Fixed Income Interview Questions: Worked Answers by Desk and Family

A fixed income interview tests working more than definitions. Asked about duration, a strong candidate prices the move. Asked about a distressed company, they find the class of debt where the value runs out. Asked about the Federal Reserve, they give the decision, the date and what the 2-year Treasury yield did that day.

The questions fall into ten families, and each desk leans on a different few: a rates seat on the curve and the Fed, a credit seat on seniority and covenants, a securitized seat on prepayment. What follows takes the families in turn, with the arithmetic worked wherever there is arithmetic, and the process around them, from the first screen to the superday.

The question families, at a glance

Ten families cover almost everything a fixed income interview asks. Every one of them gets a full section below, and the ones with a deeper page of their own are linked from that section.

How a fixed income interview runs

No bank publishes a stage-by-stage map of how it interviews for a markets seat, but the shape candidates describe is consistent: an online application, a recorded or live first screen, and a final round of several back-to-back interviews in one day.

HSBC sets out three steps for students: an application form, a 90 minute online assessment, and a three hour assessment centre. Its own guide adds that the assessment centre is called a super day in the United States, the same stage under a different word. The question families below do not change with the label.

What happens inside those interviews varies less by bank than by two other things: the seat, and which desk is asking.

Sales

• Brain teasers and rapid mental math are rare
• Weight falls on communication and a defensible pitch
• The same market and fit questions still apply

Trading

• Brain teasers and mental math show up often
• Weight falls on speed and quantitative depth under pressure
• The same market and fit questions still apply

Candidates describe brain teasers and mental math as a trading habit rather than a bank habit: a trading candidate at almost any dealer should expect them, and a sales candidate at the same dealer mostly should not.

At four of the big dealers, the desk is settled after the offer, not in the interview. Bank of America, Citi, Morgan Stanley and BNP Paribas all rotate summer analysts across two or more desks during the internship, and Bank of America says placement is determined after the hiring process. Citi asks applicants to name a preferred area of Markets. So "why this desk" tests whether a first choice can be defended, not whether it is final.

The stage-by-stage mechanics of the superday itself, including how a HireVue screen is timed, sit on Sales and Trading Superday.

Knowing what a bank calls its own trading floor matters too, since the same function carries different names across dealers.

The calendar is a season, not a date. For the 2027 summer, BNP Paribas posted its New York internship on 16 December 2025 and Citi posted its London one on 1 September 2026, while Bank of America set its London deadline at 11 October 2026 and HSBC opened its Singapore postings on 3 September 2026 for a 31 October close.

The full calendar, bank by bank and region by region, sits on Sales and Trading Internship.

Why fixed income

Every fixed income interview opens somewhere near this question, and the honest answer needs a specific desk in it. Naming fixed income without narrowing it further signals that the different desks have not really been considered.

• Rates
• Repo
• Credit
• High yield
• Distressed
• CLOs
• Securitized products
• Emerging markets credit

These are separate businesses with separate skills, clients and risk. A strong answer picks one, ties it to a real reason such as a product's complexity or a specific kind of client, and can defend that choice for the rest of the interview.

A closely related question, "which desk within fixed income interests you most," asks the same thing from the other direction: it wants consistency with the rest of the interview, not a new answer. A candidate who spends the conversation on macro views and then names a corporate credit desk reads as one who has not thought it through.

The full version of this question, including how it differs from why sales and trading more broadly, is on Why Fixed Income.

Markets and macro

This family tests whether a candidate actually follows markets, not whether they can recite a definition. Candidates describe the most common version as anchored to the latest Federal Reserve meeting: what the committee decided, what it signaled about the path ahead, and how yields moved once the decision was announced. Even a general market view should carry a specific level, a real yield or a real spread, rather than a mood.

A candidate who connects a data point to a trade, even a simple one such as a curve steepener if cuts look likely or credit protection if spreads look tight, stands out more than one who only recites the headline.

A second version hands a candidate a hypothetical sum of money and asks what they would do with it. The weak answer names an asset class. The strong answer names one trade, one reason for it tied to a real market view, and one way the trade could go wrong.

A worked answer: the December 2025 cut

Take the meeting that ended on 10 December 2025 and answer it in four moves.

The decision. The Federal Open Market Committee cut the target range for the federal funds rate by a quarter point, to 3-1/2 to 3-3/4 percent. The vote was 9-3, with dissents on both sides: one member wanted a half-point cut, two wanted no change. The statement also said the Fed would begin buying shorter-term Treasury securities to keep reserves ample.

What was priced. The minutes record that market participants generally expected a 25 basis point cut, so the cut itself was not the news.

The move. Treasury par yields fell along the curve, most at the 2-year:

The reason. With the cut priced, the answer has to find the news elsewhere. A strong candidate names what else changed, the split vote and the new purchases, and says how they would test which one moved the front end. A fuller answer on a later meeting, maturity by maturity, is on Sales and Trading Technical Questions.

Bond math

Every fixed income seat tests bond math at some level, and duration and convexity are the two ideas that come up most. Duration estimates how much a bond's price moves for a given change in yield. Convexity corrects that estimate for how much the price and yield relationship actually curves.

Take a bond with a modified duration of about 9.15 years and a convexity of about 115. Duration alone predicts that a 100 basis point rise in yield cuts the bond's price by roughly 9.15 percent. Convexity adds back a little more than half a percentage point, because the price and yield relationship curves in the bondholder's favor, so the fuller estimate lands closer to 8.6 percent.

Two more questions from the same family, each short enough to work aloud:

• "Price a two-year bond with a 5% annual coupon when yields are 6%." Discount each cash flow: 5 divided by 1.06 is 4.72, and 105 divided by 1.06 squared is 93.45, so the price is about 98.17. A coupon below the yield means a price below par; at a 5% yield the same bond prices at exactly 100.
• "What is the DV01 on $10 million of that 9.15-duration bond, priced at 98.85?" Modified duration times price times face times one basis point: 9.1527 × 0.98846 × $10 million × 0.0001 is about $9,047 for each basis point the yield moves.

The full derivation, plus the mental-math questions that come packaged with it, is on Bond Math Interview Questions. The bond calculator works the same numbers for any coupon, yield and maturity.

Rates

A benchmark Treasury is the most recently auctioned issue at a given maturity, called on the run. Every earlier issue at that maturity is off the run, and trading concentrates in the newest issue: on the electronic interdealer platforms, Treasury coupon trading happens in the on-the-run securities.

A classic technical question gives two spot rates and asks for the rate implied between them. Say the one-year Treasury yields 1 percent and the two-year yields 1.5 percent. Compound the two-year rate, 1.015 squared is 1.030225, and divide by the one-year, 1.01: the implied one-year rate, one year forward, is about 2 percent. That is the rate the curve implies for a year from now, extracted from prices rather than guessed at.

How a rates desk trades the curve day to day is on Rates Trading.

Repo

A repurchase agreement is, in the New York Fed's own description, economically similar to a loan collateralized by securities. One side sells a security and agrees to buy it back at a set price on a set date; the other side is doing the reverse.

Candidates describe the desk's job as funding the bank's own inventory and its clients' positions, with its profit coming more from allocating that inventory well across many trades than from directional risk.

A common question asks why repo rates spike around quarter end and year end, even when nothing has changed in the underlying market. The answer is balance sheet, not credit.

Dealers whose parent reports its leverage from the quarter-end date itself pull back from tri-party repo into that date, around $170 billion of it on average. US regulators measure the ratio from a daily average instead, so the pull-back sits with non-US parents, and the price of that scarce capacity rises for a few days before falling back once the date passes.

A candidate who can explain a price move through a regulatory calendar rather than a market event is showing exactly the kind of structural knowledge this desk wants. The full mechanics of general collateral, specials, tri-party settlement and the desk itself sit on Repo Trading.

Credit, high yield and distressed

Investment grade and high yield are separated by one thing: the credit rating. A bond rated BB+ or lower at S&P, or Ba1 or lower at Moody's, is high yield, and distressed is a further, informal step down.

The rating line also shows in how the bonds trade. In 2024, 48% of US investment grade corporate bond volume traded electronically, against 33% of high yield.

US investment grade

48%

US high yield

33%

Investment grade drove the year's rise in electronic trading, while the high yield share was flat on the year. Source: Coalition Greenwich data, reported by The Desk, January 2025.

A strong answer to "how do the two differ" uses both facts: the rating line, and a high yield market that still did two thirds of its volume away from the screen.

A cov-lite loan carries incurrence covenants, which only restrict set actions such as issuing more debt, rather than maintenance covenants, which the borrower must meet every quarter on pain of handing control to its lenders. That difference is why cov-lite structures give borrowers more room and give lenders less warning before something goes wrong.

Interviewers also test whether a candidate keeps two kinds of default straight. A technical default is a covenant violation, a breach of a rule in the loan agreement, and can happen even when every payment has been made on time. A payment default is a missed interest or principal payment, the more serious of the two.

Conflating the two is a common, and noticeable, mistake. The desk-by-desk depth behind flow credit, high yield and distressed trading sits on Credit Trading, High Yield Trading and Distressed Debt Trading, and the questions a dedicated credit research seat is asked sit on Credit Analysis Interview Questions.

The classic distressed question gives a capital structure and asks where the value sits. Take a company worth $850 million, with $700 million of senior bank debt and $300 million of bonds beneath it. Pay the claims from the top down: the bank debt takes the first $700 million and recovers 100 cents on the dollar, the $150 million left covers half the bonds, 50 cents, and the equity gets nothing.

Repaid from the top down. Each band is sized by its share of the structure.

Bank debt70% · senior, recovers 100 cents

Bonds← the fulcrum security30% · recovers 50 cents

Repaid from the top down: $850 million covers the $700 million of bank debt in full and $150 million of the $300 million of bonds. The equity beneath them gets nothing.

That makes the bonds the fulcrum security: the most senior class not covered in full, and in a restructuring often the class best placed to take the new equity. The number that matters from the bond chair is $700 million. Below it the bonds recover nothing, and every dollar of value between $700 million and $1 billion goes to them.

CLO and securitized

A CLO pools leveraged loans, mostly senior secured loans to below investment grade companies, and slices the cash flows into tranches by seniority. The top tranches are engineered for an investment grade rating even though the underlying loans are not, because losses are absorbed from the bottom up: the equity tranche takes the first loss and the AAA tranche the last.

Scale is part of a good answer. CLOs were a $1.4 trillion asset class inside a $13.3 trillion structured credit market at the end of 2024, and they bought 61% of that year's new-issue leveraged loans.

$1.4tn

CLO market size

in $13.3tn of structured credit, 31 Dec 2024

61%

of 2024 new-issue leveraged loans bought by CLOs

Bank of America data, 30 Apr 2025

64%

of the leveraged loan market owned by CLOs

Bank of America data, 30 Apr 2025

A related question asks a candidate to separate a CLO from a CDO, since the two are often used loosely as if they were the same thing. A CLO is backed specifically by a pool of leveraged loans to corporate borrowers.

A CDO is the broader category, and can be backed by a wider mix of debt instruments, including bonds and, in the structures that mattered most in 2008, tranches of subprime mortgage-backed securities. Getting this distinction right, unprompted, signals that a candidate understands the product rather than the acronym. The full tranche mechanics, warehousing and manager tiering sit on CLO Trading.

Mortgage-backed securities carry a different wrinkle: negative convexity. When rates fall, homeowners refinance, the mortgage pool prepays faster, and the bond's effective life shortens right when a bondholder would rather it lengthen. That is the opposite of the convexity a plain government or corporate bond carries, and a securitized interviewer will want it explained correctly. The deeper mechanics of prepayment and the agency market are on Securitized Products.

Emerging markets

Candidates describe an emerging markets credit desk as covering more than one product for a given country: hard-currency sovereign bonds, sometimes local-currency debt, and often sovereign credit default swaps on the same names. A candidate should be able to name at least one country-specific driver of a bond's price: a commodity export dependency, an upcoming election, an IMF program's status.

That beats treating emerging markets as one undifferentiated block. For hard-currency sovereign debt, the index to know by name is J.P. Morgan's EMBI Global Diversified, which tracks liquid dollar bonds issued by sovereign and quasi-sovereign borrowers.

A second question worth preparing for asks why a candidate would want a hard-currency bond over a local-currency one, or the reverse. A hard-currency bond, usually issued in dollars, removes currency risk from the trade but leaves the issuing country's ability to earn or borrow dollars as the real question.

A local-currency bond adds currency risk on top of credit risk, which is a different, and often larger, source of return and loss. Naming that trade-off directly is worth more than naming either instrument alone. The full desk breakdown for emerging markets credit sits on Emerging Markets Credit.

Brain teasers

Candidates describe brain teasers and rapid mental math as far more common on the trading side than the sales side. The point is rarely the final number. It is whether a candidate thinks out loud, breaks a hard calculation into easy pieces, and stays composed when the first approach does not work.

• Decompose, do not memorize: 47 times 5 becomes forty times five plus seven times five, or two hundred plus thirty five, reached in two easy steps instead of one hard one.
• State an assumption out loud before using it, rather than silently picking one.
• If a probability question sounds like a trick, check for independence before computing anything.
• When a game has a decision in it, find the threshold first: keep anything worth more than what you would give up to change it.

A classic version: roll a die and be paid its face value in dollars. If you dislike the first roll you may roll again, but then you must keep the second. What would you pay to play? A fresh roll is worth 3.50 on average, so keep a 4, 5 or 6 and reroll a 1, 2 or 3.

Half the time you keep a roll that averages 5, and half the time you take a fresh roll worth 3.50, so the game is worth 0.5 × 5 plus 0.5 × 3.50, or $4.25. Checking all 36 pairs of rolls under that rule gives 4.25 again, and moving the threshold either way lowers the value to about 4.17. Pay anything up to $4.25.

The number matters less than the threshold, and the threshold is just what a reroll is worth. The fuller set of market-view and brain teaser questions sits on Sales and Trading Technical Questions.

Behavioural

Fit questions in a fixed income interview reward the same story structure investment banking does, situation, task, action, result, but score it against different traits: risk judgment, composure under pressure, and a genuine reason for wanting this specific seat rather than a seat in general.

The most useful preparation before any fit question is not a rehearsed script. It is knowing the Treasury curve, the most recent Fed decision, and one live commodity or currency move well enough to fold into an answer without being asked directly for it.

A handful of behavioural questions recur often enough to prepare word for word:

• "Are you more risk averse or more risk seeking, and why does that suit this desk." The honest answer for most fixed income seats leans toward risk aware rather than either extreme: fixed income investors are mostly paid for identifying and pricing risk precisely, not for chasing the largest possible payoff.
• "Tell me about a time you were wrong and had to change your mind quickly." A desk wants to see the update happen fast and without ego, since a trader who defends a wrong position too long is the expensive kind of wrong.
• "What is the greatest risk you have taken, and how did you size it." The story matters less than the sizing logic: a candidate who describes deciding how much to risk, not just what to risk, is closer to how the seat actually works.

Mistakes interviewers name

A narrow set of habits costs candidates points before the technical questions even get hard.

• Naming fixed income without a specific desk, when asked why this seat.
• Reciting a memorized definition of duration without being able to compute anything from it.
• Having no market view beyond a headline, when asked what is happening in rates or credit.
• Treating a brain teaser as a test to pass rather than a chance to show how you think.
• Arriving at a superday able to discuss the bank but not the specific desk.

The four week preparation plan

The bottom line

A fixed income interview does not reward the candidate who has memorized the most definitions. It rewards the one who can walk into any of these ten families, from why fixed income to a distressed capital structure, and show the work: which desk, which number, which reason. That habit is built before the superday, not during it.

The interview readiness tool turns this page into a short diagnostic if a faster read on where the gaps sit is more useful than working through every family in order.