Ultimate Guide to Distressed Debt Investing

When Frontier Communications reorganized in 2021, its senior unsecured bondholders received 100 percent of the new stock, and the company shed about $11 billion of debt. The people who bought those bonds cheaply ended up owning the company. Distressed debt investing is buying the loans and bonds of companies in financial trouble at a discount, either to collect more than you paid when the debt is repaid or restructured, or to end up owning the company when that debt is converted into equity. For many of these investors, the bankruptcy is where the trade pays off.

I went to Harvard and worked at McKinsey, and I've run a lot of mock technical interviews with students headed into banking. By the end of this guide you'll be able to walk a recovery waterfall and find the class of debt that ends up owning the company, explain the two main trades and why the valuation fight decides them, tell the two kinds of distressed funds apart, and answer the questions a distressed interview asks.

What Is Distressed Debt?

Distressed debt is debt that trades far below what the borrower promised to pay, because the market doubts it will get paid. In practice that means a bond or loan trading somewhere below 70 or 80 cents on the dollar, or yielding 10 percentage points or more above Treasuries. The company behind it might be missing payments already, or heading toward a bankruptcy filing, or in the middle of one. That's why it's priced on what the lender will eventually recover, whatever the company originally borrowed.

How Do Distressed Debt Investors Make Money?

There are two basic trades. In the first, the investor buys debt at a deep discount and gets paid back more than they paid. That can happen because the company recovers, because it refinances, or because it goes through bankruptcy and pays that class of debt in full or in part. In the second, called loan-to-own, the investor buys the class of debt that will be converted into ownership of the company when it reorganizes, then makes money as the owner. The price you pay is set by the market's fear, and the money you make is set by what the bankruptcy process eventually hands your class of debt.

1. You're Buying a Claim, Not a Security

The first shift in thinking is this. A bond or a loan looks like a security, something with a price on a screen. When the company is healthy, that's a fine way to think about it. When the company is in trouble, the loan becomes something else: a legal claim on the company's assets and cash, with a specific place in line.

That place in line is set by two things. The first is whether the claim is secured. A secured lender has a lien, which is a legal right to specific assets of the company if it isn't paid. A first-lien lender stands first on those assets. A second-lien lender stands behind the first lien on the same assets. An unsecured lender has no lien at all and shares whatever is left after the secured claims are dealt with. The second is where the claim sits in the corporate structure. A loan made to the operating company that actually owns the factories ranks ahead of a bond issued by a holding company that only owns the shares of the operating company.

So when a distressed investor looks at a loan trading at 60 cents, the question isn't "will the price go back to 100." The question is "if this company goes through a restructuring, what does this claim get?" If the honest answer is 80 cents, buying at 60 is a good trade even if the company never recovers. If the honest answer is 20 cents, buying at 60 is a disaster even if everyone on the news says the company is too big to fail. In distressed investing, the price is the market's guess and the recovery is the answer, and the whole job is being better than the market at the guess.

2. The Waterfall Decides Who Gets Paid, and the Fulcrum Security Is the Seat That Gets the Company

Every distressed analysis starts with the same exercise. You line up the company's debt in order of priority, estimate what the company is worth, and pour the value down the line until it runs out. That's the waterfall.

Here's one with round numbers. A company has $1.2 billion of first-lien loans, $400 million of unsecured bonds, and equity held by a private equity sponsor. Its business has deteriorated, and after doing the valuation work you think the whole enterprise is worth $1 billion.

Pour the $1 billion down the stack. The first lien is owed $1.2 billion and there's only $1 billion available, so the first lien takes all of it and still comes up $200 million short. That's a recovery of $1 billion divided by $1.2 billion, about 83 cents on the dollar. The unsecured bonds get nothing. The equity gets nothing.

Now change one number. Suppose the company is worth $1.5 billion instead. The first lien is paid in full, $1.2 billion. The remaining $300 million goes to the $400 million of unsecured bonds, a recovery of 75 cents. The equity still gets nothing.

The class of debt where the value runs out is called the fulcrum security. In the first case it's the first lien. In the second case it's the unsecured bonds. The fulcrum matters because of a rule in the Bankruptcy Code called absolute priority. A class can't receive anything under a plan of reorganization unless every class senior to it has been paid in full or has agreed to take less. The rule lives in Section 1129(b) of the Code. So the fulcrum class is the one that's partly paid and partly not, and instead of cash for the unpaid part, it usually gets the new equity in the reorganized company. Everyone senior gets paid. Everyone junior gets wiped out or gets a small tip to avoid a fight. The fulcrum becomes the owner.

That's why the valuation number matters so much. Move the enterprise value from $1 billion to $1.5 billion and the fulcrum shifts from the lenders to the bondholders, and a completely different group of people ends up owning the company. Whoever holds the fulcrum security walks away with the reorganized company.

3. Trade One: Buy Below Recovery Value and Collect the Difference

The simpler trade is a math problem with a time dimension.

Say you've done the waterfall above and believe the first-lien loans will recover about 83 cents through a bankruptcy. The loans are trading at 60 because most holders are banks and mutual funds that either can't or don't want to own a bankrupt company's paper, and they're selling. You buy $50 million face value at 60, which costs you $30 million. Eighteen months later the plan is confirmed and your class receives 83 cents of value, some in cash and some in new debt or stock. That's $41.5 million on your $30 million, a gain of $11.5 million, about 38 percent.

Now the time dimension. If the case takes 18 months, 38 percent works out to roughly 24 percent a year. If a valuation fight drags the case to three years, the same gain is about 11 percent a year, and you've spent three years of legal fees to get it. Distressed investors think in annualized returns, so a slow case can turn a good trade into a mediocre one even when the recovery number was right.

Two more things eat into the trade. Legal and advisory fees paid by the company come out of the estate before creditors get anything, and in a large case they run into the tens of millions of dollars. And once the company files, interest on most of its old debt stops accruing, so unless you're a secured lender receiving protection payments during the case, your return is entirely in the final recovery.

The trade works when the investor knows something the sellers don't, or is willing to hold through a process the sellers can't. Both are real edges. Buying at 60 what will recover 83 is the trade; being right about the 83, and about how long it takes, is the job.

4. Trade Two: Loan-to-Own, Where the Debt Becomes the Company

The second trade is bigger, slower, and where most of the famous distressed returns came from.

Loan-to-own means buying the fulcrum security on purpose, so that when the company reorganizes and that class receives the new equity, you're the new owner. You've bought a company through its debt at a price set by frightened lenders, rather than through its stock at a price set by an auction.

The recent examples run through most of the big bankruptcies you've heard of. Frontier Communications is one: on top of the stock, its bondholders received $150 million of cash and $750 million of new notes. When Chesapeake Energy emerged in early 2021, the lenders under its first-lien last-out term loan received 76 percent of the new equity before new money. Second-lien holders and unsecured creditors got 12 percent each plus warrants. Intelsat came out of Chapter 11 in February 2022 as a private company owned by its former bondholders, with PIMCO the largest holder at about a third. Its debt went from roughly $16 billion to $7 billion. Revlon emerged in May 2023 owned by a group of its lenders, including Glendon, King Street, Angelo Gordon, Antara, Nut Tree, Oak Hill, and Cyrus, having eliminated more than $2.7 billion of debt.

The same thing happens outside court. In 2024, Vista Equity handed Pluralsight to its private credit lenders, a group that included Blue Owl, Ares, Golub, and Oaktree. The deal cut about $1.3 billion of debt and put in $250 million of new capital. Vista marked its equity to zero. The mechanism that makes loan-to-own possible in a sale process is credit bidding. Under Section 363(k) of the Bankruptcy Code, a secured lender can bid the face value of its debt instead of cash when the company's assets are auctioned. So a lender who bought the loan at 50 cents can bid 100 cents of paper for the business.

The reason loan-to-own pays so well is that the buyer controls the two numbers that matter. The entry price is whatever the loan was trading at. The exit value is whatever the company is worth once the debt is gone and someone competent is running it. Loan-to-own is buying a company for the price of its broken debt, and the bankruptcy process is the closing.

5. The Valuation Fight Is the Whole Game

If the fulcrum shifts with the valuation, then everyone in a bankruptcy has a reason to argue about the valuation, and they do.

The company, which is usually run by management and the sponsor, wants a high number so that junior classes, including the sponsor's equity, get something. The senior lenders want a low number so that the value "runs out" in their class and they get the equity cheaply. The unsecured creditors, who typically organize into an official committee, want a number high enough to put them in the money. Each side hires bankers to build the DCF and the comparables, and the court decides. In a contested case the judge is picking between two expert valuations that can be a billion dollars apart.

Two cases show how far the number can move. Hertz filed in May 2020 with about $19 billion of debt and every expectation that its shareholders would get nothing. Then used-car prices surged during the case, and a group led by Knighthead, Certares, and Apollo bid for the company. Under the 2021 plan, shareholders received $239 million in cash, 3 percent of the new equity, and warrants for another 18 percent, a package worth about $8 a share, in a case that began with the stock treated as worthless. The value moved during the case, and the fulcrum moved with it, all the way down to the equity.

Yellow Corp went the other way. When the trucking company shut down and filed in 2023, its real estate looked like it would sell for enough to pay everyone, and the stock traded as if shareholders would get a check. The real estate did sell, for about $2.4 billion. Then a court ruled in September 2024 that Yellow owed withdrawal liability to its pension funds, which had claimed $6.5 billion, and the liquidation plan confirmed in November 2025 put shareholders last in line, behind creditors the court thought were unlikely to be paid in full. The value was real. The claims ahead of the equity were bigger than the market had counted.

This is why the analyst's job is only half financial modeling. The other half is reading: the credit agreements, the indentures, the intercreditor agreements that set the order between lenders, and the court docket where the claims are filed. An analyst who builds a perfect DCF but misses a $6.5 billion pension claim gets the fulcrum wrong. In a bankruptcy, the recovery is set by the valuation the court accepts and the claims it counts, so the analyst who reads the documents beats the analyst who only builds the model.

6. In 2026 the Game Is Played Out of Court First

Ten years ago, this guide could have stopped at bankruptcy. It can't now, because most distressed situations get restructured out of court before anyone files, through deals called liability management exercises, or LMEs.

An LME is a transaction in which a stressed company and a subset of its lenders rewrite the capital structure using the loan documents themselves. The typical version is called an uptier. A majority of lenders lend new money in a new senior tranche and exchange their old loans into a second new tranche. The lenders who weren't invited get pushed to the back of the line. The J.Crew, Serta, and Del Monte deals are the standard examples. In 2025, about 65 percent of leveraged loan defaults were LMEs rather than bankruptcy filings or missed payments.

For a distressed investor this changed two things. First, the waterfall can be rearranged before you get to court. A loan you bought as first lien can become third-out overnight if you're on the wrong side of a majority vote. In July 2026, a Houston bankruptcy judge awarded the lenders who'd been excluded from Serta's 2020 uptier $261 million in damages plus interest, six years after the deal, though credit for earlier settlements later cut the payout to about $105 million plus interest. That's a long time to wait for a recovery. Second, recoveries themselves have fallen. Fitch's numbers on 2024 bankruptcies show it: first-lien debt of companies that had done an LME recovered 23 cents on the dollar, against 53 cents where there'd been no LME. In April 2026 Fitch reported that Chapter 11 recoveries in 2025 fell to a ten-year low.

The response from investors has been the cooperation agreement, where lenders holding a majority of a loan agree in advance to negotiate as a group and not to cut side deals. 9fin counted roughly 45 of them in the U.S. in 2024, up from about four a year over the six years before.

The practical result is that the modern distressed analyst spends as much time on documents as on models. Baskets, unrestricted subsidiaries, pro rata clauses, and the exceptions to them decide whether your first lien is actually first. In 2026, the first question about any distressed loan is what the documents let a majority of lenders do to you, before anyone gets to what the company is worth.

7. The Funds, and What Kind of Distressed Investor Each One Is

"Distressed debt fund" covers two different businesses, and you should know which one you're talking to.

The first is the hedge fund model. These funds hold liquid positions, trade in and out, and run the buy-below-recovery trade at scale, with loan-to-own when the opportunity is right. Elliott, Davidson Kempner, King Street, Silver Point, Monarch, Diameter, Brigade, Marathon, Canyon, Beach Point, Redwood, Whitebox, Mudrick, and Owl Creek all live here, in different sizes and styles. Anchorage Capital, once one of the largest, announced in December 2021 that it would close its $7.4 billion flagship fund, and it kept its structured credit and private-equity-style funds going.

The second is the drawdown fund model, which looks more like private equity. The fund raises committed capital from institutions, calls it over several years, and makes concentrated, often control-oriented investments. Oaktree is the reference name; its Opportunities Fund XII closed in February 2025 with about $16 billion. Centerbridge, Apollo's Hybrid Value strategy, Blackstone Credit and Insurance, Bain Capital Special Situations, Ares' opportunistic credit strategy, Sixth Street, HPS, Fortress, and Cerberus run versions of it.

The difference shows up in the job. At a hedge fund you might cover fifteen names, trade around them, and sit on an ad hoc lender group in two of them. At a drawdown fund you might spend a year on one situation, join the board after the reorganization, and hold the company for five years. Both need the same waterfall skills. They need different temperaments.

Right now, both are busy. S&P Global counted 785 U.S. corporate bankruptcy filings in 2025, the most since 2010, and 372 in the first half of 2026, the highest first half since 2010. The biggest of the year is still unfolding. First Brands, the auto parts company, filed in September 2025 with a $4.4 billion DIP loan, $3.3 billion of it a roll-up of existing debt. It never confirmed a plan. After the judge rejected its plan in late August 2026, the case was converted to a Chapter 7 liquidation, and a trustee is now selling what's left.

8. What the Interview Asks, and What a Good Answer Sounds Like

Distressed seats are rarely filled straight from undergrad. Most analysts spend two years in a restructuring group at PJT Partners, Houlihan Lokey, Evercore, Lazard, or Moelis, or in leveraged finance, and then recruit. A few funds do hire directly. Davidson Kempner runs a two-year rotational analyst program. Oaktree has advertised undergraduate summer internships on its Special Situations team, and Sixth Street, Blackstone, and Apollo take summer analysts who can land on credit teams. WSG's distressed debt career path guide covers the routes in detail. Whichever door you use, the interview tests the same four things.

The waterfall. You'll be handed a capital structure and a value, or a range of values, and asked to walk it. Practice section 2 until you can do it on a whiteboard in three minutes with the fulcrum identified at each value.

A pitch. "Give me a distressed credit you like." The answer needs a name, a price, your recovery estimate, what has to happen for you to be paid, and what breaks the trade. The interviewer will usually push on what happens if your catalyst falls through, because they're testing whether you've thought about being wrong, so have a downside recovery ready before you walk in.

The documents. "What happens to a second-lien lender in an uptier?" "What's a J.Crew blocker?" "What's the difference between a prepack and a free-fall?" These are checks on whether you've read a real case rather than a textbook.

And your reason for being there.

Why distressed debt?

Don't say: "I find it intellectually interesting because it combines law and finance."

Say: "I read the Serta docket after the Fifth Circuit ruling and realized the whole outcome turned on the words 'open market purchase.' I want to be in the seat where reading the document that closely is the job, and where the analysis ends in owning the company rather than advising on it."

What's a distressed credit you like?

Don't say: "I think the retail sector has a lot of opportunities right now."

Say: "[Company]'s first-lien term loan at 71. I have the enterprise value at [X] on [these comps], which puts first-lien recovery in the high 80s. The catalyst is the asset sale the company has to close by [date] under its DIP milestones. The trade breaks if the sale price comes in under [Y], where my recovery drops to the low 60s, so I'd size it small until the bids are in." Fill in the brackets with a real case you've followed. The shape is what the interviewer is grading.

A distressed interview tests whether you can pour value down a stack, defend a price, and read a document, so practice all three on a real case before the interview.

Where to Start

Pick a live case and follow it the way an analyst would. Saks Global filed in January 2026, First Brands is now in liquidation, and both have public dockets on their claims agents' websites. Pull the first-day declaration, which is the company's own account of what went wrong, build the capital structure from it, and write down a recovery for each class at three enterprise values. Then subscribe to the Pari Passu newsletter and read Stephen Moyer's Distressed Debt Analysis, a standard text on the subject. Everything you need to practice distressed analysis, from first-day declarations to claims registers, is public.