FRM Part 2 Credit Risk: Master Every Tested Concept
Why Credit Risk Is the Make-or-Break Topic on FRM Part 2
If you've mapped out the FRM Part 2 curriculum, you already know that Credit Risk is not a supporting act — it's one of the exam's heaviest content areas, accounting for roughly 20% of the total exam weight according to GARP's official topic area breakdown. That means walking into the exam with shaky credit risk fundamentals is one of the fastest ways to fail a paper you otherwise could have passed.
The frustrating reality? Credit risk on FRM Part 2 isn't just a plug-and-chug exercise. GARP constructs questions that require you to hold multiple connected concepts in your head simultaneously — default probability, loss given default, credit valuation adjustment, and how Basel III pulls all of that into a regulatory capital framework. If you've been studying definitions in isolation, that approach will cost you.
This guide walks you through the concepts that actually show up, the connections that exam writers exploit, and the mindset you need to stop losing marks you've already earned.
---
The Core Building Blocks: PD, LGD, EAD, and Expected Loss
Every credit risk question on FRM Part 2 ultimately traces back to a small set of foundational variables. Master these cold.
Probability of Default (PD)
PD represents the likelihood that a counterparty fails to meet its obligations over a given time horizon. GARP tests two distinct approaches:
- Structural models (Merton model): treat the firm's equity as a call option on its assets. Default occurs when asset value falls below the debt face value at maturity. You need to be comfortable with the intuition and the mechanics — distance to default, N(d2) as the risk-neutral PD, and how leverage amplifies default risk.
- Reduced-form models (hazard rate models): treat default as a surprise event driven by a Poisson process. Here, you work with hazard rates and survival probabilities. Know how to back out implied PDs from credit spreads using the approximation: Credit Spread ≈ PD × LGD.
A common exam trap: conflating risk-neutral PD (derived from market prices) with real-world PD (estimated from historical data). They are not the same, and GARP knows candidates who study carelessly will mix them up.
Loss Given Default (LGD) and Recovery Rate
LGD is simply 1 − Recovery Rate. If a bond recovers 40 cents on the dollar, LGD = 60%. Conceptually simple — but exam questions will embed LGD in multi-step calculations where an error early compounds downstream. Know that recovery rates vary by seniority and collateral, and be ready to use LGD in both expected loss and CVA computations.
Exposure at Default (EAD)
EAD is particularly nuanced for derivatives and revolving credit facilities, where the exposure isn't fixed. For derivatives, current exposure and potential future exposure (PFE) feed into EAD estimates. GARP will test whether you understand that EAD for a loan differs conceptually from EAD for an OTC swap.
Expected Loss (EL) and Unexpected Loss (UL)
Expected Loss = PD × LGD × EAD. Regulatory capital is held against unexpected loss — not expected loss, which is already priced into spreads and loan-loss provisions. This distinction matters for Basel questions. Candidates who blur EL and UL will consistently misread capital adequacy questions.
---
Credit Valuation Adjustment (CVA): The Concept GARP Loves to Test
CVA is the market value of counterparty credit risk embedded in a derivative position. It represents the expected cost of counterparty default, discounted to present value. The formula candidates need to internalize:
CVA ≈ LGD × Σ (Discount Factor × PD in period t × Expected Exposure in period t)
What makes CVA questions difficult isn't the formula — it's the moving parts:
- Wrong-way risk: when the counterparty's creditworthiness deteriorates precisely when your exposure to them increases. This is a specific FRM Part 2 concept with real-world examples (e.g., a bank buying CDS protection from a counterparty whose credit worsens as the reference entity deteriorates).
- DVA (Debt Valuation Adjustment): the mirror image — the value of your own default risk from the counterparty's perspective. Conceptually controversial but tested.
- Bilateral CVA: accounts for both counterparty default risk and own-default risk simultaneously.
If a question gives you exposure profiles and survival probabilities across multiple periods, slow down. These are multi-step aggregation problems, not single-formula applications.
---
Basel III Credit Risk Capital Requirements
Basel III is not just a regulatory backdrop on FRM Part 2 — it's tested directly and in depth. Know both the Standardized Approach and the Internal Ratings-Based (IRB) approaches.
Standardized Approach
Under the Standardized Approach, risk weights are assigned based on external credit ratings and asset class. Know the key risk weight categories: sovereigns, banks, corporates, retail exposures, and residential mortgages. GARP will give you an exposure and ask you to compute the risk-weighted asset (RWA) or the resulting capital charge.
Internal Ratings-Based (IRB) Approaches
The Foundation IRB approach allows banks to estimate their own PD while regulators supply LGD and EAD. The Advanced IRB approach lets banks estimate all three internally. The capital formula under IRB is complex and relies on a single-factor ASRF model — know that capital requirements increase with PD and asset correlation, and that the correlation parameters differ by asset class.
Credit Risk Mitigation
Basel also tests credit risk mitigation — netting agreements, collateral, and guarantees. Understand how a valid netting agreement reduces EAD for a derivatives portfolio, and how collateral haircuts work under both financial collateral simple and comprehensive approaches.
---
Portfolio Credit Risk: Correlation Is Everything
Isolated single-name credit risk is manageable. Portfolio credit risk — where default correlation changes the entire loss distribution — is where FRM Part 2 separates candidates.
Credit Correlation and Copulas
Default correlation measures the tendency of obligors to default together. The Gaussian copula became infamous after the financial crisis for its role in CDO pricing, but GARP tests it because it illustrates how correlation assumptions drive tail risk in credit portfolios. Understand the mechanics: marginal distributions are combined using a copula to produce a joint default distribution.
Credit Metrics and Credit VaR
CreditMetrics is a full-distribution approach to credit portfolio risk. It models rating migrations and defaults across the portfolio to produce a loss distribution, from which Credit VaR is extracted (analogous to market risk VaR, but for credit losses at a given confidence level). Know the concept, and know that Credit VaR = Unexpected Loss at a given confidence level.
---
The Study Trap That Kills FRM Part 2 Credit Risk Scores
The most dangerous study habit for this topic is reading coverage and believing you understand it. Credit risk on FRM Part 2 is a topic where surface familiarity feels like mastery — until the exam presents a question that connects three concepts at once.
The fix is deliberate: after every concept, close the notes and reconstruct it from scratch. Can you derive why risk-neutral PD differs from real-world PD? Can you explain why wrong-way risk increases CVA without looking at a formula? If not, you have a conceptual gap — not a memory gap.
This is precisely the kind of reasoning that separates candidates who pass from candidates who walk out saying "I knew that material."
---
How Clavis Trains Credit Risk Reasoning for FRM Part 2
Clavis is built specifically for candidates preparing for high-stakes finance exams — and FRM Part 2 credit risk is one of the areas where AI-native training makes the biggest difference.
Rather than presenting static question banks where you pattern-match to answers you've seen before, Clavis asks follow-up questions that force you to explain your reasoning. Get a CVA question right? Clavis probes whether you actually understand wrong-way risk or just happened to select the correct formula. Get one wrong? Clavis diagnoses whether you have a knowledge gap, a conceptual misunderstanding, or a careless error — and routes your next questions accordingly.
For FRM Part 2 credit risk, where GARP explicitly rewards candidates who can reason through novel applications of familiar frameworks, training conceptual fluency isn't optional. It's the only approach that actually works.
If you're serious about passing FRM Part 2, start building that fluency now at clavis.study.