The firm has implemented a VaR measure that approximates the non-linear behavior of options with linear polynomials, which is likely to produce misleading VaR measurements. This relates to Type A, model specification risk.
This problem relates to Type C, model application risk. Employees are supposed to be using output from the VaR measure to understand the risks being taken, but they are failing to do so.
The VaR measure is being applied in a new environment for which it is unsuited. This relates to Type C, model application risk.
This situation relates to both Type A, model specification risk, and Type C, model application risk. Obviously, the portfolio manager’s staff is “implementing” a flawed VaR measure, but the portfolio manager is also misapplying that measure by setting expectations for its output that no VaR measure could satisfy.
The VaR measure has been implemented in a manner inconsistent with its design, which relates to Type B, model implementation risk.
This situation relates to Type C, model application risk. The VaR measure is being used in a manner that is likely to mislead.