Market Risk vs Credit Risk vs Model Risk: Risk Management Roles and Interview Questions
Updated 2026-10-27
Market risk covers losses from price moves, and its core tools are VaR, stress testing and the Greeks. Credit risk covers losses when a borrower or counterparty defaults, built around PD, LGD, EAD and credit analysis. Model risk checks whether the bank's other models are right, and leans on statistics and programming.
If you are strong in programming and statistics, look at model risk and market risk first; if you like reading financial statements and analyzing companies, look at credit risk first. Interview questions differ a lot by sub-area, so choosing one before you prepare works much better than memorizing generic risk answers.
Five risk sub-areas in one table
| What you do | Core concepts and tools | Background that helps | |
|---|---|---|---|
| Market risk | Run daily VaR and stress tests, monitor trading desk limits, explain changes in P&L and risk numbers | VaR (historical, parametric, Monte Carlo), Expected Shortfall, Greeks, duration and convexity, backtesting, FRTB | Financial engineering, math, statistics; derivatives knowledge; Python or SQL |
| Credit risk | Assess whether companies or individuals will default, approve credit, monitor loan portfolios, estimate expected loss and reserves | PD, LGD, EAD, expected loss, rating migration, IFRS 9 / CECL, counterparty risk (CVA, wrong-way risk) | Finance, accounting, economics; reading the three statements; credit analysis |
| Model risk / validation | Independently test the assumptions, data and code of pricing, risk and credit scoring models, and write validation reports | Backtesting, sensitivity analysis, benchmark models, model risk management guidance (SR 11-7) | Statistics, math, financial engineering, CS; the highest coding bar |
| Liquidity risk | Check whether the bank has enough cash and liquid assets under stress, and run funding gap analysis | LCR, NSFR, HQLA, funding gaps, deposit outflow assumptions | Finance, economics; a feel for the balance sheet |
| Operational risk | Record and analyze losses from failed processes, systems and human error, and test controls | Loss event data, risk and control self-assessment (RCSA), key risk indicators (KRI), scenario analysis | The widest range of backgrounds; attention to detail and communication matter most |
Banks organize these differently: some put counterparty risk under credit risk and some make it a separate team; some place model validation inside risk and some keep it independent. Go by the job description.
What market risk interviews ask
- What VaR is, the pros and cons of the three calculation methods, its limitations, and why regulators moved toward Expected Shortfall.
- How to estimate a 10-day 99% VaR from a 1-day 95% VaR, and what has to be true for that scaling to hold.
- Given the standalone VaR of two assets, the largest and smallest possible portfolio VaR and the correlation behind each.
- Why a historical simulation VaR might barely move the day after a market crash.
- Six losses beyond 99% VaR in 250 days: can the model still be used, and how do you explain it to the head of the desk.
- How much a portfolio with duration 5 moves when rates rise 50 basis points, and when convexity matters.
What credit risk interviews ask
- What PD, LGD and EAD are, how to compute expected loss, and which parameter a falling collateral value hits hardest.
- Why exposure at default on a partly drawn revolver is more than the drawn amount.
- Point-in-time vs through-the-cycle PD, and how each moves in a recession.
- The three stages of IFRS 9 expected credit loss and how CECL differs.
- What counterparty credit risk and CVA are; an example of wrong-way risk.
- Given a mid-sized company's loan application, which financial and non-financial factors you would look at.
What model risk interviews ask
- What model risk is and how banks manage it.
- The steps you would take to validate a new VaR engine or credit scoring model.
- What worries you most if the business wants to replace a logistic regression scorecard with XGBoost (explainability, overfitting, stability).
- Whether to backtest against actual or hypothetical P&L, and why.
- Statistics and coding fundamentals: hypothesis tests, regression diagnostics, overfitting, and a live Python or SQL exercise.
The evidence your resume should show
Risk recruiters look for three things in each experience: what you analyzed (a portfolio, loans, counterparties, a model), which method you used (VaR, stress testing, credit analysis, backtesting), and what you concluded or changed. Missing any one of the three makes it hard for a screener to count the experience as risk-related.
For example, "assisted the team with risk reports" says almost nothing. "Ran a +200bp rate shock on a 200-bond portfolio in Python, identified two sub-portfolios over their duration limits and escalated them to the portfolio manager" covers the object, the method and the conclusion.
FAQ
- Can I apply for market risk without a risk internship?
- Yes. Build the evidence through coursework or a personal project: compute VaR and run a stress test on a portfolio using public data, then write up the method and conclusion in your resume's projects section.
- Is the FRM worth it?
- It helps for risk roles because it shows you have studied risk methods systematically, but it does not replace concrete evidence from projects and internships. If time is short, get your interview answers and resume solid first.
- Can I move between market risk and credit risk later?
- Yes, especially through hybrid roles like counterparty risk. Interview preparation at entry level differs a lot, though, so focus on one sub-area first.
Free tools
The FindAlly interface is in Chinese; mock interviews are conducted in English.