Recovery Rate vs. Default Rate: The Expected Loss Formula Lenders Actually Use

"Walk me through how you'd estimate expected loss on this loan" is the question recovery rate and default rate are both building toward, and a lot of candidates answer it by defining the two terms separately instead of showing how they combine. Default rate and recovery rate measure two different things entirely. One tells you how often a borrower stops paying. The other tells you how much of your money you get back when that happens. An interviewer pairing them wants to see that you know they multiply together into a single number, not that you can recite two glossary definitions back to back.

That single number is expected loss, and it's the actual metric a lender, a CLO manager, or a private credit fund underwriting a deal cares about. A loan with a high default rate and a high recovery rate can lose less money than a loan with a low default rate and a low recovery rate. Getting that ordering backwards in an interview is the kind of mistake that signals you memorized the vocabulary without understanding what it's for.

Default Rate and Recovery Rate Answer Different Questions

Default rate is a frequency. It's the percentage of loans (or the percentage of par value outstanding) that fail to make a scheduled payment, breach a covenant that triggers default, or file for bankruptcy within a given period, usually measured over a trailing twelve months on a par-weighted basis across a pool of loans.

Recovery rate is a severity measure, and it only applies after a default has already happened. It's the percentage of a loan's face value that a lender actually collects, whether through a distressed sale of the loan itself, cash and new securities received in a bankruptcy reorganization, or proceeds from liquidating collateral. A 100% recovery rate means the lender is made whole despite the default. A 0% recovery rate means the loan is a total loss.

The gap between the two becomes obvious with a simple comparison. A loan with a 5% default rate and an 80% recovery rate is losing less capital on a dollar basis than a loan with a 2% default rate and a 20% recovery rate, even though the second loan defaults far less often. Frequency alone tells you nothing about severity, and severity alone tells you nothing about frequency. Expected loss is what forces the two together.

The Expected Loss Formula That Ties Them Together

The standard credit risk formula is:

Expected Loss = Probability of Default × Loss Given Default × Exposure at Default

Probability of Default (PD) is essentially the default rate applied to a single credit or pool, the likelihood of default over the measurement horizon. Exposure at Default (EAD) is the amount outstanding when default occurs. For a fully funded term loan, EAD is just the current balance. For a revolving credit facility, EAD has to account for the borrower drawing down more of the unused commitment as it heads toward default, which is why revolvers carry higher EAD assumptions than their current drawn balance would suggest.

Loss Given Default (LGD) is where recovery rate comes in directly:

LGD = 1 - Recovery Rate

If a loan recovers 65 cents on the dollar after default, its LGD is 35%. That's the piece of exposure the lender actually loses, and it's the number that gets multiplied against PD and EAD to produce expected loss.

A Worked Example

Take a $50 million first-lien term loan with a 1-year PD of 3% and an expected recovery rate of 60% if it defaults.

LGD = 1 - 60% = 40%

Expected Loss = 3% × 40% × $50 million = $600,000

Now compare that to a second-lien tranche in the same capital structure, sized at $20 million, with a similar 3% PD but a much lower expected recovery rate of 25% given its subordinated claim on collateral.

LGD = 1 - 25% = 75%

Expected Loss = 3% × 75% × $20 million = $450,000

The second-lien piece has a lower expected loss in dollar terms purely because it's a smaller exposure, even though its loss rate as a percentage of face value (2.25%, versus 1.2% for the first lien) is almost double. An interviewer asking you to compare the two tranches wants you to notice that distinction: expected loss in dollars and expected loss as a percentage of exposure can tell different stories, and which one matters depends on whether you're sizing a single position or comparing relative credit quality across tranches.

Why Recovery Rate Matters More Once Defaults Start Climbing

Here's the part most prep material skips entirely: the relationship between default rate and recovery rate isn't static. How much a recovery assumption actually moves your loss number depends heavily on where the default rate sits to begin with.

At a low default rate, say 2%, cutting the recovery assumption from 70% to a stressed 30% only moves the loss rate from about 0.6% to 1.4%. That's an increase, but it's not the headline number in a healthy credit environment. At a 10% default rate, the same recovery cut from 70% to 30% moves the loss rate from roughly 3.0% to 7.0%, more than doubling it. The recovery assumption barely matters when defaults are rare and becomes the dominant driver of loss once defaults become common.

That's exactly why recovery rate gets so much attention from credit investors during a downturn specifically, rather than being treated as a constant input. When the cycle turns and default rates climb, a lender's or a CLO's realized losses are driven far more by how aggressive or conservative their recovery assumptions turn out to be than by small moves in the default rate itself. First-lien leveraged loan recoveries have actually been trending down in recent cycles, with research from JPMorgan putting recent recoveries around 44%, the lowest level since at least 2008 and below the roughly 60 to 70% recoveries that older historical averages would suggest. Interviewers in credit and private credit seats are aware of this trend, and naming it shows you're tracking the market rather than reciting a textbook number.

How Lenders Estimate a Recovery Rate

Recovery rate sounds like a single clean number, but there are two different ways to measure it, and interviewers sometimes probe which one you mean.

Trading-price-based recovery uses the market price of the defaulted loan roughly 30 days after the default event, which is the convention rating agencies like Moody's and S&P use for their historical recovery studies. It's observable and comparable across a large dataset, but it reflects the market's uncertainty about the outcome rather than the outcome itself, since the restructuring hasn't actually happened yet.

Ultimate recovery is the actual value received once the bankruptcy or workout concludes, whether that's cash, new debt, or equity in the reorganized company, valued as of the resolution date. Ultimate recovery is frequently higher than the 30-day trading price because the market tends to overdiscount for uncertainty in the immediate aftermath of a default, and it can take a year or more for a Chapter 11 case to resolve.

Beyond which measurement convention applies, the actual recovery rate a lender expects to realize is driven by a few structural factors that are worth naming directly:

• Seniority and collateral. A first-lien secured loan recovers more than a second-lien loan, which recovers more than unsecured debt, which recovers more than subordinated debt, following the same rank-order logic that governs how the bankruptcy waterfall actually pays out. Whether a lender is underwriting against hard collateral or enterprise value changes the recovery analysis from the start, the same distinction covered in asset-based versus cash flow lending.
• Covenant package. A loan with maintenance covenants gives lenders an early trip wire that can force a restructuring while enterprise value is still largely intact. A covenant-lite loan removes that trip wire, so by the time a default actually occurs, usually triggered by a missed payment or a bankruptcy filing rather than a covenant breach, enterprise value has often eroded further, which is part of why recoveries have drifted lower as cov-lite structures took over the market.
• Liability management transactions. Borrowers under stress increasingly use uptiering and other liability management exercises to prime existing lenders with new, more senior debt before a formal default, which directly reduces what the primed lenders recover if a default eventually does occur.

Common Follow-Up Questions

"Is recovery rate the same thing as the price a distressed loan trades at?" Not necessarily. The 30-day trading price is one specific, commonly cited measure of recovery, but it's a snapshot taken under uncertainty. The recovery rate a lender ultimately realizes, the ultimate recovery, can differ meaningfully from where the loan traded right after default.

"Does expected loss use a one-year PD or a lifetime PD?" It depends on the application. Regulatory capital frameworks like Basel typically use a one-year PD for most exposures. A lender pricing a multi-year term loan or a CLO manager assessing a portfolio over its reinvestment period will often work with a lifetime (or multi-year) PD instead, since a one-year snapshot understates the cumulative chance of default over the life of the loan.

"Why would a lender ever accept a lower margin on a loan with a higher default rate?" Because margin gets priced against expected loss, not against default rate alone. A loan with a higher PD but strong collateral and a correspondingly high expected recovery rate can carry a lower expected loss, and therefore a tighter margin, than a lower-PD loan with weak collateral and a low expected recovery.

"How does this connect to CLO structuring?" A CLO's overcollateralization tests are built around the par value of its loan portfolio, and defaults that come with low recoveries erode that collateral base faster than defaults with high recoveries, which is exactly the stress scenario that can trip a CLO's coverage tests and cut off payments to equity.

Getting Comfortable With Expected Loss

The framework above covers what most interviewers are testing: why default rate and recovery rate measure two separate things, how they combine into expected loss through the PD times LGD times EAD formula, and why recovery rate becomes the dominant driver of losses once default rates climb. Offcycle has this alongside the broader credit and leveraged finance mechanics that come up in the same conversation, including how a loan's pricing gets built from Term SOFR and the credit spread adjustment, built into structured flashcards and quizzes you can work through by topic and difficulty.

Beyond technical flashcards, you get custom practice sets built around whatever you're weakest on, mock interview questions that test the same material out loud, and a readiness score that tracks your progress across credit, LBO, and valuation topics so you know exactly where the gaps are before you're in the room.