This guide develops the decisions behind enterprise risk frameworks, operational controls, credit exposure, market measurement, liquidity, funds transfer pricing, and professional conduct. Twenty original four-choice questions include one linked five-question case; sixty lessons explain mechanisms, calculations, and historical-case applications. It follows the supplied post-August-2024 programme and March 2025 guidebook for the requested 17 October 2026 scope. The published exam has 84 questions in four hours, including four five-question cases; this is a representative preview. Editorial allocations follow supplied item counts. Source citations establish syllabus relevance; worked calculations use stated assumptions and established risk methods. No future syllabus is blended into this guide.
What this guide covers
Build PRM Exam 2 skills with 20 original questions, a linked case, and 60 practical risk management lessons.
Enterprise Risk Management Frameworks and Operational Risk Management Practices
1. Three Lines and Board Risk Accountability
The Three Lines model separates management's ownership of risk, specialist oversight and challenge, and independent assurance. Business managers own exposures and controls; risk and compliance functions provide expertise, frameworks, and challenge; internal audit assesses governance and control effectiveness independently. The board oversees strategy, appetite, and accountability without operating individual controls. External auditors, regulators, and rating agencies provide additional perspectives but do not replace internal ownership. Apply the model by tracing who makes a decision, who challenges it, and who independently verifies that the arrangement works, including escalation routes when responsibilities conflict.
Worked example: A lending head approves exposures within delegated authority, credit risk independently challenges ratings and concentrations, and internal audit tests whether approvals actually follow policy. The board receives aggregate exceptions and requires management to address repeated breaches rather than approving every individual loan.
Common mistake: Calling the risk department the owner of every risk can weaken business accountability and confuse independent challenge with transaction management.
Official syllabus reference: [1]
2. Risk Appetite, Tolerance, and Capacity
Risk appetite describes the risk an organization chooses to accept in pursuing objectives. Tolerances and limits translate that choice into actionable boundaries, while capacity describes the resources or constraints that ultimately limit risk absorption. Expected-loss appetite concerns recurring losses accepted in pricing and budgets; stress-loss capacity concerns survival under adverse outcomes. These are related but cannot substitute for one another. A useful framework links each measure to an owner, reporting frequency, escalation rule, and corrective action. Numerical boundaries must reflect the organization's circumstances rather than being presented as universal prudential thresholds.
Worked example: A bank budgets USD 3 million annual fraud losses, sets an escalation trigger at USD 4 million, and estimates that USD 20 million of sudden losses would exhaust its available operational-risk buffer. These figures serve different decisions even though all measure money.
Common mistake: A business can remain within its expected-loss budget while accumulating a concentrated stress exposure that threatens the organization's loss-absorbing capacity.
Official syllabus reference: [1]
3. COSO ERM and ISO 31000 Integration
COSO ERM connects governance, culture, strategy, performance, review, and reporting so risk influences business choices. ISO 31000 organizes risk management through principles, a supporting framework, and a process involving scope, assessment, treatment, communication, and review. Both support integration into decisions rather than a detached risk register. Their structures differ, so avoid treating their terminology as interchangeable checklists. Apply either by identifying an objective, evaluating uncertainty around that objective, selecting treatments, and incorporating results into resource allocation. A completed register alone does not demonstrate that strategy or incentives changed.
Worked example: Before entering a new payments market, management evaluates revenue alongside settlement, fraud, compliance, and service-disruption exposures. The decision includes investment in screening capacity and revised return targets, showing how risk analysis changes strategy rather than merely producing an additional report.
Common mistake: Implementing a named framework does not prove effective risk management if its assessments never influence objectives, incentives, investment decisions, or accountability.
Official syllabus reference: [1]
4. RCSA: Inherent Risk and Residual Risk
A risk and control self-assessment identifies process objectives, plausible failure events, underlying causes, existing controls, and remaining exposure. Inherent risk represents exposure before the specified controls; residual risk reflects their actual design and operating effectiveness. Distinguish a well-designed control from one consistently performed, because documented procedures do not prove execution. Assess likelihood and impact using defined scales or quantitative estimates, record evidence, and assign treatment owners. RCSA judgments can reveal emerging weaknesses but remain vulnerable to optimistic self-reporting. Independent challenge and incident evidence help test whether residual-risk assessments are credible.
Worked example: A treasury team rates duplicate-payment exposure as high before controls. A payment-system block is well designed, but testing finds that manual uploads bypass it. Residual risk therefore remains elevated until the bypass is closed and operating evidence confirms the change.
Common mistake: Reducing a risk rating because a policy exists confuses control documentation with evidence that the control actually prevents or detects failure.
Official syllabus reference: [1]
5. Operational Events: Cause, Event, and Effect
Operational-risk taxonomy distinguishes the source of a failure, the event itself, and its consequences. A weak access-control process is a cause; unauthorized payment release is an event; financial loss, customer harm, and regulatory action are effects. Categories such as internal fraud, external fraud, execution failures, and business disruption help aggregate comparable events. Classification should preserve multiple contributing causes without counting one loss repeatedly. Market or credit consequences can originate in an operational failure, so a useful taxonomy records both the initiating control breakdown and the financial exposure through which the loss materialized.
Worked example: A dealer enters a trade using the wrong currency because reference data are incorrect. Correcting the position creates an exchange-rate loss. The initiating event is a processing failure, while the exchange-rate movement determines the financial consequence and its eventual size.
Common mistake: Classifying every loss by its final financial instrument can conceal operational causes and lead management to improve the wrong control.
Official syllabus reference: [1]
6. KRI Calibration and Escalation Design
A key risk indicator measures a condition associated with exposure, control weakness, or emerging deterioration. Leading indicators, such as unresolved critical vulnerabilities, differ from lagging indicators, such as realized fraud losses. Normalize counts where business volume changes, establish reliable data definitions, and test whether movements have decision value. Thresholds should reflect risk appetite, historical behavior, operational capacity, and the cost of missed warnings. Attach each trigger to an owner and action. An indicator can be informative without predicting every loss, but frequent unexplained alerts or easily manipulated measures reduce its usefulness.
Worked example: A processing center reports thirty overdue exceptions, up from twenty. Because transaction volume doubled, the exception rate actually fell. Management therefore examines overdue exceptions per thousand transactions and the age of the oldest critical exception before deciding whether controls deteriorated.
Common mistake: An increasing raw count does not necessarily indicate worsening risk when the underlying exposure population or transaction volume has also changed.
Official syllabus reference: [1]
7. Gross Loss, Recoveries, and Event Aggregation
Operational loss data should distinguish gross loss, recoveries, timing, business activity, and the event connecting individual accounting entries. Insurance proceeds do not erase the original control failure, and delayed recovery affects liquidity separately from ultimate net loss. Related charges may belong to one event even when recorded across periods or departments; unrelated incidents should not be merged merely because they share a description. Collection thresholds and missing small losses can bias frequency and severity estimates. Preserve the original amounts and recovery records so internal analysis can reconcile economic experience to accounting and control evidence.
Worked example: One outage generates USD 400,000 compensation and USD 100,000 repair costs in different departments. An insurer later pays USD 150,000. Record a USD 500,000 gross event with a separate recovery, retaining payment dates to analyze both control failure and temporary funding demand.
Common mistake: Recording only net losses hides recovery uncertainty and can make an unchanged control environment appear safer after insurance coverage improves.
Official syllabus reference: [1]
8. Root-Cause Investigation and Barrier Analysis
A loss investigation reconstructs the event timeline, secures evidence, identifies failed barriers, and distinguishes initiating causes from conditions that amplified harm. Techniques such as fault trees, causal mapping, and repeated why-questions help organize hypotheses, but conclusions require corroboration. Barrier analysis asks which preventive, detective, and recovery controls should have interrupted the sequence and why they did not. Avoid ending the investigation at human error when system design, incentives, workload, or access rights made the error likely. Effective remediation targets a demonstrated causal mechanism and specifies evidence that the replacement barrier works.
Worked example: An operator releases the wrong payment file. Investigation shows identical filenames, absent preview totals, and a second approval that checks only the operator's email. Renaming files alone is insufficient; system-level batch validation and genuinely independent approval address separate failed barriers.
Common mistake: Naming the last person involved does not establish root cause or explain why preventive and detective controls failed to interrupt the event.
Official syllabus reference: [1]
9. Compound Loss Models: Frequency and Severity
The loss distribution approach models aggregate loss as the sum of a random number of event severities. Under independent, identically distributed severities independent of event count, expected aggregate loss is E[N]E[X]. Its variance is E[N]Var(X) + Var(N)(E[X])². A Poisson count assumes equal count mean and variance; clustering or overdispersion can require alternatives such as a negative binomial model. Severity choices must reflect skewness, reporting thresholds, and extreme events. Expected loss alone does not determine a tail quantile, and dependence between counts, severities, or business units changes aggregation.
Worked example: Suppose annual incident count is Poisson with mean six and average severity is USD 20,000. Expected annual loss is USD 120,000. If large losses cluster during outages, the independence assumption may understate tail risk even though that expected-loss calculation remains a useful baseline.
Common mistake: Multiplying average frequency by average severity estimates the mean, not a high-confidence capital amount or a worst-case annual loss.
Official syllabus reference: [1]
10. Extreme Value Theory and Threshold Selection
Peaks-over-threshold extreme value methods approximate sufficiently high threshold exceedances using a generalized Pareto distribution. The tail-shape parameter determines how rapidly extreme-loss probabilities decay; positive values permit heavy tails. A low threshold supplies more observations but can violate the tail approximation, while a high threshold reduces bias at the cost of substantial estimation uncertainty. Examine parameter stability, diagnostics, truncation, and event dependence rather than selecting a threshold solely to obtain a convenient capital result. EVT extrapolates beyond observed losses under assumptions; it does not create reliable information where data are sparse or structurally obsolete.
Worked example: An analyst compares fits above USD 100,000, USD 200,000, and USD 400,000. The highest threshold leaves only eight independent events, producing unstable tail estimates. The analyst reports sensitivity across defensible thresholds instead of presenting one precise extreme quantile as certain.
Common mistake: A statistically sophisticated tail model can still produce fragile results when threshold choices, dependent observations, or sparse exceedances dominate the estimate.
Official syllabus reference: [1]
11. Scenario Analysis and Common-Cause Dependence
Scenario analysis constructs plausible severe pathways where historical data are insufficient or the business has changed. Define the trigger, affected processes, control failures, durations, losses, and recoveries explicitly. Common causes can make separate business losses occur together, while shared remediation can create overlapping costs. Aggregate mutually exclusive outcomes or model dependencies to avoid both understatement and double counting. Expert judgments should be challenged against external events and internal exposure data. Scenarios inform resilience and capital decisions, but their assumed probabilities and severity ranges should remain distinguishable from frequencies estimated from a stable statistical sample.
Worked example: A regional power failure disables both a data center and its nearby recovery site. Treating their outages as independent understates joint disruption. The scenario includes common power dependence, remote staff availability, transaction backlog, and one shared customer-remediation program rather than counting that program twice.
Common mistake: Two risks appearing in different departmental registers does not establish independence when they share infrastructure, suppliers, customers, or an external trigger.
Official syllabus reference: [1]
12. Operational Capital: Standardized and Internal Measures
Operational-risk capital can be assessed through internal economic models or prescribed standardized methods. Internal models may use loss distributions, scenarios, and dependency assumptions to estimate a selected tail-loss measure. Standardized approaches can use business-volume indicators and, where implemented, loss-history components; details depend on the applicable jurisdiction and implementation. Regulatory capital and internal economic capital therefore need not coincide. Compare their purposes, inputs, sensitivities, and limitations rather than assuming either directly measures current control quality. Capital absorbs financial consequences, whereas preventive controls and resilience capabilities reduce event likelihood, severity, or disruption.
Worked example: A bank's standardized operational charge rises after its business expands, while internal scenario analysis shows reduced outage exposure after infrastructure diversification. Both results can be coherent: the prescribed measure responds to business scale, while the internal analysis captures a particular control improvement.
Common mistake: A larger operational capital charge does not automatically prove weaker controls, and a lower modeled charge does not establish better resilience.
Official syllabus reference: [1]
13. Impact Tolerance, RTO, and RPO
Operational resilience focuses on delivering important services through disruption. Impact tolerance expresses the maximum tolerable disruption using relevant dimensions such as duration, transaction backlog, or customer harm. Recovery time objective is the targeted time to restore a capability; recovery point objective specifies acceptable data loss measured backward from disruption. Map people, technology, information, facilities, and suppliers supporting the service, then test severe plausible failures across that chain. Internal recovery targets should support service-level tolerance, but meeting one metric does not establish overall resilience. Definitions and numerical targets must be specified for the organization concerned.
Worked example: A settlement service tolerates four hours of disruption, targets restoration within two hours, and allows ten minutes of data loss. Recovery after three hours with five minutes missing meets the stated outer tolerance and data objective but misses the restoration target.
Common mistake: Restoring a server does not necessarily restore the customer service when data integrity, staffing, network access, or downstream processing remains unavailable.
Official syllabus reference: [1]
14. Third-Party Concentration and Exit Feasibility
Outsourcing transfers activities but leaves the organization exposed to service failure, data misuse, contractual gaps, and concentration. Assess criticality, subcontractors, access rights, service evidence, portability, and the resources needed to switch providers. Two fintech vendors may depend on the same cloud region or identity service, creating hidden common-cause exposure. An exit plan is credible only if data can be extracted, replacement capacity obtained, and operations restored within acceptable disruption. Contracts and indemnities can support recovery but cannot guarantee immediate service continuity or eliminate the organization's responsibility for managing its dependencies.
Worked example: A bank uses separate vendors for fraud screening and payment initiation, but both depend on one cloud region. Its diversification claim is therefore incomplete. A practical test exports data, provisions an alternative region, and measures whether the complete payment service can restart.
Common mistake: Counting vendor names can exaggerate diversification when supposedly separate providers share infrastructure, subcontractors, proprietary formats, or other difficult-to-replace dependencies.
Official syllabus reference: [1]
15. Compliance Obligations and Control Testing
Compliance risk arises from failure to meet applicable legal, regulatory, contractual, or conduct obligations, potentially causing sanctions, remediation costs, restrictions, and reputational harm. An enterprise framework maps obligations to products, processes, accountable owners, controls, and testing evidence. Distinguish preventive controls, such as eligibility checks, from detective monitoring and corrective remediation. Regulatory change management must determine applicability and translate requirements into operating procedures rather than merely circulating updates. Testing should assess both design and execution using suitable populations and samples. Specific duties depend on jurisdiction and activity; a general framework cannot establish an unstated legal requirement.
Worked example: A firm introduces a product into a new jurisdiction. Compliance maps the applicable disclosure obligation to the onboarding process, checks template design, and samples completed customer files. Discovering missing disclosures triggers both customer remediation and correction of the automated document-generation control.
Common mistake: A completed training course or policy attestation does not establish that customer transactions consistently satisfy the obligations applicable to them.
Official syllabus reference: [1]
Credit Risk and Counterparty Credit Risk
16. Credit Lifecycle and Covenant Intervention
The credit lifecycle runs from origination and screening through underwriting, approval, documentation, disbursement, monitoring, repayment, and possible restructuring or recovery. Underwriting examines repayment capacity, business risk, leverage, cash-flow resilience, and security. Covenants create information and intervention rights, but their value depends on definitions, reporting timeliness, and enforceability. Monitoring should reassess the original repayment thesis rather than simply confirm scheduled payments. A covenant breach may signal deterioration without being identical to payment default. Appropriate action depends on contractual rights, borrower prospects, and recovery economics rather than an automatic acceleration rule.
Worked example: A distributor continues paying interest but breaches a leverage covenant after inventory accumulates. The lender updates cash-flow forecasts, verifies receivables, and evaluates a waiver with tighter reporting. Timely intervention can address deterioration before the borrower misses a contractual payment.
Common mistake: Current interest payments do not prove unchanged credit quality when cash conversion, leverage, liquidity, or covenant headroom has materially deteriorated.
Official syllabus reference: [1]
17. PD Calibration, Rating Migration, and Score Discrimination
Probability of default must specify a horizon, default definition, and conditioning information. Point-in-time estimates respond to current conditions, while through-the-cycle measures smooth cyclical variation. Rating migration matrices describe movements between credit states, including default, over a defined interval. A scorecard or machine-learning model can rank borrowers well yet produce poorly calibrated default probabilities. Discrimination measures ordering; calibration compares predicted frequencies with realized outcomes. Validate both using representative, time-separated data and examine drift. Multi-period default estimates require coherent transition or survival assumptions rather than multiplying a one-year PD indiscriminately.
Worked example: A model assigns two borrower groups PDs of 1% and 3%, while realized rates are 2% and 6%. It ranks the groups correctly but understates risk in both. Recalibration may improve probability estimates without changing the ordering of individual borrowers.
Common mistake: Strong ranking performance does not establish that predicted default probabilities are numerically accurate or suitable for pricing and expected-loss calculations.
Official syllabus reference: [1]
18. Merton Structural and Reduced-Form Credit Models
Structural credit models link default to an economic balance-sheet condition. In the basic Merton model, equity resembles a call option on firm assets, and default at debt maturity occurs when asset value is below promised debt. Asset volatility, leverage, and maturity shape default risk. Reduced-form models instead describe default arrival through an intensity process calibrated to observed credit information. Structural models provide an economic mechanism but require difficult asset-value estimates; intensity models are flexible for pricing but need not explain the firm's failure mechanism. Market-implied default probabilities also embed pricing assumptions and risk premia.
Worked example: Two firms have equal asset values and debt obligations, but one has much higher asset volatility. In the basic structural model, that firm's debt is riskier because more probability lies below the repayment boundary, even though its current accounting leverage matches the other's.
Common mistake: A default intensity calibrated from market spreads is not automatically the same as a real-world default frequency used for credit budgeting.
Official syllabus reference: [1]
19. LGD, Discounted Recoveries, and Downturn Dependence
Loss given default measures the proportion of exposure not recovered after default, accounting for the chosen recovery convention. Economic recovery analysis discounts future collections and deducts workout costs; collateral face value is not equivalent to immediate realizable proceeds. Seniority, security, legal enforceability, realization delays, and borrower condition affect recovery. Downturns can increase defaults while reducing collateral values, creating dependence between PD and LGD. Use consistent exposure dates and discount assumptions when comparing estimates. A negative accounting write-off in one period or later recovery does not by itself define the complete lifetime economic loss.
Worked example: A USD 1 million defaulted loan yields USD 700,000 after one year, with USD 50,000 recovery costs at that time. At a stipulated 5% discount rate, net recovery is about USD 619,048 today, implying economic LGD of approximately 38.1%.
Common mistake: Dividing nominal future recoveries by today's exposure ignores workout costs and the economic effect of delayed collection on the lender.
Official syllabus reference: [1]
20. EAD and Credit Conversion Factors
Exposure at default estimates the amount outstanding when default occurs, which can exceed today's balance when borrowers draw committed lines before failure. A simple facility model adds drawn exposure to a credit conversion factor multiplied by the undrawn commitment. The factor represents assumed additional utilization under the specified methodology and horizon. Product terms, cancellation rights, borrower behavior, and stress conditions influence the estimate. It is distinct from PD and LGD and should not be applied mechanically to already drawn balances. Derivative EAD requires a different treatment of market-driven exposure, netting, and collateral.
Worked example: A company has drawn USD 30 million of an USD 80 million committed line. With an assumed 40% conversion factor on the remaining USD 50 million, EAD is USD 50 million. Using today's drawn amount alone would miss the additional USD 20 million exposure.
Common mistake: An undrawn commitment can create material credit exposure even when it produces no current principal balance on the lender's balance sheet.
Official syllabus reference: [1]
21. Expected Loss, Economic Capital, and RAROC
Expected credit loss is a probability-weighted average, commonly simplified as PD × LGD × EAD when inputs are consistent. Unexpected loss concerns variation around that mean; economic capital is often defined using a selected tail-loss measure after allowing for expected loss. RAROC compares a consistently defined risk-adjusted income measure with allocated capital. Specify whether expected loss, operating costs, tax, and funding costs are already deducted to avoid double counting. A profitable loan can consume disproportionate capital through concentration or dependence, and portfolio capital need not equal the sum of standalone capital amounts.
Worked example: A loan generates USD 2 million revenue after funding and operating costs, has USD 500,000 expected loss, and receives USD 10 million economic capital. Under a pretax definition deducting expected loss, RAROC is 15%, which can be compared with the bank's stated hurdle.
Common mistake: Expected loss is neither a worst-case loss estimate nor sufficient information for calculating portfolio economic capital or risk-adjusted profitability.
Official syllabus reference: [1]
22. CreditMetrics, CreditRisk+, and Concentration
CreditMetrics-style models consider rating migration and associated changes in credit value, including default. CreditRisk+ uses an actuarial default-count framework with sector-driven variability to model portfolio losses. Their modeling boundaries differ: migration-related mark-to-market changes are central to one approach, while default-loss aggregation motivates the other. Both require assumptions about dependence, exposures, and recoveries. Concentration increases vulnerability when a large obligor, sector, geography, or common risk factor dominates losses. Exposure shares and concentration indices are useful screens, but economic concentration also depends on correlations and nonlinear loss contributions across the portfolio.
Worked example: A portfolio contains one hundred equal loans, but eighty borrowers depend on the same commodity. Borrower counts suggest granularity, while sector stress reveals common default dependence. A migration model may also recognize spread-related value losses before any borrower actually defaults.
Common mistake: A large number of legal borrowers does not guarantee diversification when their repayment capacity depends on the same underlying economic factor.
Official syllabus reference: [1]
23. Replacement Cost, PFE, and Legal Netting
Counterparty credit exposure on derivatives changes with market values. Current replacement cost measures today's positive exposure after applicable netting and collateral; potential future exposure describes a selected high percentile of future exposure. Expected exposure averages the positive exposure distribution at a future date. These measures serve different limit, pricing, and capital purposes. Legally enforceable close-out netting allows positive and negative values to offset within the relevant agreement, but economic offsets across separate counterparties do not establish legal netting. Calculate exposure by enforceable netting set before aggregating, respecting collateral allocation and applicable floors.
Worked example: Two trades with one counterparty are worth positive USD 9 million and negative USD 6 million under enforceable netting. Current pre-collateral replacement cost is USD 3 million. If the negative trade belongs to another counterparty, it cannot ordinarily reduce the first exposure.
Common mistake: A small net market position across a trading book does not imply small counterparty exposure when legal rights to offset are absent.
Official syllabus reference: [1]
24. Variation Margin, Initial Margin, and Margin Period Risk
Variation margin transfers collateral against changes in current mark-to-market exposure. Initial margin provides a buffer against adverse exposure changes during the period needed to close out or replace positions after default. Thresholds, minimum transfer amounts, settlement lags, disputes, and collateral haircuts can leave residual exposure even with regular margining. The margin period of risk links market volatility and liquidation time to that residual vulnerability. Collateral eligibility also matters: a volatile token or correlated security may be difficult to realize precisely when needed. Margin reduces exposure but introduces funding, custody, and operational demands.
Worked example: A derivative is fully variation-margined at yesterday's close, but the counterparty defaults before today's transfer. Replacement prices move by USD 2 million during a five-day close-out period. Initial margin may cover that gap, subject to its amount, accessibility, and collateral realization.
Common mistake: Daily collateral exchange does not eliminate exposure from intraday moves, settlement delays, disputed valuations, close-out periods, or inaccessible collateral arrangements.
Official syllabus reference: [1]
25. CVA and Default-Probability Increments
Unilateral credit valuation adjustment represents the discounted expected loss from counterparty default under a specified pricing measure. A discrete approximation sums discounted exposure conditional on interval default, multiplied by the interval's unconditional default probability and LGD. If exposure is assumed independent of default, ordinary expected exposure may replace conditional exposure; wrong-way dependence invalidates that simplification. Conditional default probabilities require survival weighting, while unconditional increments already contain it. Maintain consistent dates, exposure definitions, recovery assumptions, and pricing conventions. CVA changes can arise from exposure, credit spreads, recovery expectations, or dependence, even without actual default.
Worked example: For one interval, expected exposure conditional on default is USD 5 million, unconditional default probability is 2%, discount factor is 0.97, and LGD is 60%. Its CVA contribution is USD 58,200, reducing value from the counterparty-credit perspective.
Common mistake: Applying another survival factor to an already unconditional interval default probability counts survival twice and understates the calculated default-loss contribution.
Official syllabus reference: [1]
26. DVA, FVA, and MVA Boundaries
Debit valuation adjustment reflects the valuation effect of the institution's own default risk on liabilities. Funding valuation adjustment addresses specified funding costs or benefits, while margin valuation adjustment concerns funding initial margin. These adjustments depend on modeling, collateral, close-out, and accounting conventions; they cannot always be added as independent charges. Avoid double counting credit components embedded in funding spreads or costs already incorporated in another adjustment. An own-credit deterioration can create a DVA valuation benefit while worsening the institution's economic health. Distinguish valuation effects, cash funding requirements, and realized trading performance when interpreting XVA.
Worked example: A collateral agreement requires USD 8 million initial margin. With an assumed 1.5% annual funding spread and constant margin for one year, its undiscounted funding cost is USD 120,000. An MVA model would additionally reflect the margin path, survival, discounting, and chosen conventions.
Common mistake: A valuation gain from worsening own credit does not mean the institution has become safer, more liquid, or more economically profitable.
Official syllabus reference: [1]
27. Wrong-Way and Right-Way Dependence
Wrong-way risk occurs when exposure increases, or protection weakens, as counterparty credit quality deteriorates. Right-way risk describes the opposite dependence. Specific wrong-way risk arises from a direct transaction or issuer connection; general wrong-way risk can arise through broader economic factors. A collateralized loan secured by the borrower's own shares illustrates a direct link. Evaluate exposure conditional on distress rather than assuming exposure and default are independent. Ordinary market volatility or an imperfect hedge is insufficient to establish wrong-way risk unless a relationship with counterparty default or credit deterioration is identified.
Worked example: A bank buys currency protection from a domestic financial institution against a severe domestic currency collapse. If that collapse also undermines the institution's solvency, the protection becomes most valuable when payment is least reliable, creating general wrong-way dependence through a shared macroeconomic shock.
Common mistake: Collateral reduces counterparty risk only to the extent its value, enforceability, and availability remain reliable when the counterparty becomes distressed.
Official syllabus reference: [1]
28. CDS Protection and Securitization Risk Transfer
A credit default swap transfers specified credit-event losses from protection buyer to seller in exchange for premium, subject to contractual terms and seller performance. A simplified small-interval fair-spread approximation is hazard rate multiplied by LGD, with fuller pricing equating discounted premium and protection legs. Securitization funds assets through claims with different priorities and loss attachment points. Selling assets or buying protection does not guarantee complete economic risk transfer when recourse, retained tranches, basis differences, or counterparty risk remain. Distinguish funding obtained, credit risk transferred, and contingent support retained by the originator.
Worked example: A lender securitizes USD 100 million loans but retains the first USD 5 million of losses and a liquidity commitment. It obtains funding, yet retains junior credit exposure and contingent funding demands. The transaction's accounting presentation alone cannot establish the extent of risk reduction.
Common mistake: Removing an asset from a reported balance-sheet category does not prove that its credit, liquidity, or reputational support risks disappeared.
Official syllabus reference: [1]
29. CCP Default Waterfalls and Liquidity Concentration
A central counterparty interposes itself between clearing members, enabling multilateral netting and standardized margining. Default resources commonly include the defaulter's margin and default-fund contribution, specified CCP resources, and mutualized resources in an order set by the CCP's rules. Exact waterfalls and recovery tools vary. Clearing transforms bilateral exposures into exposures to the CCP and clearing infrastructure rather than eliminating risk. Large variation-margin calls, initial-margin increases, member defaults, and concentrated service dependencies can create liquidity stress. Assess access, portability, settlement arrangements, and the institution's ability to meet calls under market disruption.
Worked example: A clearing member's portfolio remains economically hedged, but volatility causes a USD 40 million initial-margin increase at its CCP. The member needs cash or eligible collateral immediately, demonstrating why reduced bilateral credit exposure can coexist with substantial centralized funding demands.
Common mistake: Central clearing does not eliminate default, liquidity, operational, or concentration risk; it changes their location, mutualization, and management arrangements.
Official syllabus reference: [1]
Market Risk, Asset Liability Management, and Funds Transfer Pricing
30. ALCO and Structural Funding Mismatch
Asset-liability management coordinates interest-rate, liquidity, funding, and balance-sheet risks rather than optimizing each product independently. An asset-liability committee typically reviews structural mismatches, assumptions, limits, hedges, and funding plans within delegated governance. Long-lived assets funded by short-term liabilities create refinancing exposure even when current interest margins are positive. Currency mismatches add dependence on foreign-exchange and cross-currency funding markets. Distinguish contractual maturity from behavioral maturity, especially for deposits and prepayable assets. Decisions require a joint view of earnings, economic value, funding availability, and capital constraints, with accountability for assumptions and exceptions.
Worked example: A bank funds five-year fixed-rate loans with three-month wholesale deposits. The initial spread is attractive, but funding must repeatedly roll and may reprice sharply. ALCO compares term funding, rate hedges, liquidity buffers, and loan repricing before approving further balance-sheet growth.
Common mistake: A positive current net interest margin does not establish sustainable profitability when assets and funding reprice or mature on different schedules.
Official syllabus reference: [1]
31. Liquidity Ladders and Survival Horizons
A liquidity ladder places cash inflows, outflows, and usable counterbalancing resources into time buckets. Calculate cumulative cash positions because a temporary early deficit cannot be repaired by an inflow arriving after payment is due. A survival horizon identifies how long resources remain sufficient under a specified stress and monetization strategy. Apply realistic haircuts, settlement delays, encumbrance, and currency restrictions to assets. Contractual ladders provide a baseline, while behavioral assumptions capture deposit runoff, commitment draws, and other contingent flows. The maximum cumulative deficit determines funding needs under the modeled timing assumptions.
Worked example: A bank starts with USD 15 million cash, pays USD 20 million tomorrow, and receives USD 30 million next week. Although total future inflows exceed outflows, tomorrow's USD 5 million deficit requires timely funding or asset monetization before the later receipt arrives.
Common mistake: Positive net cash flow over an entire month does not prevent an earlier payment failure caused by mismatched intramonth timing.
Official syllabus reference: [1]
32. LCR and NSFR Measure Different Horizons
The liquidity coverage ratio compares eligible high-quality liquid assets with modeled net cash outflows over its prescribed short stress horizon. The net stable funding ratio compares available stable funding with required stable funding associated with assets and activities over a structural horizon. They address complementary vulnerabilities: immediate stress liquidity and longer-term funding structure. Asset eligibility, weighting, inflow restrictions, implementation details, and required levels must be taken from the applicable rules rather than invented. Neither ratio proves currency-specific liquidity or successful operational monetization. Internal stress ladders remain necessary even when reported regulatory ratios appear comfortable.
Worked example: Using stipulated simplified inputs, a bank has USD 120 million eligible liquid assets and USD 100 million modeled net outflows, giving an LCR of 120%. Its stable-funding position requires a separate calculation; the short-horizon ratio cannot establish whether long assets have durable financing.
Common mistake: A strong short-term liquidity ratio does not by itself demonstrate a stable long-term funding structure or eliminate concentrated deposit-run exposure.
Official syllabus reference: [1]
33. Contingency Funding and Asset Monetization
A contingency funding plan connects stress triggers to executable actions, responsible decision makers, communication, and cash mobilization. Potential resources include asset sales, secured borrowing, committed lines, and other eligible facilities, but each needs operational readiness and conservative availability assumptions. Test ownership, encumbrance, documentation, collateral delivery, settlement timing, and currency conversion. Avoid counting one asset as available for multiple simultaneous funding actions. Market liquidity can deteriorate as funding needs rise, making yesterday's valuation an unreliable measure of cash obtainable today. Historical LIBOR and EONIA examples require legacy treatment rather than use as ordinary current funding benchmarks.
Worked example: Treasury lists USD 50 million securities as emergency liquidity, but USD 20 million are pledged and the remainder face a 10% stress haircut. Before timing effects, only USD 27 million is usable. A drill verifies that custody transfers can complete before critical payments.
Common mistake: An untested borrowing line or an asset's market value is not equivalent to immediately available cash under the institution's stress conditions.
Official syllabus reference: [1]
34. NII Sensitivity and Deposit Betas
Net interest income sensitivity measures how interest earnings change over a stated horizon under specified rate and behavioral assumptions. A simple repricing-gap approximation multiplies rate-sensitive assets minus liabilities by the rate change and relevant time fraction. More realistic analysis models reset dates, deposit betas, floors, loan prepayments, reinvestment, and balance-sheet evolution. Deposit beta measures the assumed pass-through from a reference-rate change to a deposit rate; it can differ across products and rate directions. Earnings sensitivity is distinct from discounted economic-value sensitivity, so a favorable short-term NII result can coexist with economic-value losses.
Worked example: A bank has USD 100 million floating assets that pass through a 1% rate increase fully and USD 100 million deposits with an assumed 40% beta. Holding balances constant for one year, income rises USD 1 million and expense USD 400,000, increasing NII USD 600,000.
Common mistake: Equal nominal asset and deposit balances do not imply zero earnings sensitivity when their reset timing, floors, or rate pass-through differ.
Official syllabus reference: [1]
35. Economic Value, Duration, and Convexity
Economic value of equity is the present value of assets minus liabilities and relevant off-balance-sheet cash flows under stated assumptions. Modified duration approximates proportional price sensitivity as −DΔy; convexity adds a second-order correction of one-half C(Δy)². Liability value changes must be subtracted when translating instrument changes into equity value. Duration is a local approximation and becomes less reliable for large shocks or cash flows affected by embedded options. Callable assets and prepayable loans can display changing duration and adverse convexity as customer exercise behavior shifts with interest rates.
Worked example: Assets worth USD 100 million have duration five, while liabilities worth USD 90 million have duration two. For a small 1% parallel increase, asset value falls about USD 5 million and liability value USD 1.8 million, reducing equity value about USD 3.2 million.
Common mistake: A fall in liability value offsets part of an asset-value loss when measuring equity sensitivity; adding both declines overstates that loss.
Official syllabus reference: [1]
36. Key-Rate Duration and Interest-Rate Basis
Key-rate sensitivities measure value changes from movements at selected yield-curve maturities, allowing twists and steepening to be distinguished from parallel shifts. A portfolio can have zero aggregate duration while retaining large offsetting sensitivities at different maturities. Interest-rate basis risk arises when related reference rates move differently, even at similar maturities. Construct hedges using a sensitivity matrix and solve for positions that address the relevant factors, subject to liquidity and transaction constraints. Check sign conventions carefully: a quoted DV01 may be positive risk magnitude or signed profit-and-loss sensitivity, and the hedge equations depend on that definition.
Worked example: A portfolio loses USD 10,000 per basis point at two years and gains USD 10,000 at ten years. Its parallel-shift sensitivity sums to zero, but a ten-basis-point rise only at two years produces an approximate USD 100,000 loss before nonlinear effects.
Common mistake: Neutralizing total duration does not neutralize curve twists, benchmark spreads, or sensitivities concentrated at particular points on the yield curve.
Official syllabus reference: [1]
37. Parametric, Historical, and Monte Carlo VaR
Value at risk is a loss quantile for a specified horizon, confidence level, portfolio, and modeling distribution. Parametric methods estimate it from assumed distributions and sensitivities; historical simulation revalues today's portfolio under observed historical changes; Monte Carlo simulation generates scenarios from a chosen model. Historical methods inherit the sample's regimes, while simulations inherit model and calibration assumptions. Nonlinear instruments may require full revaluation rather than linear approximations. VaR identifies a threshold, not the average loss beyond it, and time scaling by the square root of time requires restrictive assumptions about independent, stable increments.
Worked example: A USD 2 million one-day 99% VaR means the model places approximately 1% probability beyond that loss threshold for the stated portfolio and horizon. It does not imply losses beyond USD 2 million are small or that one breach must occur every hundred days.
Common mistake: VaR is neither a maximum possible loss nor a guarantee about the exact spacing or severity of future loss exceedances.
Official syllabus reference: [1]
38. Expected Shortfall and Stressed Calibration
Expected shortfall summarizes the average loss in the selected tail, using a definition that handles probability mass at the quantile consistently. It therefore conveys tail severity that VaR alone omits. Stressed VaR instead retains a quantile measure while calibrating to a specified stressed period or distribution. The two concepts change different dimensions: statistic versus calibration environment. Compare measures only when horizon, confidence, portfolio, and valuation assumptions are understood. Tail estimates can be highly uncertain with limited data, and liquidity or nonlinear behavior may require additional scenarios beyond either a quantile or tail average.
Worked example: In one hundred equally likely loss scenarios, the worst five losses are USD 2 million, USD 3 million, USD 4 million, USD 6 million, and USD 10 million. The average of that worst 5% is USD 5 million, illustrating tail severity beyond a boundary statistic.
Common mistake: Changing from VaR to expected shortfall does not automatically incorporate stressed market conditions, illiquidity, or an appropriate historical calibration period.
Official syllabus reference: [1]
39. Backtesting, Exception Clustering, and Model Leakage
Backtesting compares model forecasts with appropriately matched realized or hypothetical profit and loss. Coverage examines whether exceedance frequency is consistent with the forecast level; independence examines whether exceptions cluster, potentially indicating missing volatility dynamics. Align positions, horizons, fees, and intraday activity before interpreting failures. For machine-learning risk models, use chronological validation and prevent future information from entering training features or calibration. Good historical performance does not establish robustness to a new regime. Model governance should combine statistical tests, conceptual review, sensitivity analysis, data controls, and documented limitations rather than relying on one pass-fail statistic.
Worked example: A model has an acceptable total exception count, but every exceedance occurs during one volatile fortnight. That clustering suggests slow volatility adaptation. A revised model is evaluated on later untouched observations, avoiding the optimistic result that would follow repeated tuning against the same test period.
Common mistake: An acceptable exception total can conceal clustered failures, mismatched profit-and-loss definitions, or information leakage that makes historical performance misleading.
Official syllabus reference: [1]
40. Full-Revaluation Stress and Liquidation Horizons
Market stress testing applies severe historical, hypothetical, or reverse-designed scenarios to current exposures. Full revaluation captures nonlinear option behavior that delta approximations can miss. Liquidity-sensitive scenarios also consider wider bid-ask spreads, reduced market depth, delayed liquidation, and changing correlations. Reverse stress testing starts from a specified failure condition and identifies plausible combinations capable of causing it. Keep scenario assumptions coherent across prices, volatility, credit, and funding. A loss estimated under instantaneous liquidation may be unrealistic for a concentrated commodity or structured position whose exit itself moves prices and extends exposure duration.
Worked example: A commodity option book appears protected against a small spot move, but stress combines a price jump, higher volatility, and a longer exit period. Full repricing reveals option losses and additional margin needs that a delta-only estimate misses, despite the book's initially low net delta.
Common mistake: A stress calculation using normal-market bid-ask spreads and immediate liquidation can materially understate losses on concentrated or illiquid positions.
Official syllabus reference: [1]
41. Commodity Carry, Convenience Yield, and Basis
Commodity forward pricing reflects financing, storage, and the convenience yield from holding usable physical inventory. Under simplified continuous-carry assumptions, F = S exp[(r + u − y)T], where u is storage cost and y convenience yield. Physical scarcity can raise convenience yield and alter the forward curve. Hedge effectiveness also depends on grade, location, delivery date, transport, and benchmark alignment. A futures hedge may remove much broad price exposure while leaving local basis risk and margin liquidity demands. Storage capacity constraints and delivery obligations can invalidate convenient assumptions about frictionless arbitrage.
Worked example: A refinery hedges regional crude purchases with a global benchmark. A local pipeline outage raises regional prices while the benchmark is unchanged. The futures position does not offset that increase, illustrating location basis risk despite matching the expected number of barrels.
Common mistake: Matching physical and futures quantities does not guarantee a complete hedge when delivery location, quality, timing, or contractual reference prices differ.
Official syllabus reference: [1]
42. Balance-Sheet Optimization Under Multiple Constraints
Balance-sheet management allocates scarce capital, stable funding, liquidity, and risk capacity across activities. A high accounting return may be unattractive after recognizing funding duration, liquidity consumption, concentration, and capital requirements. Optimization should identify binding constraints and the marginal resource cost of expanding an activity, rather than ranking products by gross yield alone. Actions such as securitization, term issuance, asset sales, hedging, and repricing affect several constraints simultaneously. Their benefit depends on retained risks, transaction costs, and market access. A decision that improves one reported ratio can worsen earnings resilience or another scarce-resource constraint.
Worked example: Two assets each earn a 2% spread, but one requires twice the stable funding and materially more capital. If those resources are binding, the bank may prefer the lower-consumption asset or reprice the other, even though their unadjusted spreads are identical.
Common mistake: The highest-yielding asset is not necessarily the best use of balance-sheet capacity once funding, liquidity, capital, and concentration constraints are recognized.
Official syllabus reference: [1]
43. Matched-Maturity FTP and Embedded Options
Funds transfer pricing allocates funding and liquidity economics between central treasury and business units. Matched-maturity FTP charges an asset business for funding aligned with the asset's relevant tenor and credits liability businesses for funding supplied. Separate components can address reference interest rates, term liquidity, contingent commitments, and embedded options. Prepayable loans and behaviorally stable deposits require assumptions beyond contractual maturity. Clearly define included costs so credit losses, operating expenses, and capital charges are not counted twice. Good FTP improves pricing incentives; an overnight transfer rate for every asset can encourage long-term lending financed by unstable short-term funds.
Worked example: A five-year fixed loan earns 6%, while its matched funding and liquidity charge is 4%. The remaining 2% must still cover credit, operating, and any separate capital costs. Charging only today's overnight rate would overstate the business's sustainable contribution.
Common mistake: Using one short-term transfer rate for all products can conceal maturity transformation and reward businesses for risks actually borne by treasury.
Official syllabus reference: [1]
Case Study Practicum
44. Event Chains and Cross-Risk Transmission
Case analysis should distinguish the initiating failure from the channels that transmit and amplify damage. An operational breakdown can create an incorrect market position, whose losses trigger margin calls and then funding stress. Mapping the sequence preserves each mechanism without forcing the entire event into one risk category. Identify exposures, controls, contractual triggers, timing, and feedback loops at each stage. The useful lesson is a causal explanation that predicts how a similar situation could develop elsewhere. Merely attaching several taxonomy labels does not show which intervention would have broken the chain.
Worked example: A settlement-interface error leaves a currency hedge unexecuted. Exchange rates move adversely, the institution loses money, and counterparties demand additional collateral. Repairing the interface addresses initiation, while liquidity buffers and exposure reconciliation address different stages of the same developing event.
Common mistake: Listing operational, market, and liquidity risk without explaining their sequence does not identify the control or decision that could interrupt transmission.
Official syllabus reference: [1]
45. Barings Lessons: Independent Position Evidence
The Barings case illustrates how unauthorized trading can grow when trading and settlement responsibilities are insufficiently separated and exception accounts escape effective challenge. The transferable mechanism is control over both transaction creation and the evidence used to verify it. Independent exchange, clearing, bank, and counterparty records help establish position completeness and cash movements. Investigate persistent reconciliation differences, unusual funding demands, and profits inconsistent with the stated strategy together. The appropriate lesson extends beyond appointing another reviewer: the reviewer must receive independent data, understand the activity, and have authority to escalate unresolved discrepancies.
Worked example: A dealer reports a low-risk arbitrage strategy with unusually stable profits, while clearing cash demands rise sharply. Operations compares exchange positions with internal records and finds unmatched contracts. Connecting the funding symptom to independent position evidence reveals exposure that summary profit reports concealed.
Common mistake: A second signature adds little assurance when both reviewers depend on records and explanations controlled by the person whose activity they examine.
Official syllabus reference: [1]
46. LTCM Lessons: Convergence and Endogenous Liquidity
Convergence trades seek gains when related prices move toward an expected relationship. The LTCM experience illustrates how leverage, crowded positions, and dependence on financing can overwhelm an apparently attractive long-run valuation thesis. Spreads can widen before converging, correlations can change, and counterparties can demand collateral when liquidation is costly. Endogenous liquidity means the trader's own sale, and similar sales by others, can worsen prices and funding pressure. Evaluate gross exposures, financing terms, liquidation capacity, and common positions across counterparties instead of relying solely on small net market exposure or historically stable correlations.
Worked example: A fund buys a cheaper bond and shorts a similar expensive bond, expecting their spread to narrow. The spread widens during a market withdrawal, generating losses and margin calls. Forced sales then worsen the same spread, making the exit price dependent on the fund's actions.
Common mistake: A trade expected to converge eventually can still fail if leverage and collateral demands force liquidation before convergence occurs.
Official syllabus reference: [1]
47. Metallgesellschaft Lessons: Hedge Cash-Flow Timing
The Metallgesellschaft case highlights the distinction between the economic purpose of a hedge and the cash-flow pattern required to maintain it. Short-dated exchange-traded hedges against longer-dated customer commitments require rolling and daily settlement. Adverse interim movements can create immediate margin payments even when another leg is expected to benefit over time. Roll basis, term structure changes, customer optionality, and the ability to finance interim cash flows all matter. Analyze the complete contract structure and hedge horizon rather than concluding that an eventual offset guarantees liquidity or that every interim accounting loss proves hedge failure.
Worked example: A supplier promises fixed-price deliveries over several years and uses nearby futures as part of its hedge. Futures losses require cash today, while offsetting customer-contract benefits arrive with later deliveries. Treasury therefore models daily collateral requirements separately from the ultimate combined contract economics.
Common mistake: An economically motivated hedge can create severe funding pressure when losses settle immediately and offsetting benefits are delayed or difficult to monetize.
Official syllabus reference: [1]
48. Orange County Lessons: Leverage and Inverse Floaters
The Orange County case provides lessons about leveraged interest-rate exposure, structured securities, and reliance on short-term financing. An inverse floater pays a coupon that decreases as a reference rate rises, often with a multiplier that amplifies sensitivity. Its price behavior can be more adverse than the coupon description alone suggests, especially when leverage and embedded options are present. Borrowing to hold longer-duration assets adds funding and liquidation vulnerability. Examine effective duration, nonlinear cash flows, collateral terms, and governing investment constraints together, rather than treating a government-related underlying issuer as evidence that the investment strategy is low risk.
Worked example: A security pays a stipulated coupon of 12% minus twice a reference rate, subject to stated limits. If that rate rises from 3% to 4%, the coupon falls from 6% to 4%. Borrowing short-term to buy it can simultaneously raise funding expense.
Common mistake: Low issuer default risk does not eliminate interest-rate, embedded-option, leverage, or liquidity risk in a structured investment strategy.
Official syllabus reference: [1]
49. Bankers Trust Lessons: Embedded Leverage and Suitability
Structured-derivative disputes associated with Bankers Trust emphasize the need to understand contractual payoff formulas, embedded leverage, valuation, and client communication. A modest initial premium or attractive coupon can hide nonlinear sensitivity to rates or spreads. Decompose the instrument into simpler exposures, evaluate scenarios outside the recent trading range, and compare the payoff with the customer's stated objective and capacity. Product understanding and suitability require more than a signed document containing technical terms. In an exam scenario, distinguish market loss from the separate conduct and disclosure failures that can amplify financial and reputational consequences.
Worked example: A corporate treasurer seeks stable funding costs but receives a swap whose payment rises three times a benchmark spread beyond a trigger. A clear payoff table shows the mismatch between the stability objective and the embedded leverage before the contract is approved.
Common mistake: A low upfront cost or detailed contract does not demonstrate that a customer understands a structured payoff or that the product meets its objective.
Official syllabus reference: [1]
50. Enron Lessons: SPVs and Economic Recourse
The Enron case illustrates why risk analysis must look through legal entities and presentation to economic obligations, related-party incentives, and contingent support. A special-purpose vehicle may appear separate while guarantees, retained interests, liquidity commitments, or reputational incentives return losses to the sponsor. Identify who ultimately funds adverse cash flows and absorbs asset-value declines. Related-party transactions require particular scrutiny because nominally independent prices and decisions may share incentives. Accounting treatment and legal form are relevant evidence, but neither alone establishes effective economic risk transfer. Consolidate the scenario's obligations without double counting the same loss through several contractual channels.
Worked example: A sponsor sells assets to a vehicle but guarantees senior investors and retains the junior tranche. The sale provides cash, yet the sponsor remains exposed through two specified channels. Stress analysis traces the waterfall and guarantee conditions rather than assuming the entire asset risk disappeared.
Common mistake: Treating a legally separate vehicle as economically independent can conceal guarantees, retained losses, related-party incentives, and contingent funding commitments.
Official syllabus reference: [1]
51. Northern Rock Lessons: Wholesale Funding Dependence
Northern Rock illustrates the vulnerability of funding long-term lending through markets that must remain continuously accessible. A bank can have performing loans yet face acute stress when wholesale refinancing or securitization capacity closes. Funding concentration involves investor type, maturity, instrument, currency, and shared sensitivity to market confidence. A retail deposit run may amplify a problem that began with structural funding dependence. Examine refinancing schedules, contingent collateral needs, usable buffers, and credible alternative channels. Solvency analysis and liquidity analysis answer different questions, although forced asset sales and expensive emergency funding can transmit liquidity stress into capital losses.
Worked example: A mortgage lender expects to refinance USD 500 million of short-term borrowing through a new securitization. Investor demand disappears before maturity. Even if mortgage payments continue as forecast, the lender must replace the maturing cash funding or sell assets under pressure.
Common mistake: Performing long-term loans do not supply the immediate cash required to repay concentrated short-term creditors when refinancing markets become unavailable.
Official syllabus reference: [1]
52. Lehman Lessons: Repo Haircuts and Confidence
The Lehman experience highlights interactions among illiquid assets, leverage, secured funding, valuation uncertainty, and counterparty confidence. Repo funding can contract when lenders raise haircuts, reject collateral, shorten maturities, or decline renewal. Higher haircuts require more equity or cash for the same asset position, while forced sales can depress values further. A reported balance-sheet snapshot may miss intraperiod funding dependence and contingent collateral demands. Assess whether resources are freely available across entities and currencies. Distinguish uncertain asset solvency from immediate funding ability, while recognizing that deteriorating confidence can connect the two through adverse feedback.
Worked example: A dealer finances USD 100 million securities with a 5% haircut, borrowing USD 95 million. If the haircut rises to 15% with asset value unchanged, borrowing falls to USD 85 million. The dealer must provide USD 10 million additional funding or reduce the position.
Common mistake: Secured funding is not automatically stable because collateral eligibility, haircuts, lender confidence, and renewal decisions can change rapidly during stress.
Official syllabus reference: [1]
53. AIG Lessons: Rating Triggers and Collateral Calls
AIG-related lessons include the funding consequences of derivative collateral terms, credit-rating triggers, and concentrated credit protection. A downgrade can increase collateral requirements independently of an immediate realized default loss on referenced assets. Market deterioration may simultaneously increase derivative exposure and weaken the protection seller's funding capacity. Map contractual triggers, valuation disputes, eligible collateral, and timing under combined credit and market shocks. The distinction between ultimate asset loss and immediate collateral outflow is central. An institution can face a large liquidity requirement while contractual loss estimates remain uncertain, so capital and cash adequacy need separate stress calculations.
Worked example: A protection seller owes USD 12 million variation margin after spread changes. A simultaneous downgrade removes a USD 5 million unsecured threshold, adding another USD 5 million call. Treasury needs USD 17 million eligible collateral even though no reference borrower has yet defaulted.
Common mistake: No realized reference-asset default does not mean a credit-derivative portfolio cannot generate an immediate and substantial collateral funding requirement.
Official syllabus reference: [1]
54. Option-Valuation Failures and Volatility Surfaces
Option-valuation failures, including lessons associated with NatWest Markets, show why a model name is insufficient evidence of reliable pricing. Options with different strikes and maturities may require a volatility surface rather than one convenient volatility input. Stale parameters, unsupported extrapolation, and trader-controlled marks can create false profits. Independent price verification should challenge model inputs, calibration instruments, reserves, and valuation uncertainty, particularly for illiquid options. Distinguish model limitations from deliberate input manipulation: both can produce misvaluation, but their causal controls differ. Reconcile valuation changes with risk sensitivities and independently observable market movements.
Worked example: A trader marks a long-dated option using a short-dated volatility quote because no matching transaction exists. Independent valuation tests term-structure assumptions and alternative calibration ranges, showing that the reported profit depends heavily on an unsupported input rather than an observable trading gain.
Common mistake: A widely used pricing formula can still produce unreliable values when its volatility inputs, calibration evidence, or extrapolation assumptions are inappropriate.
Official syllabus reference: [1]
55. Rogue Trading and Fictitious Hedge Detection
Rogue-trading cases such as Société Générale demonstrate how fictitious offsetting trades can conceal large directional exposures when controls rely on internal representations. Repeated cancellations, delayed confirmations, unusual counterparties, and amendments near reporting dates can be important patterns. Verify trade existence against external confirmations, clearing records, cash settlement, and counterparty identities, while restricting inappropriate system privileges. A low reported net position is insufficient when its hedge leg may not exist. Combine market-risk analysis with operational evidence so apparent hedges receive economic recognition only when their contractual existence, terms, and enforceability are supported.
Worked example: A desk reports a large futures position offset by an OTC trade. The OTC transaction is repeatedly canceled before confirmation and rebooked with a later date. External verification finds no counterparty agreement, so risk reporting removes the unsupported offset and reveals the directional exposure.
Common mistake: Risk calculations can be mathematically correct and still conceal a dangerous position when fictitious trades are accepted as valid hedge evidence.
Official syllabus reference: [1]
56. Virtual-Asset Custody and Ownership Verification
Applying historical custody and reconciliation lessons to virtual assets requires distinguishing market value, legal entitlement, and operational control. An on-chain balance can demonstrate assets at an address without proving ownership, freedom from encumbrance, or completeness of customer liabilities. Private-key access creates operational and fraud exposure; exchange claims introduce intermediary credit and withdrawal risk. Analyze segregation, authorization, reconciliation, valuation sources, and recovery arrangements separately. Proof of reserves without corresponding liability evidence cannot establish solvency. Virtual-asset market volatility is therefore only one component of a scenario that may also involve custody, counterparty, liquidity, and governance failures.
Worked example: A platform publishes a wallet holding USD 30 million equivalent tokens but owes customers USD 45 million and has pledged part of the wallet. The visible token balance alone cannot establish customer coverage, unencumbered resources, or the platform's ability to honor withdrawals.
Common mistake: A visible blockchain address or reserve attestation does not by itself establish complete liabilities, legal ownership, unencumbered assets, or reliable customer access.
Official syllabus reference: [1]
57. Counterfactual Remediation Testing
A useful post-incident lesson specifies a control change and tests whether it would interrupt the documented failure mechanism. Counterfactual testing replays the event with the proposed barrier, including realistic timing, data, permissions, and escalation. Distinguish preventing initiation, detecting exposure earlier, and reducing eventual consequences; each can be valuable but produces a different outcome. Remediation evidence should show sustained operating effectiveness rather than task completion alone. Compare residual vulnerabilities after the change, including displacement to manual workarounds. This converts a historical narrative into a testable decision instead of assuming that additional training or reporting automatically solves the cause.
Worked example: After a missed collateral call, management adds an automated alert. Replaying the incident shows the alert would arrive after the payment deadline because source data update overnight. Moving the data feed earlier, assigning ownership, and testing escalation creates a control that can actually change the outcome.
Common mistake: Closing a remediation action proves an activity was completed, not that the revised process would prevent, detect, or contain the original failure.
Official syllabus reference: [1]
PRMIA Standards and Governance
58. Group of Thirty Derivatives Best Practices
Group of Thirty derivatives best-practice recommendations emphasize senior management oversight, clear policies, sound valuation, risk measurement, independent market and credit risk functions, documentation, and operational capability. Their practical purpose is to connect the economic complexity of derivatives with controls capable of measuring and governing it. Daily valuation and independent challenge reduce reliance on a trader's own account of performance; enforceable agreements and settlement controls address different risks. Recommendations are professional guidance rather than a universal statutory code. Apply them by identifying the relevant weakness and the concrete governance, valuation, credit, or operational practice needed to address it.
Worked example: A company begins trading swaps without independent valuation or counterparty limits. Applying these best-practice themes means establishing management-approved policies, independent price checks, exposure aggregation, credit limits, and reliable documentation before increasing activity, rather than treating a market-risk number as a complete control framework.
Common mistake: A derivatives policy is incomplete if it measures market risk while leaving valuation independence, counterparty exposure, legal documentation, or settlement controls unsupported.
Official syllabus reference: [1]
59. Governance Principles Versus Association Bylaws
Governance principles guide how authority, accountability, oversight, transparency, and stakeholder interests should operate in an organization. Bylaws serve a different function: they establish an association's constitutional arrangements and formal procedures, including how its governing bodies and membership processes operate. For PRMIA-related questions, distinguish the purpose of its governance principles from the binding organizational provisions of its bylaws and use supplied text for clause-specific decisions. Board oversight should permit effective independent challenge and clear delegation. A formal governance structure is valuable only when information, incentives, and escalation allow those assigned responsibilities to exercise them effectively.
Worked example: A case asks whether management's risk decision received adequate independent oversight. Governance principles inform that analysis. A separate question about the association's formal election process requires the relevant bylaw provision; a broad principle of transparency alone cannot determine an unstated procedural rule.
Common mistake: General governance principles cannot establish an exact voting requirement, membership power, or election procedure when the applicable bylaw text has not been supplied.
Official syllabus reference: [1]
60. Ethical Risk Reporting and Professional Judgment
Professional ethical practice requires honest representation of analysis, competent work, protection of confidential information, and appropriate management of conflicts. A risk report should disclose material assumptions and limitations so decision makers are not misled by selective metrics or false precision. Confidentiality does not authorize fabricating results or suppressing material weaknesses; escalation and disclosure decisions must respect applicable duties and authorized channels. Separate a defensible difference in modeling judgment from deliberate misrepresentation. When personal incentives, client interests, or management pressure threaten independence, disclose the conflict and implement proportionate safeguards before the affected decision is finalized.
Worked example: Management asks an analyst to remove the stress scenario producing the largest loss because it appears improbable. The analyst documents its assumptions, explains its decision relevance, and uses established escalation if exclusion would mislead the committee, while keeping customer-level confidential data within authorized channels.
Common mistake: A technically accurate number can still mislead when material limitations, adverse scenarios, conflicting incentives, or relevant uncertainty are deliberately omitted from its presentation.
Official syllabus reference: [1]