RISK MANAGEMENT MODULE C QUIZ 1

Market risk is the risk of loss arising from adverse movements in market variables such as interest rates, exchange rates, equity prices and commodity prices. A rise in yields generally reduces the price of existing fixed-rate bonds. Credit risk is about counterparty/issuer default, while operational risk arises from failed processes, people, systems or external events.

The trading book generally contains instruments held with trading intent or to hedge other trading-book positions, subject to regulatory criteria. Positions are actively managed and valued using market prices or valuation models. A loan originated and held to collect contractual cash flows is not ordinarily a trading-book position merely because its value can fluctuate.

Fixed-rate bond prices and market yields generally move inversely. When market yields rise, the present value of the bond’s fixed future cash flows falls, so its market price declines. The size of the change depends on duration, convexity, cash flows and yield movement.

A net long foreign-currency position benefits when that currency appreciates against the domestic currency and loses when it depreciates. If the rupee appreciates, one US dollar converts into fewer rupees, reducing the rupee value of the bank’s net long dollar position.

Equity price risk is the risk of loss due to adverse movements in share prices or equity indices. A fall in the market price of shares held by the bank reduces the portfolio’s value, assuming no offsetting hedge.

Commodity price risk arises from adverse movements in commodity prices such as oil, metals or agricultural products. The exact impact depends on whether the position is long or short and the contract terms; the scenario explicitly states that the position loses when crude prices rise.

A sound control structure separates front-office dealing from independent risk measurement/monitoring and back-office confirmation and settlement. The risk function should be able to challenge positions and limits independently of traders, helping reduce conflicts of interest and unauthorised exposure.

A limit breach should be identified, reported promptly and escalated under documented procedures. The bank should assess the exposure and require authorised remediation or approval where permitted. Quietly ignoring or retrospectively concealing a breach undermines the risk framework.

A one-day 99% VaR of ₹10 crore means the model estimates a 1% probability that the loss will exceed ₹10 crore over the one-day horizon, under the chosen methodology and assumptions. It does not mean the maximum possible loss is ₹10 crore, nor does it quantify the size of losses beyond that threshold.

Historical simulation applies historical changes in market risk factors to the current portfolio and derives a loss distribution from the resulting hypothetical portfolio outcomes. It does not require assuming that returns are normally distributed, but it depends on the relevance and sufficiency of the historical sample.

Parametric (variance-covariance) VaR estimates portfolio risk using statistical parameters such as volatilities and correlations, commonly under a distributional assumption such as normality. Its simplicity can be useful, but results may be less reliable when distributions are non-normal or positions have nonlinear payoffs.

Monte Carlo VaR generates many simulated scenarios for market risk factors from specified stochastic models, revalues the portfolio under those scenarios and estimates a loss quantile. The result depends on model specification, calibration and the number and quality of simulations.

VaR identifies a loss quantile at a confidence level but does not describe how severe losses may be beyond that quantile. It can also be sensitive to assumptions and historical data. Stress testing and Expected Shortfall can provide additional insight into tail risk.

Expected Shortfall estimates the average loss conditional on losses being beyond the corresponding VaR threshold, subject to the precise convention used. It therefore provides information about the tail beyond the quantile, unlike VaR alone.

Stress testing evaluates the effect of severe but plausible or otherwise relevant extreme scenarios on a portfolio. Scenarios may involve sharp rate moves, currency depreciation, equity-market falls, volatility spikes or correlations changing. It complements model-based measures such as VaR.

Backtesting compares model forecasts with realised or appropriately constructed hypothetical outcomes to assess model performance. For VaR, exceptions occur when losses exceed the VaR estimate; their frequency and pattern help identify model weaknesses and support validation.

Modified duration estimates the percentage price sensitivity of a bond to a small change in yield. Approximately, percentage price change equals negative modified duration multiplied by the yield change, with yield expressed consistently. Convexity improves the approximation for larger changes.

PVBP/DV01 measures the approximate change in portfolio value for a one-basis-point change in yield. It helps quantify interest-rate sensitivity and can be used to compare or hedge exposures, while sign conventions depend on whether the measure is expressed as gain/loss or absolute sensitivity.

Duration provides a first-order, approximately linear estimate of price sensitivity. Bond price-yield relationships are curved; convexity captures a second-order effect and improves the estimate, especially for larger yield changes. It is not a direct measure of credit quality.

Portfolio risk depends not only on each position’s individual volatility but also on how risk factors move together. Correlations may reduce diversification benefits when low, but can rise during market stress, causing losses across assets to occur together. Correlation assumptions therefore need monitoring and stress testing.