In this post we will discuss what VaR is, why it is used and what it can – and cannot – tell us; from a user point of view.
VaR has three main goals:
- It puts a figure on the risk taken by a business, in normal circumstances.
- It allows consolidating many different businesses and their related market risks into one nice number.
- It enforces a disciplined and organised way of looking at the books, with the option being available to cap the risk appetite with risk limits – VaR or notional based.
Normal as in “in normal market conditions”; this means “as the market has behaved in a more or less distant past, over a certain period”. It is always based on the past behaviour of the portfolio/book constituents. Therefore, by nature, VaR will NEVER capture extreme market situations, unexpected events or “fat tail” risks
Just one number
It produces one number that encapsulates a certain vision of the loss that a portfolio made of many different assets could experience, under normal market conditions. It is not a magic number that tells a manager all about his risks.
Widely promoted (if not imposed) by regulators
This has successfully allowed them to:
- Provide a number to evaluate of risk on which they could base capital requirements calculations;
- Put pressure on Financial Institutions to get organised and pay “some attention” to their risks. How many banks did not have a risk department when VaR was imposed?
Not a standalone number
It was not and is not supposed to be the only number to use. However, as a trend, it can be interesting. A growing VaR number is also a clear sign that “something is happening” and that some investigation is worth doing. It should focus minds and trigger discussions. Another benefit is that producing VaR numbers requires a high quality infrastructure, which can deliver other risk/p&l analysis and enforces operational discipline.
VaR gives a view on the relative contribution of various assets or portfolios. It is a tool. It cannot produce a full picture of what the future could be. It is an attempt to anticipate what could happen in the future with an evaluation based on “rules”. That means you must be aware of the assumptions in order to understand its limitations.
All VaR does is give us an estimate of the maximum loss a portfolio should experience, under a certain probability; assuming the markets behave as they usually do.
Academics and practitioners have established a variety of numbers and concepts to capture other aspects of risk. Stress tests, scenario analysis and historical analysis all provide useful information, with some trying to address more specifically the infamous “the fat tail” risks. However, they only tell part of the story: using VaR coupled with these measures will contribute to producing a clearer, more all encompassing picture of risk. But nothing offers a full insurance against any possible future events.
VaR is part of the toolbox to understand risk
The first job of a CEO of a financial institution is to make sure the company under his watch survives to fight another day; and not to bet the shop in order to increase profits. It is impossible to foresee how risk will materialise. The only thing that experienced people know is that it will. Risk measures will help keep it under control, with resulting consequences at reasonable levels; good judgement – and sensible sizing- will make sure the unexpected is not fatal. This is ultimately the talent of management, the role and responsibility of the captain of the ship. Risk management is just a helper; and VaR part of the toolbox.