Risk managers should park their traditional models and head to the beach if they want to understand the latest techniques, writes Brian Caplen.
Risk management has been completely overhauled since the financial crisis but the biggest developments may be still to come. One problem that needs solving is that many approaches are backward looking and static, whereas complex financial systems are dynamically responding to an unfolding situation.
This has led risk management experts to draw on the natural sciences to come up with better models, especially at the supervisory level where the greatest complexity lies. The Banker’s economics editor, Silvia Pavoni, explored this in our March cover story.
Risk managers who want to be on top of all this should start their research by heading to the beach.
Consider grains of sand piling on top of each other. The pile is stable for a long time and then suddenly a single additional grain collapses everything. But can we work out which grain will do this? The science for finding out is known as agent-based modelling and is starting to be applied to financial systems to understand how they can appear stable for a long period and then brought down by a single small inflection.
Agent-based modelling has many financial applications. It can be used to determine behavioural change such as the likelihood of an individual counterparty to pull funds and cause a knock-on effect. It can also be used to model ‘fat-tail’ risks that are low-probability but with high impact. Cyber-attacks, for example, are attempted all the time but how to identify the particular one that finally breaks through?
Banks need to be sure they have the right people. Before the financial crisis they hired candidates with a PhD in maths or science to design complex derivatives – the so-called rocket scientists. In the new environment it makes sense to transfer some of these brains to risk management.
Brian Caplen is the editor of The Banker. Follow him on Twitter @BrianCaplen
Register to receive my blog and in-depth coverage from the banking industry through the weekly e-newsletter.
