The risk of risk management
Risk management is about identifying and quantifying risks. Based on this, management measures are established. Risks with a high impact are especially relevant. By analogy with Newton’s second law (impact is mass times velocity), the impact of a risk is the result of the probability times the consequence. A large consequence with a minimal chance is generally not important, just like a large chance with a minimal consequence. As a result, risk managers remarkably often have an eye for the wrong details. This is because risk managers – they are like people – put a lot of emphasis on risks that occur frequently. In the financial world, risks are usually about exposure to financial risks. Now, these risks are all too often estimated as if they were normally distributed. The principle of Value at Risk or VaR is a method of determining the financial risk of a position over a given period of time, again focusing on risks that occur frequently. VaR provides an estimate of the potential loss and the likelihood of that loss occurring. Risks that occur frequently are thus accepted and are no longer a risk. The real risk is something that is not seen as a risk, but for which you have to look outside the normal distribution.
The remarkable thing is that thanks to VaR, the real risk, permanent loss of wealth, actually increases. Indeed, VaR provides a false sense of security, as fat tails and skewed distributions are common in financial markets. An important explanation for this is the use of leverage. Working with borrowed money can have devastating consequences on the normal distribution. In this respect, another phenomenon from physics is relevant and that is stability. Without stability, there is no matter. One of the successes of quantum mechanics is that it explains the stability of atoms. When there are many neutrons in an atom, the atom becomes unstable, and eventually, the nucleus splits, converting part of the mass into energy. A lot of energy, because energy is the mass times the speed of light squared. The speed of light is three hundred million metres per second. The square of that is 90 trillion. Stability is therefore not something static, but dynamic. Something appears stable for a long time, but ultimately it only takes a small change to create chaos. Time plays an important role. So systems that appear stable for a long time, and to which more and more leverage is added, become increasingly unstable. Hyman Minsky has translated this process to the financial world. According to Minsky, stability destabilises. In an environment that appears stable, the chances (the return) are small and the consequences (if things go wrong) are also small. Based on VaR, there are then no risks. This invites people to take more risk, usually by using leverage. This leverage makes a system less stable. Whereas in a calm, stable-looking situation it is possible to borrow heavily, in times of chaos credit can be withdrawn completely. The longer a system appears stable, the more leverage is applied, making it increasingly unstable. The moment this instability becomes visible, there is a Minsky moment. Chaos ensues, which ultimately forms the basis for future stability. So there is a cycle.
More than ten years after the Great Financial Crisis, one would think that the financial world would have learned its lessons, but the long period of stability seems to be causing the financial system to become unstable again. Because of the apparent stability, risk managers seem to be asleep and the use of leverage is increasing. What Greensill and Archegos have in common is the irresponsible use of leverage. Even now that stock markets are at an all-time high and credit spreads are extremely low, the leverage used is already causing instability. In both cases, the risk managers involved appear to be blind in vision, but thanks to VaR, they probably pay a great deal of attention to the wrong details. Risk managers also have a tendency to be very precise, which makes it difficult to oversee the bigger picture. You then quickly miss the elephant in the room. Once the VaR models have identified this risk, it is too late. A wave of selling then follows and the real risk is realised, which is permanent loss of assets. The realisation of the loss ensures that everyone knows again what risk is. This time, no Great Financial Crisis is needed. For the stability of the system, it is occasionally good that there is a renewed awareness of what risk really is. And let’s finally stop with the VaR false security.