Which of the following is a common limitation of VaR?

• A. It assumes normal distribution of returns and may misstate tail risk.
• B. It relies on current market prices only.
• C. It excludes liquidity risk completely.
• D. It relies on historical data and may underestimate tail risk.

Answer: (D)

Value at Risk is built on a model of how losses behave based on past data. If those past observations don’t capture extreme events or shifts in how markets move, the VaR estimate can understate what could happen in a future crisis. In particular, the tail of the loss distribution—losses beyond the VaR cutoff—can be much larger than what the model implies, so relying on historical data tends to underestimate tail risk. This makes it a common and important limitation to remember when using VaR for risk assessment.

The other ideas aren’t as universally true across all VaR implementations: assuming a normal distribution is only a feature of certain parametric methods and isn’t inherent to all VaR approaches; some methods don’t rely solely on current prices; and while VaR can overlook liquidity effects, that critique isn’t as central or universal as the historical-data tail-risk issue.