Glossary term
Value at Risk (VaR)
Subject: Risk Management, Risk Assessment and Quantification
Also known as: VaR
Value at Risk is the loss that a portfolio is not expected to exceed over a set period at a stated confidence level, such as 99% over one day.
Explanation
A one-day 99% VaR of 5 million means that, under the model, losses should be larger than 5 million on only about one day in a hundred. It condenses a whole portfolio's market risk into a single figure that managers and boards can compare with limits. Three methods are common: historical simulation using past price moves, the variance-covariance method using assumed statistical distributions, and Monte Carlo simulation. Banks have long used it for trading limits and market risk capital, and backtest it against actual results. Its main limit is silence about the size of losses beyond the threshold, which is why expected shortfall now complements or replaces it. It also depends on past data, so it can look calm just before a crisis. Limit-setting, daily reporting and capital rules all use this single figure, which gave firms a shared language for market risk.
Example
A trading desk reports a one-day 99% VaR of 4 against a limit of 5. Positions are increased on Tuesday and the figure rises to 5.6. The risk team notifies the desk head, who must reduce positions by close of day, and the breach is logged for the risk committee.
Not to be confused with
- Expected Shortfall
- Expected shortfall averages the losses beyond the VaR threshold. VaR only marks where that threshold sits.