Risk Management: Flawed Fantasy Or Achievable Challenge?

By Brendan Hogan, a Masters of Applied Finance student

Lights are flashing. The screams of children are echoing off the walls. The cabin has become an ecosystem of fear and trepidation. There has been no word from the cockpit in over 20 minutes and the rattling of the Rolls-Royce engines mimic the velocity of an earthquake. A flight stewardess appears from the galley and assures the passengers everything is going to be alright. She asks everyone to put their life vests on as a precaution: “standard procedure,” she says.

Thousands of flights circle the globe every day, but very few descend to a deadly fate. Risk management is much the same. The chances of a catastrophic market crash are minute, but as history tells us, they do happen. How do we see foresee these events happening? Well, just as it’s hard to predict the safety of your flight, it’s hard to measure the risk of a crash of the finance kind. While we have invented flawed models like ‘Value at Risk’ (VaR) to measure risk, the truth is we are fantasising if we think it’s achievable.

In fact, one would be living in a fantasy land, complete with talking unicorns and houses made from candy if one would believe risk management was simple and could predict the unpredictable. There is no sure-fire way to measure risk.

In 1996 the Basel Committee on banking supervision made VaR the mandatory measure to determine credit, market and operational risk. This has been reaffirmed in the 2004 and 2013 amendments of the Basel Accords Agreement. During this time VaR has been accepted by many as the only relevant measure of risk exposure, therefore solidifying its standing as ‘best practices’.

Advocates of VaR talk of the objective nature of the measure; given a set of historical data, everyone will agree on the outcome. VaR is also intuitive. It allows risk managers to think of risk in terms of dollars of loss, whereas standard deviation defines risk in terms of deviation (either above or below expected return). Finally, it is universally applicable, that is, it can be applied to any asset class.
However, VaR’s use of historical data, generally just 250 scenarios, to compute future positions and its assumption of a normal distribution (in some cases) is perhaps its greatest travesty. Whilst these assumptions ensure the simplicity of the measure, it also means that VaR is likely to give a bad approximation when returns are not ‘normal’.

VaR’s true purpose is also commonly misinterpreted in the marketplace. It was never meant to accurately predict all possible risk and yet risk managers and traders have continued a seemingly monogamous relationship with the model.

My main problem with VaR is not so much that its forecasts won’t be accurate, but more that it can be quite easy to get a VaR that is very low. You essentially just need a portfolio of assets that have recently enjoyed benevolent calm or little correlation with each other. If a risk manager is able to construct such a cluster, the model will scream ‘sound and riskless’ operation. Hence the notion that VaR has, in essence, been providing risk managers and traders with an ‘alibi’ for their deficiencies. This is exactly what was going on in the fiery pits of Wall Street all those years prior to late 2007. According to VaR, things were looking pretty darn good, but we all know how that ended.

Due to these reasons, and not surprisingly, there are many well publicised critics of the VaR model; none more so than acclaimed full-time ‘VaR critic’ Nassim Taleb. When Taleb is not touring the globe speaking of the atrocity that is VaR, he is putting out books like his 2007 effort ‘The Black Swan’. A ‘Black Swan’ event can be described as a surprise event which has a major effect and is often inappropriately rationalised after the fact with the benefit of hindsight. Taleb contends that banks and trading firms are very vulnerable to hazardous Black Swan events and are exposed to losses beyond those that are predicted by their defective financial models. History is testimony to this when we consider the financial catastrophes of 1987, 1997, 2001 and 2008. What has happened in the past doesn’t necessarily tell us what could happen tomorrow. In fact, it rarely does.

Taleb’s arguments are strong, and for the most part warranted, but what Taleb doesn’t offer is a viable solution and alternative to identify and mitigate ‘fat-tail’ risks.

So what are the alternatives? Well that’s just it; they are few and far between. While critics are quick to shoot down VaR, they offer no viable alternative. One of the few alternatives to VaR which has had some strong support is the ‘Expected Shortfall’ methodology. The Expected Shortfall methodology is more sensitive to the shape of the loss distribution in the tail of the distribution than VaR. However, it carries greater complexity than traditional VaR methods and requires greater computing power and back testing. Whilst some organisations are using Expected Shortfall internally, regulators are still advocating the use of VaR and its importance as a risk measure is therefore unlikely to diminish in the short term.

There is no such thing as a perfect model (Heidi Klum aside) and if we are being honest, truly ‘managing’ risk is a flawed fantasy. We are hell bent on using a flawed model. Whilst VaR may provide us with a sense of security, in reality no amount of risk management will be able to protect investors against a market crash. There is no sure way to predict the future or lessen the blow of a Black Swan event. Consider the use of a life vest on the doomed plane: if the plane hits the water do you really think the life vests will be of any real help? Nevertheless, they give passengers a sense of security and perhaps that is all people desire…

Tags: , , , , , ,

Comments are closed.

Privacy | Copyright matters | Accessibility
Contact us | Feedback | Disclaimer
Opinions expressed in this blog are those of the individual contributors only.
QUT Home | Blog Home