Debt Capital Markets interview preparation

Questions

100

Traced to a firm

45

Firms

26

Updated

September 2026

Asked at

All firmsTSTruist Securities5PIMCO4TD Securities4Apollo Global Management3Nomura3Scotiabank3Bain Capital2Houlihan Lokey2Mizuho2Neuberger Berman2Oaktree Capital Management2RCRBC Capital Markets2Carlyle Group1Deutsche Bank1Golub Capital1HPS Investment Partners1Invesco1KKR1Lazard1Moelis & Company1Moody's1Northern Trust1NUNuveen1Rothschild & Co1S&P Global1Wells Fargo Securities1

Topic

All topicsBond mechanics11Duration and convexity6Yield curve and rates5Credit spreads5Credit analysis and ratings13Credit modelling9Primary issuance9Syndication and loans10Structured credit7Covenants and documentation5Liability management5Indian debt markets7Fit8

Level

AnyCoreIntermediateHard

Type

AnyTechnicalBrainteaserMarket viewCaseFit

1. 012What is duration, explained the way you would explain it to a client?Duration and convexityCorephone / first roundPIMCODebt Capital Markets · San Diego · 2026

   Say this

   Duration is how much your bond's price moves for a one percent change in yields. A duration of 7 means roughly a 7 percent price fall if yields rise 100 basis points. Underneath, it is the weighted average time until you get your money back.

   Then walk it

   1. Two readings of the same number. Macaulay duration is a time: the present-value-weighted average years to the cash flows, quoted in years. Modified duration is a sensitivity: the percentage price change per 100 basis points.
   2. Modified equals Macaulay divided by one plus the periodic yield, so for normal yields they are close, and people use the words loosely. Say which one you mean.
   3. What makes duration long: long maturity, low coupon, low yield. All three push more of the present value further into the future.
   4. The client version: duration is your interest rate risk budget. A fund with duration 2 loses about 2 percent if rates rise 100 basis points. A fund with duration 15 loses about 15. Same credit, totally different instrument.
   5. It is a first-order approximation, valid for small moves. For a 200 basis point move you need convexity, which corrects the fact that the price-yield curve bends.
   6. And the framing that matters on a debt desk: duration is what a rates trader hedges and what a credit investor tries to neutralise so that what is left is the credit view.

   Where candidates lose it

   Conflating Macaulay and modified duration, or reciting 'weighted average time to cash flows' without ever saying what it is used for. A client and an interviewer both want the sensitivity first, then the definition.

   Expect next

   • So how does duration affect what happens when rates move?
   • What is DV01 and how is it different?
   • How would you reduce the duration of a portfolio?

   Reported by candidates at PIMCO (Debt Capital Markets, San Diego, 2026). Source: Wall Street Oasis.

2. 013What is DV01, and how would you actually use it on a desk?Duration and convexityIntermediatetechnicalSyndicate desksFixed income asset management

   Say this

   DV01 is the dollar change in the value of a position for a one basis point move in yield. It converts a percentage sensitivity into money, which is what you need to size a hedge.

   Then walk it

   1. The arithmetic: DV01 is roughly modified duration times market value times 0.0001. On 100 million of a bond with duration 7, that is about 70,000 dollars per basis point.
   2. Why money rather than percent: you cannot hedge a 7 percent sensitivity, you hedge 70,000 dollars a basis point. So you sell enough Treasury futures or pay enough on a swap to produce minus 70,000 a basis point.
   3. The hedge ratio is just the ratio of the two DV01s. If the cheapest-to-deliver 10-year note has a DV01 of 780 dollars per contract per basis point, you need about 90 contracts.
   4. It is also how risk limits are written. A syndicate or trading desk carrying a new issue overnight has a DV01 limit, not a duration limit, because the risk manager cares about dollars at stake.
   5. Related measures on the same logic: CS01 or spread DV01 for a one basis point move in credit spread, and that is the number a credit desk watches, because their rate risk is hedged out.
   6. The limitation: DV01 is linear and local. For large moves convexity matters, and for a callable bond DV01 itself changes as rates move, so you re-hedge rather than set and forget.

   Where candidates lose it

   Defining DV01 and stopping. The question is 'how would you use it', so give the hedge ratio. If you cannot say that the hedge is the ratio of DV01s, the answer reads as textbook.

   Expect next

   • Work out the DV01 on 250 million of a 5-duration bond.
   • What is CS01?
   • How would you hedge the rate risk on a new issue you are holding overnight?
3. 014What is convexity, and why do investors pay for it?Duration and convexityHardsuperdayFixed income asset management

   Say this

   Convexity is the curvature of the price-yield relationship — the rate at which duration itself changes as yields move. Positive convexity means you gain more when rates fall than you lose when they rise by the same amount, so it is a free asymmetry and investors pay for it in yield.

   Then walk it

   1. Duration is the first derivative, convexity the second. The price change is minus duration times the yield move, plus a half times convexity times the move squared. The squared term is always positive for a bullet bond, whichever way rates go.
   2. Concretely: a bond with duration 10 and convexity 100. Rates fall 100 basis points, you make 10 plus 0.5 percent, so 10.5. Rates rise 100, you lose 10 minus 0.5, so 9.5. That one point of asymmetry is the convexity.
   3. It matters more the bigger the move and the longer the bond. For a 30-year at low yields, convexity can be several points over a 200 basis point move, and ignoring it is a serious mispricing, not a rounding error.
   4. Because it is an asset, it is priced. Two bonds with the same duration but different convexity will not have the same yield: the more convex one yields less. Barbell versus bullet portfolio construction is exactly this trade.
   5. Negative convexity is where the money is lost. Callable bonds and mortgage-backed securities have it: as rates fall, the borrower refinances, so your upside is truncated. You are short an option and the yield is your premium.
   6. The practical honesty: a long-convexity position bleeds carry. You are paying up in yield every day for protection against a big move, and if the move never comes you underperform.

   Where candidates lose it

   Describing convexity as 'the second derivative' and leaving it there. Interviewers want the asymmetry stated in numbers, and they want to hear negative convexity named for callables and MBS, because that is where convexity actually costs people money.

   Expect next

   • Which has more convexity, a barbell or a bullet?
   • Why do mortgage-backed securities have negative convexity?
   • Who pays for convexity and who sells it?
4. 017What is spread duration, and how is it different from interest rate duration?Duration and convexityHardsuperdayCredit researchFixed income asset management

   Say this

   Spread duration measures the price change for a one percent move in the credit spread; rate duration measures it for a move in the underlying risk-free yield. For a fixed-rate bullet they are almost identical, but for a floater they are completely different.

   Then walk it

   1. The reason they usually match: price is discounted at the risk-free rate plus the spread, and a 100 basis point move in either component moves the discount rate the same amount. So for a fixed bullet, spread duration is effectively modified duration.
   2. Where they diverge is a floating rate note. Rate duration is about 0.25 years, because the coupon resets quarterly. Spread duration is the full maturity, because the fixed spread over the reference rate does not reset.
   3. That is exactly why leveraged loans and floaters are the instrument of choice for someone who wants credit risk and no duration.
   4. For a credit portfolio it is the risk number that matters. A manager running duration-hedged credit measures spread duration times notional, sometimes called DTS or duration times spread, because spread volatility scales with the level of spread.
   5. The DTS insight worth quoting: a 5-year bond at 500 basis points has roughly the same spread risk as a 10-year bond at 250. Wide credits behave like long duration credits, and a book that ignores this is badly mis-sized.
   6. Limitation: spread duration assumes a parallel shift in the spread curve, and in a credit selloff short-dated distressed paper often widens far more than long-dated, so the linear measure underestimates the tail.

   Where candidates lose it

   Saying they are the same thing. They are the same number for a fixed bullet, and radically different for a floater, a loan or a CLO note. Naming the floater case is what proves you understand why the distinction exists.

   Expect next

   • What is the spread duration of a 7-year leveraged loan?
   • What is duration times spread?
   • How would you hedge spread risk?
5. 018What drives the shape of the yield curve?Yield curve and ratesIntermediatetechnicalFixed income asset management

   Say this

   Three things stacked on top of each other: where the market thinks policy rates are going, a term premium for holding duration, and supply and demand at specific maturities. The front end is almost entirely central bank expectations; the long end is mostly term premium and flows.

   Then walk it

   1. Expectations first. The 2-year is roughly the average expected policy rate over two years, so if the market prices cuts, the front end falls and the curve steepens from the front.
   2. Term premium second. Lending for 30 years carries inflation and policy uncertainty you cannot diversify, so investors demand extra yield. That premium expands when inflation is volatile and compresses when it is boring.
   3. Supply and demand third, and it is bigger than textbooks suggest. Pension and insurance demand anchors the long end; heavy government issuance at a particular tenor cheapens it. Quantitative easing suppressed term premium directly by taking duration out of the market.
   4. Put it together for the standard shapes. Upward sloping is the normal state: rates expected stable and a positive term premium. Inverted means the market expects cuts, which usually means it expects a slowdown. Humped usually means near-term hikes followed by cuts.
   5. A real example: the US curve inverted through 2023 with 2s10s at about minus 100 basis points at the extreme, then steepened back as cuts got priced. Same curve, two completely different messages about the cycle.
   6. The honest caveat: you cannot separate expectations from term premium observably. Models like ACM decompose them, and they disagree. So be careful about claiming the curve is 'predicting' anything specific.

   Where candidates lose it

   Giving only the expectations story. If the curve were pure expectations, the term premium would be zero and 30-year bonds would be as safe as bills. Naming term premium and supply-demand is what makes the answer sound like a rates desk rather than a textbook.

   Expect next

   • What is the term premium and can you observe it?
   • Why did QE flatten the curve?
   • What does an inverted curve tell you?
6. 019What does an inverted yield curve tell you, and what does it not tell you?Yield curve and ratesIntermediatetechnicalCredit researchFixed income asset management

   Say this

   It tells you the market expects policy rates to be lower in future than they are now, which usually means it expects growth to weaken. It does not tell you when, it does not tell you by how much, and it is not itself a cause of anything.

   Then walk it

   1. Mechanically, the long rate is an average of expected short rates plus a term premium. For the long rate to sit below the short rate, the market must be pricing meaningful cuts.
   2. The historical record is genuinely strong: 2s10s inversion has preceded every US recession since the 1960s. But the lag has ranged from about 6 to 24 months, which makes it useless as a timing tool.
   3. It has also produced false signals, and the 2022 to 2024 inversion is the live example — the deepest inversion in forty years without the recession arriving on schedule. Anyone who positioned purely on the signal lost money for two years.
   4. What it does to a DCM desk is concrete and immediate. Inversion means short funding costs more than long funding, so issuers term out debt and the long end of the new issue calendar gets busy. It also crushes bank net interest margins, because banks borrow short and lend long.
   5. For credit specifically, inversion plus tight spreads is the uncomfortable combination: the rates market is pricing a slowdown and the credit market is not. That divergence is worth flagging in an interview because it is a real analytical tension.
   6. The limitation to state: it is a market expectation, not a forecast with a track record of calibration. And the curve can un-invert either because growth recovers or because the front end collapses in a crisis — same shape change, opposite story.

   Where candidates lose it

   Saying 'an inverted curve predicts a recession' as a flat fact. The 2022 to 2024 episode is the obvious counter and an interviewer will produce it. Give the mechanism, then the record, then the false-signal caveat, then what it means for issuance.

   Expect next

   • So why did the 2022 inversion not produce a recession on schedule?
   • What does inversion do to bank margins?
   • Which part of the curve do you watch, 2s10s or 3m10y?
7. 023What does a credit spread actually compensate an investor for?Credit spreadsIntermediatetechnicalCredit researchFixed income asset management

   Say this

   Four things, and only the first is default. Expected credit loss, the risk that losses come out worse than expected, illiquidity, and a residual that is really just supply and demand for credit risk. Expected loss is usually the smallest component.

   Then walk it

   1. Expected loss is probability of default times loss given default. For a BBB name at roughly 0.2 percent annual default probability and 60 percent loss severity, that is about 12 basis points a year.
   2. But BBB spreads have historically averaged around 150 basis points. So the vast majority of the spread is not compensation for expected default at all.
   3. The rest is: a risk premium for the fact that defaults cluster in recessions exactly when your other assets are falling, so credit losses are systematic and undiversifiable; a liquidity premium, because corporate bonds trade by appointment and bid-offer widens when you most want to sell; and a plain supply-demand residual.
   4. The credit spread puzzle is the academic name for this gap, and it is worth naming. It is why a diversified investment grade portfolio has historically earned a real excess return over Treasuries despite realised losses being tiny.
   5. For high yield the balance shifts. A single-B name at 4 percent annual default probability and 60 percent severity carries about 240 basis points of expected loss against maybe 450 of spread, so default compensation is now a majority of the spread.
   6. The honest caveat: spreads also contain a technical component that has nothing to do with credit. Index inclusion, central bank purchases, and fund flows can move investment grade spreads 50 basis points with no change in fundamentals.

   Where candidates lose it

   Answering 'default risk' and stopping. The interesting fact is that expected default loss explains only a small fraction of investment grade spread, and being able to put numbers on that gap is what makes you sound like a credit analyst rather than a student.

   Expect next

   • So what is the credit spread puzzle?
   • How would you estimate loss given default?
   • Why do spreads widen before defaults rise?
8. 024Walk me through the different spread measures — I-spread, asset swap spread, Z-spread and OAS. When would you use each?Credit spreadsHardtechnicalSyndicate desksCredit research

   Say this

   They are four increasingly careful ways of asking the same question: how much am I being paid over risk-free for this credit. I-spread is the crude one over an interpolated swap rate; Z-spread discounts every cash flow off the whole curve; OAS strips out the value of any embedded option.

   Then walk it

   1. I-spread, or interpolated spread: the bond's yield minus the swap rate at the same maturity. Fast and fine for quoting, but it compares a single yield number against a single point on the curve, so it ignores curve shape.
   2. Asset swap spread: the spread you earn if you buy the bond and swap the fixed coupons into floating. It is what a funded investor actually realises, and it is the number a syndicate desk uses to compare a new euro issue against outstanding paper.
   3. Z-spread, or zero-volatility spread: the constant spread you add to every point on the zero curve so the discounted cash flows equal the market price. This is the correct measure for a bullet, because it respects the shape of the curve rather than one point on it.
   4. OAS, option-adjusted spread: Z-spread minus the value of the embedded option, computed over simulated rate paths. For a callable high yield bond or an MBS, Z-spread overstates the credit compensation because part of it is the option premium you sold.
   5. So the rule: bullets, use Z-spread. Callables and mortgages, use OAS. Cross-currency or floating comparisons, use asset swap. Quick quotes, I-spread.
   6. One number to show the difference matters: on a callable high yield bond, Z-spread minus OAS can easily be 80 to 150 basis points. Ranking two bonds on Z-spread when one is callable will give you the wrong answer.

   Where candidates lose it

   Using Z-spread and OAS interchangeably. They are identical only when there is no embedded option. Saying 'OAS equals Z-spread for a bullet' unprompted is the single cleanest way to show you understand the family.

   Expect next

   • When are Z-spread and OAS the same?
   • How is OAS actually computed?
   • Which measure would you use to price a new euro issue?
9. 027Can debt ever be more expensive than equity, and in what scenario?Credit spreadsHardsuperdayTD SecuritiesCapital Markets · New York · 2025

   Say this

   Yes, and it happens regularly in distress. When a borrower is close to default, rescue debt can price at 15 to 20 percent cash plus fees, warrants and PIK, while the equity stub is an out-of-the-money option that costs the sponsor nothing in cash. At that point issuing equity is cheaper than borrowing.

   Then walk it

   1. The normal ordering holds for two reasons: debt is senior so it takes less risk, and interest is tax-deductible so the after-tax cost is lower. Both can break.
   2. The tax shield disappears first. A company with no taxable income gets no deduction, so the after-tax cost of debt equals the pre-tax cost. Interest limitation rules, like the 30 percent of EBITDA cap in the US, do the same thing at high leverage.
   3. Then the risk ordering flips in effect. In distress, existing bonds trade at 20 or 30 percent yields and new money demands super-priority, a 15 percent coupon, 3 points of fees and warrants. All-in that can be 25 percent.
   4. Meanwhile the equity has convexity. A sponsor putting fresh equity into a levered business is buying a long-dated option on recovery, and the required return on that, in the sponsor's own hands, can be lower than the rescue lender's price.
   5. Concrete cases: 2020 pandemic rescue financings and several 2023 to 2024 liability management exercises priced new money at levels no equity investor would have demanded. Sponsors chose to write equity cheques instead.
   6. There is also a mundane version: a small private company with no ratings and no assets may simply not be able to borrow at any price, which is an infinite cost of debt. Availability, not just price, is part of the cost of capital.

   Where candidates lose it

   Answering flatly no because debt is senior and tax-deductible. That is the textbook line and the question exists to test whether you can break it. Name the tax shield disappearing and the distress case, and give an all-in rescue-financing number.

   Expect next

   • What happens to the tax shield if there is no taxable income?
   • Why would a sponsor write an equity cheque instead?
   • How would you price rescue financing?

   Reported by candidates at TD Securities (Capital Markets, New York, 2025). Source: Wall Street Oasis.

10. 028Walk me through the ratings process for a debut issuer.Credit analysis and ratingsIntermediatetechnicalRating agenciesCorporate banking

    Say this

    Six to ten weeks, and it is a managed process rather than a submission. You prepare a rating presentation and a model, hold a management meeting with the analytical team, they take it to a committee, you get an indicative or final rating with an outlook, and the issuer has one appeal before it is published.

    Then walk it

    1. Step one is choosing agencies. Most benchmark issuers take two, because index eligibility and many investor mandates require two ratings. A third is bought when you expect it to be better or when you need a specific investor base.
    2. Then the rating presentation, which the DCM team drafts with the company: business description, market position, financial policy, a five-year model, and crucially a stated leverage target. The financial policy commitment is often what decides the notch.
    3. The management meeting is the substance. The agency's lead analyst and team question the CFO and treasurer for several hours. Rehearsing the CFO for this is real ratings advisory work, not window dressing.
    4. Committee follows, and the analyst is one vote. Then the agency issues an indicative rating confidentially, which lets the issuer decide whether to proceed or fix something first — refine the structure, add security, change the maturity profile.
    5. Appeal is one shot and needs new information, not a better argument. Then the rating publishes with an outlook and the issuer enters ongoing surveillance, with an annual review and rating triggers.
    6. The practical number: a notch on the boundary is worth real money. Crossing from BBB minus to BB plus can widen the coupon by 100 to 200 basis points and shrink your investor base sharply, which is why ratings advisory happens before the deal, not during it.

    Where candidates lose it

    Describing it as a passive assessment the company receives. It is a negotiation with an evidence standard, and the commercially important insight is the indicative rating step, which lets the issuer restructure before anything is published.

    Expect next

    • Why would an issuer take two ratings rather than one?
    • What is ratings advisory?
    • What happens if the indicative rating comes back a notch below target?

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.