Risk Management interview preparation
Market, credit and operational risk, plus model validation, regulatory capital, liquidity and ALM, the statistical foundations and the Indian regulatory syllabus. Every question is either traced to a named firm from a public candidate report, or tagged at desk level when we could not trace it — and answers lead with the point, then the mechanism, then the limitation.
100 questions, mapped to the firms that asked them
- Questions
- 100
- Traced to a firm
- 37
- Firms
- 12
- Updated
- September 2026
080What is the broad range of risks a bank faces, and what is the greatest one?UBSPrivate Wealth Management · New York · 2026
Say this
Credit, market, liquidity and operational are the four you capitalise, then interest rate risk in the banking book, conduct, model, strategic and reputational risk on top. The greatest is liquidity, because it's the one that kills a bank in days rather than years.
Then walk it
- Credit risk is the largest by capital, usually 80 to 90 percent of a commercial bank's RWA, and it's the one that does most of the slow damage. Almost every banking crisis starts as a credit cycle.
- Market risk is small for most commercial banks and large for a trading house. Operational risk includes conduct, which has produced some of the biggest single losses in banking history.
- Liquidity risk is the answer to 'greatest', and the argument is about speed and irreversibility. A capital problem gives you quarters to raise equity or shrink. A funding problem gives you a day. Northern Rock, Lehman, Credit Suisse and SVB were all liquidity events at the end, whatever started them.
- The nuance that makes it a better answer: the cause is usually credit or rate risk and the mechanism of death is liquidity. So the greatest risk is the interaction, not any one silo. Solvency doubts cause funding to disappear, and forced sales turn doubts into insolvency.
- For a specific bank the answer changes and you should say so. For an Indian public sector bank it's concentrated corporate credit. For an NBFC it's asset-liability mismatch and wholesale funding dependence. For a custodian it's operational and technology risk. For a private bank it's conduct and suitability.
- The risk I'd flag as most underweighted relative to its importance: third-party and technology concentration. A handful of cloud providers, core banking vendors and payment rails now underpin the system, and that exposure sits in nobody's capital calculation.
- So my framing: capital protects you from credit and market losses, and only liquidity and governance protect you from the failure mode that actually happens.
Where candidates lose it
Listing risk types with no view. The question explicitly asks which is greatest, so refusing to pick is a fail. Pick liquidity, justify it on speed, then show sophistication by saying the cause is usually credit and the mechanism is liquidity, and that the answer depends on the institution.
Expect next
- Why liquidity and not credit?
- What's the greatest risk for a private wealth business specifically?
- Which risk do you think is most underpriced today?
Reported by candidates at UBS (Private Wealth Management, New York, 2026). Source: Wall Street Oasis.
081What is a risk appetite framework, and what makes one actually work?Regulatory reporting
Say this
It's the board's statement of how much risk the firm will take to pursue its strategy, translated into measurable limits that cascade down to desks. It works when it constrains a real decision. If no business has ever been turned down because of it, it's a document, not a framework.
Then walk it
- The structure has three layers. Risk capacity, the maximum the firm could absorb before failing. Risk appetite, what the board chooses to take, deliberately inside capacity. Risk tolerance and limits, the operational thresholds that keep you inside appetite.
- It has to be expressed in metrics with numbers, not adjectives. 'We have a conservative appetite for credit risk' is meaningless. 'Maximum 15 percent of the loan book in any one sector, CET1 not below 12 percent in the adverse stress scenario, maximum 3 percent credit cost' is a framework.
- Cover the unquantifiable too, with zero-tolerance statements where that's honest: no facilitation of tax evasion, no business in sanctioned jurisdictions, no products sold to retail without a suitability assessment. Conduct appetite can't be a number, but it can be a boundary.
- Cascade is the hard part. The board sets firmwide appetite, which has to become desk limits, sector caps, product approvals and even individual mandates. The arithmetic of cascading rarely works cleanly because of diversification, and that allocation is a genuine judgement.
- Then the feedback loop: monitoring, escalation when you approach a threshold, and a defined process for breaching one. Amber triggers a conversation, red triggers a mandated action. If breaching a limit has no consequence, the limit isn't real.
- The test I'd apply, and I'd say this out loud: name a transaction the firm declined in the last year because of the risk appetite framework. If nobody can, it's decorative. That question is also the best one to ask a firm in an interview.
- The other failure mode is appetite drift. Limits get raised to accommodate whatever the business is already doing, so the framework ratifies behaviour instead of constraining it. The control against that is that increases go to the board risk committee with a documented rationale, not to a delegated authority.
Where candidates lose it
Describing capacity, appetite and tolerance as a hierarchy and never addressing whether it changes behaviour. The differentiator is the test: what did the firm say no to. And naming appetite drift, limits quietly rising to fit the business, shows you've seen how these decay in practice.
Expect next
- How would you cascade a firmwide appetite down to a desk?
- How do you express appetite for conduct risk?
- What do you do when the business wants a limit raised?
Firm tags come from public, anonymous candidate reports on Wall Street Oasis: strong signal, not sworn testimony. Firms are named as the places a question was reported, not as partners of Fin Maverick. Answers are written for this page to show how to think out loud; they are not scripts to recite.

