating meaningful risk measures across business units, asset classes and counterparties in real time is simply not feasible. Not by a long shot.

What exacerbates this feeling of frustration is that there are large numbers of technology vendors serving both the buy and sell side who claim to be able to provide the underlying architecture that supports such initiatives. But when you come to challenge such claims, you find that what they’re talking about is real-time compliance or real-time market surveillance, which are significantly removed from real-time risk management. Of course, they would argue that automatic monitoring of a trader’s positions is, by association, a risk management function, given that the technology is designed to intervene as a way of mitigating the risk of traders breaching their limits. But this is a far cry from running tens of thousands of scenario analyses on a real-time basis, which, when it comes to high-frequency trading shops is simply too complex to even start to comprehend. So, for now at least, there appears to be no end in sight to my source of frustration.