CAIIB Paper 2 Module B Quiz 9

Sensitivity measures estimate the impact of a specified change in a risk factor on value or earnings. Examples include BPV, delta and other factor-specific measures.

Delta is the first-order sensitivity of an option’s value to a small change in the underlying price. It can change as market conditions change and does not capture all nonlinear effects.

Gamma is the sensitivity of delta to the underlying price. It captures curvature in option exposure and becomes important when hedging nonlinear positions or large price moves.

Vega measures how an option’s value changes with implied volatility, all else equal. Options can be exposed to volatility even if the underlying price is unchanged.

Theta measures the effect of time passing on an option’s value, holding other variables constant. The sign and magnitude depend on the option and market conditions.

Expected Shortfall estimates the average loss in the tail beyond a chosen quantile. It complements VaR by conveying information about the severity of tail outcomes.

VaR, stress tests, sensitivity measures, limits and scenario analysis each reveal different dimensions of risk. A suite of measures reduces overreliance on one model or assumption.

A stop-loss limit defines a loss threshold that prompts specified action, such as reducing or closing a position or escalating for approval. It is a control, not a guarantee against further losses.

Concentration limits help prevent excessive exposure to correlated loss drivers. They should be supported by monitoring, escalation and an institution-wide view of connected exposures.

Metrics need consistent definitions, reliable data, a stated horizon and confidence or scenario assumptions where relevant. This makes comparisons and escalation meaningful.

A useful breach report explains the extent and duration of the breach, root cause, potential impact, responsible owner, escalation and actions to restore compliance.

Continuous or appropriately frequent monitoring helps identify changing exposures, limit breaches and emerging threats promptly. Frequency should be proportionate to the risk and speed of change.

Correlation describes statistical co-movement and does not establish causality. It can change over time and may rise during market stress, affecting portfolio diversification and VaR.

A heat map places risks according to assessed likelihood and impact to support prioritisation. Ratings depend on consistent criteria and should not replace detailed analysis of material risks.

Data integrity depends on validation, reconciliation, access controls, audit trails and traceable lineage from source systems to reports. Errors in data can distort risk decisions.

Materiality depends on potential impact and context. Banks should identify and escalate risks that could meaningfully affect financial condition, critical operations or stakeholder confidence.

Warning thresholds provide time to act before a formal limit breach. The owner should investigate drivers, assess exposure and escalate according to policy, even if no loss has yet occurred.

New products, systems, acquisitions or operating models can change the risk profile. Policies and controls should be reassessed to ensure they remain suitable and responsibilities are clear.

Control testing evaluates whether a control addresses the intended risk and operates consistently in practice. Findings should be documented, rated and remediated with follow-up.

Risk appetite can be broader than individual limits. An exposure can be formally within a limit yet inconsistent with the bank’s stated appetite, requiring analysis, escalation and management action.