Financial Risk Management Handbook

Definition of Financial Risk

Risk is the potential that a chosen action or activity will lead to a loss or an undesirable outcome. Risk has two components: Exposure (the maximum amount you tend to lose in the worst-case scenario) and Uncertainty (the probability that worst case scenario will happen). The basic formula is:

All risks are divided into 3 main categories: Systematic Risks (derived from macro environment — inflation, taxation, regulation, interest rates), Unsystematic Risks (arising from business operations — equity, credit, capital, valuation risks), and Fidelity Risks (arising from incorrect or incomplete information).

Market Risk & Value at Risk (VaR)

Market risk is exposure to the uncertain market value of a portfolio. The 5 standard market risks are equity risk, interest rate risk, liquidity risk, currency risk, and commodity risk. Value at Risk (VaR), introduced by J.P. Morgan in 1994, represents the maximum mark-to-market loss of a liquid portfolio over a short-term future with a certain confidence level.

VaR models include Parametric (RiskMetrics, GARCH), Non-parametric (Historical Simulation), and Semi-parametric (Extreme Value Theory, CAViaR). The handbook also covers 10 key market risk metrics: Standard Deviation, Absolute Deviation, VaR, Conditional VaR (Expected Shortfall), Marginal VaR, Incremental VaR, Sortino Ratio, Greeks (Delta, Gamma, Vega, Theta, Rho), Cross Greeks, and Measure of Five (Jensen Alpha, Beta, Treynor Ratio, Sharpe Ratio, Tracking Error).

Credit Risk Measurement

Credit risk is the risk of loss arising from a borrower who does not make payments as promised. The 3 key parameters for credit risk measurement are:

• EAD (Exposure at Default) — total amount a lender will lose in case of default, calculated under Foundation IRB and Advanced IRB approaches
• PD (Probability of Default) — likelihood that a loan will not be repaid, calculated under Standardized, Foundation IRB, and Advanced IRB approaches
• LGD (Loss Given Default) — credit loss incurred if an obligor defaults, calculated net of collateral and guarantees

The handbook covers structural credit risk models (Merton Model), reduced form models (Jarrow-Turnbull/KMV), credit derivatives (CDS, Total Return Swaps), and the complete Basel II IRB framework for credit risk capital requirements.

Basel Accord Framework (I, II & III)

The Basel Accords are banking supervision recommendations issued by the Basel Committee on Banking Supervision (BCBS). Basel I (1988) established 8% capital requirements focused on credit risk. Basel II (2004) introduced three pillars: Minimum Capital Requirements, Supervisory Review, and Market Discipline — with the formula:

Basel III (post-2008 crisis) strengthens bank capital requirements with 4.5% CET1, 6% Tier I, mandatory 2.5% capital conservation buffer, countercyclical buffer up to 2.5%, minimum 3% leverage ratio, and two liquidity ratios (LCR and NSFR).

Solvency Risk & Capital Adequacy

Solvency risk represents the risk that business losses will be greater than what can be absorbed by the capital side of balance sheet. Capital is divided into Tier I (permanent — common equity, retained earnings), Tier II (supplementary — subordinated debt, revaluation reserves), and Tier III (short-term subordinated debt).

Key ratios: Tier I Capital Ratio (≥ 6%), Capital Adequacy Ratio (≥ 10%), Leverage Ratio (≥ 5%). The handbook also covers Economic Capital (VaR-based), RAROC, RORAC, and RARORAC frameworks for risk-adjusted profitability measurement.

Model Validation & Risk Audit

Model validation follows 6 principles: assessing predictive ability, institution responsibility, iterative process, no single method, quantitative and qualitative elements, and independent review. The validation framework includes evaluation of conceptual soundness, ongoing monitoring (process verification and benchmarking), and outcomes analysis (back-testing).

Risk audit uses a phased 5-step process: preparation, assessment, mitigation, reporting, and follow-up. It covers risk assessment, asset identification, criticality/confidentiality levels, threat and vulnerability identification, and risk calculations. The handbook also covers PNL Attribution analysis using both Sensitivities and Revaluation methods.