FRM Part I · FRM Exam Part I
Operational Risk: formula sheet
Key formulas
- Basel definition of operational risk
- Operational risk = risk of loss from inadequate or failed internal processes, people and systems, or from external events
- Includes legal risk. Excludes strategic and reputational risk.
- Seven Basel event types
- 1 Internal fraud | 2 External fraud | 3 Employment practices and workplace safety | 4 Clients, products and business practices | 5 Damage to physical assets | 6 Business disruption and system failures | 7 Execution, delivery and process management
- Memorise the names. Questions give a scenario and ask for the category.
- Four causes in the definition
- Processes + People + Systems + External events
- A loss from any of these four sources is operational.
- Three lines of defense
- 1st line = own and manage risk | 2nd line = oversee, challenge, frame | 3rd line = independent assurance
- Internal audit is always the third line. The risk function must stay independent of the business it oversees.
- Board vs management
- Board = approves framework and risk appetite | Senior management = implements and resources it
- Board approves and oversees. It does not run day-to-day controls.
- Risk appetite vs tolerance
- Appetite = amount and type of risk accepted | Tolerance = acceptable deviation, set through limits, KRIs and triggers
- Appetite is broad and strategic. Tolerance is operational and measurable.
- Core framework tools
- Identify and assess: RCSA, loss data, scenario analysis, KRIs | Respond: control, mitigate, transfer, avoid, accept
- Remember them as the toolkit behind the Basel identification and assessment principle.
- Residual risk
- Residual risk = Inherent risk − Effect of controls
- A conceptual relationship, not a precise calculation. Rated qualitatively or on a scale in RCSA.
- Expected annual loss for a risk
- Expected annual loss = Frequency per year × Average severity
- Used in scenario analysis and risk ranking. Assumes the frequency and severity estimates are for the same event type.
- Tool orientation
- Loss data = backward-looking; RCSA, scenarios, KRIs = forward-looking
- The most tested distinction. Loss data is objective but weak on rare events; scenarios cover the tail but are subjective.
- KRI versus KPI
- KRI = signals rising risk; KPI = signals performance against target
- A KRI should have thresholds that trigger escalation.
- Basic indicator approach
- K(BIA) = α × [Σ(GI₁…ₙ) ÷ n], α = 15%
- GI is annual gross income. Include only years where GI is positive; n is the number of those years, out of the previous three.
- Standardized approach
- K(TSA) = {Σ over years 1–3 of max[Σ(GI × β), 0]} ÷ 3
- β = 12%, 15% or 18% depending on the business line. Yearly total floored at zero in Basel II.
- Beta factors by business line
- 18%: corporate finance, trading and sales, payment and settlement; 15%: commercial banking, agency services; 12%: retail banking, asset management, retail brokerage
- Eight business lines in total. Learn the 18% group first.
- SMA capital
- K(SMA) = BIC × ILM
- For banks in the larger size bucket. BIC comes from the business indicator with increasing marginal coefficients. ILM reflects average historical loss relative to BIC.
- Risk-weighted assets equivalent
- RWA = 12.5 × operational risk capital
- Capital is converted to RWA by multiplying by 12.5 (1 ÷ 8%).
- Poisson frequency
- P(N = n) = e^(-λ) × λ^n ÷ n!
- Mean and variance both equal λ. P(N = 0) = e^(-λ).
- Aggregate loss
- S = X1 + X2 + ... + XN
- Assumes severities are i.i.d. and independent of N.
- Expected aggregate loss
- E[S] = E[N] × E[X]
- Holds when N is independent of the i.i.d. severities.
- Variance of aggregate loss
- Var(S) = E[N] × Var(X) + Var(N) × (E[X])²
- For Poisson frequency this simplifies to Var(S) = λ × E[X²], where E[X²] = Var(X) + (E[X])².
- Lognormal severity mean
- E[X] = exp(μ + σ²/2)
- μ and σ are the mean and standard deviation of ln X, not of X. The median is exp(μ).
- Operational VaR
- VaR(α) = the α-quantile of S
- Basel AMA used α = 99.9% over one year.
- Unexpected loss
- UL = VaR(α) − E[S]
- Capital is intended to cover UL.
- Single-loss approximation
- VaR(α) ≈ F⁻¹(1 − (1 − α) ÷ λ)
- F is the severity CDF. It is an approximation for heavy-tailed severity and high α, with Poisson frequency. It is not exact.
- Net loss after insurance
- Net loss = Gross loss − Insurance recovery (after deductible, up to the policy limit)
- Recovery = min(max(Loss − Deductible, 0), Policy limit), assuming a simple policy with no co-insurance.
- Operational risk categories (Basel event types)
- Internal fraud; External fraud; Employment practices and workplace safety; Clients, products and business practices; Damage to physical assets; Business disruption and system failures; Execution, delivery and process management
- Match each case to one category; Barings and Société Générale are internal fraud.
- Control layers
- Prevent → Detect → Respond → Transfer
- Use this order to match a failure to the right mitigation.
- Expected value of an insured loss
- Expected net loss = Σ (probability × net loss after recovery)
- Use it to compare retained loss with and without insurance.
Quick revision
- Operational risk is loss from inadequate or failed processes, people, systems, or external events.
- Basel's definition includes legal risk but excludes strategic and reputational risk.
- Know the Basel event types, such as internal fraud, external fraud, and damage to physical assets.
- Three lines of defense: business units, independent risk function, internal audit.
- Risk appetite and tolerance are set by the board and shape limits and indicators.
- Key risk indicators are forward-looking signals; loss data is backward-looking.
- Risk and control self-assessments rely on business judgment and can be biased.
- Capital approaches move from simple formulas to more risk-sensitive measures; know the inputs of each.
- Expected loss for a period = expected frequency × expected severity.
- Frequency is usually modeled with a count distribution and severity with a fat-tailed one.
- Capital is based on a high percentile of the aggregate loss distribution, less expected loss where the rules allow.
- In case studies, find the control failure first, then match the mitigation.
Common mistakes
- Including reputational and strategic risk in the Basel definition. Fix: Remember legal risk is in, strategic and reputational risk are out.
- Classifying a rogue trader loss as market risk. Fix: Look at the root cause. Unauthorised, concealed trading is internal fraud.
- Putting the operational risk management function in the first line. Fix: It is the independent second line. The first line is the business and its support functions, which own the risk.
- Saying internal audit owns or manages operational risk controls. Fix: Audit gives independent assurance as the third line. It does not own risk or design day-to-day controls.
- Treating KRIs and KPIs as the same thing. Fix: Ask what the metric signals. A KRI warns about rising risk and has escalation triggers. A KPI tracks performance against a target.
- Calling loss event data forward-looking. Fix: Loss data is historical. It is objective but thin on rare, severe events, so scenarios and external data supplement it.
- Dividing by three when one year has negative gross income under BIA. Fix: Exclude non-positive years from both numerator and denominator.
- Using 15% for every business line under the standardized approach. Fix: Look up each line's beta: 12%, 15% or 18%.
- Using Var(S) = λ × Var(X) for a Poisson frequency. Fix: Use λ × E[X²], which equals λ × (Var(X) + (E[X])²). The general form is E[N] × Var(X) + Var(N) × (E[X])².
- Treating μ in a lognormal as the mean loss. Fix: μ and σ describe ln X. The mean loss is exp(μ + σ²/2) and the median is exp(μ).
Exam tips
- Learn the seven names in order and attach one example to each. Questions give a scenario, not the label.
- Always classify by root cause, not by where the loss appears in the accounts.
- Memorise the inclusion and exclusion: legal risk in, strategic and reputational risk out.
- Watch for distractors that are really market or credit events dressed up as operational ones.
- Expect the paper to be 100 multiple-choice questions in 4 hours. This topic usually needs only a quick classification, so do not spend long on it.
- Learn each function's verb: first line owns, second line challenges and designs, third line assures, board approves, management implements.
- When a question stresses independence or reporting to the audit committee, the answer is usually internal audit.
- Questions on Basel principles are usually conceptual. Match the scenario to the theme, such as culture, appetite or reporting, instead of recalling principle numbers.