scenario analysis

1 December 2009

The academic approach to financial risk management, and indeed all risk management, includes several key steps.

The material at the ACT teaches the following steps:

Identification
Assessment
Evaluation
Response
Reporting / Feedback

The evaluation of risk in recent years has become more and more scientific and more and more based on numerical analysis. The science and underlying mathematics has moved through several phases. From the simple variance, co-variance model that produces VaR to modern stochastic calculus and even more advanced proprietary models, it is now very difficult for treasurers to really understand them. Some measures even moved into the regulatory sphere. The US SEC required public corporations to disclose quantitative information about their derivatives activity in 1997 and many chose VaR to do this. Basel II required banks to calculate VaR as a risk measure and so this became part of the culture.

Of course, the credit crunch has caused confidence in these methods to reduce or even evaporate. However the models never promise to work under all conditions but this has produced one quote:

[These models are] “an airbag that works all the time, except when you have a car accident”.

Treasurers work in a world of both numbers and emotions and need to be able to evaluate risk in different ways.

Scenario analysis is one very useful way to approach the evaluation of risk. Scenario analysis is simply a method of asking:

“What happens if……?”

The scenario can be anything, from moves in FX rates or interest rates to state bankruptcies, bank insolvencies, staff resignation, travel problems, internet failure, share price falls, downgrading or any other question at all.

It is simple and so can be explained to staff at all levels in an organisation. It can use lists which are easy to absorb and perhaps most importantly, because the risk is actually addressed and an outcome is specified, it points towards looking out for the event as well as often suggesting a response. It is distinctly non academic.