Introduction
An underwriting commitment gives a sponsor something close to a put option on its own debt. The bank promises at signing to fund the buyout on agreed terms; if investors later refuse those terms, the company borrows anyway and the bank keeps the difference. The bank's upside is the fee. Its downside is limited only by the flex caps in the fee letter and by how far prices fall before it sells. For years the option rarely paid out, because credit markets stayed open between signing and syndication. In 2022, as the Federal Reserve raised rates, it paid out on a string of buyouts signed in the boom, the mechanism traced in how interest rates drive M&A and LBO activity. The banks learned what a hung deal costs, how one is worked out and what a client remembers, lessons a financial sponsors group (FSG) now carries into every commitment it supports.
How an Underwriting Commitment Turns Into a Hung Deal
Three things have to line up. The commitment is signed with the acquisition, often months before closing: Vista Equity Partners and Evergreen Coast Capital, an Elliott Management affiliate, agreed to buy Citrix in January 2022 and closed at the end of September. Spreads and base rates then move before the debt is sold. And the move outruns the market flex explained in how banks underwrite sponsor debt: inside the caps the borrower pays for a weaker market, beyond them the arrangers pay from their fees and then their capital.
- Hung Deal
An acquisition financing that the underwriting banks are bound to fund but cannot sell to investors at a price inside their flex caps. The banks either sell it at a loss or keep the loans and bonds on their balance sheets until demand returns, and the sponsor's acquisition closes either way.
The last clause separates a hung deal from a failed buyout. Under the limited conditionality of sponsor commitment letters, a market collapse does not let lenders refuse to fund. Wachtell, Lipton, Rosen & Katz's review of 2022 financing markets records that banks followed through on Twitter, Nielsen, Citrix and Tenneco even when the syndicated markets failed to support the deals.
Citrix, Twitter and the Rest of the 2022 Book
The 2022 cases differ less in how they hung than in how the underwriters got out.
Citrix: Sell Early, Then Sell Again
The Citrix financing totalled about $15 billion, led by Bank of America, Credit Suisse and Goldman Sachs. As investors balked, the banks reshaped it with the sponsors' agreement, adding a euro term loan and turning $3.95 billion of unsecured bond commitments into second-lien debt, then sold the first-lien pieces in late September 2022. Bloomberg reported that about $6.5 billion stayed on the banks' books. The second leg came in April 2023, when a 33-bank group led by Goldman sold notes to repay the second-lien bridge at a price International Financing Review (IFR) put at 79 cents on the dollar:
| Piece sold | Amount | Price (cents) | Investor return | Date |
|---|---|---|---|---|
| Dollar and euro term loans | $4.55bn | about 91 | Margin of 4.50% | September 2022 |
| First-lien secured notes | $4.0bn | 83.561 | 10% yield | September 2022 |
| Second-lien notes, 9% coupon | $3.8bn | 79 | About 14% yield | April 2023 |
The two legs together cost the banks more than $1.3 billion on Bloomberg's estimate. The sponsor side helped clear the book: IFR reported that Elliott bought about $975 million of the September secured notes and $550 million of the second-lien notes. The company, combined with TIBCO as Cloud Software Group, later became the subject of a record single-asset continuation vehicle.
Twitter: Hold, Mark and Wait
Twitter was not a sponsor buyout, but its financing became a hung position on a similar scale. Elon Musk's $44 billion purchase, closed in October 2022 after Musk had tried to abandon it, was financed partly by seven banks led by Morgan Stanley: a $6.5 billion secured term loan, $3 billion each of secured and unsecured loans meant for the bond market, and a $500 million revolver. The banks held the debt for more than two years, as the Twitter deal case study recounts, then sold it between February and April 2025, according to Reuters: $5.5 billion of the term loan at 97 cents, fixed-rate secured loans at par, and a final $1.2 billion at 98.
The two cases bracket the choice. The Citrix banks fixed their loss within months and freed capacity; the Twitter banks earned the interest, absorbed quarterly marks, and sold near par once demand returned.
Brightspeed and Tenneco: The Long Tail
Apollo agreed in 2021 to buy Lumen Technologies' local telephone business in 20 states for $7.5 billion, renamed Brightspeed; a Bank of America and Barclays-led group pulled its roughly $3.9 billion debt sale in late September 2022, and Bloomberg reported banks still trying to cut that exposure in October 2024. At Tenneco, the underwriters funded most of the $5.4 billion package at closing, sold more than $3 billion of it in August 2023 at about 85 cents, and were still offering about $365 million of the rest in April 2024.
How Banks Absorb the Loss
A bank holding a hung financing has four routes, each hitting a different line of its results, and most 2022 underwriters used more than one:
| Route | What the bank does | Where the cost shows | 2022 case |
|---|---|---|---|
| Sell at a discount | Sells below the committed price after using all flex | Realized loss, net of fees | Citrix first lien |
| Hold and mark | Funds and keeps the loans until demand returns | Fair-value marks, capital, limits | |
| Re-cut and refinance | Reshapes tranches with the sponsor, then sells | Loss on each leg | Citrix second lien |
| Hedge | Buys index or single-name credit protection | Gains offset part of the marks | Disclosed only in aggregate |
The marks arrive before any sale, which is why banks reported markdowns on leveraged loan positions through 2022 while most of the hung debt was still unsold.
- Loan Held for Sale
A loan a bank has made or committed to make with the intention of selling it rather than keeping it. Under US accounting rules it is carried at the lower of cost or fair value, or at fair value where the bank elects that option, so a fall in its market price reduces earnings before the loan is sold.
Holding therefore defers the realized loss rather than avoiding it, and meanwhile ties up regulatory capital and underwriting capacity that cannot serve new clients. It does earn interest: on Bloomberg's account, hefty coupons eased the pain for the Tenneco lenders before they sold.
What Changed After 2022, and What Sponsors Remember
The first response came in commitment terms: the Wachtell review describes lenders that grew reluctant to commit and demanded greater economics and more onerous flex when they did. Direct lenders filled the gap, providing some or all of the debt in six of the ten largest announced leveraged buyouts (LBOs) of 2022 by the same review's count; the bank partnerships that followed are set out in syndicated vs private credit.
The lasting change sits in the questions a commitment committee asks before it signs:
- Exposure time: how long the debt will stay unsold, including regulatory review.
- Flex room: whether the caps cover a plausible market move, not only today's spread.
- Risk partners: whether a direct lender or the sponsor's co-investors could share the risk at signing.
- Hold appetite: whether the bank would accept owning the loan for years.
The Coverage Banker on the Wrong Side of a Client's Deal
A hung deal puts the coverage banker between two interests. Leveraged finance (LevFin) and risk managers decide when to sell and at what price; the sponsor wants the sale handled without damaging the company. A deep-discount sale publishes a price for the company's debt that rating agencies and future lenders read, and can hand part of the capital structure to holders the sponsor did not choose.
The coverage job is to warn the sponsor before the desk moves, bring options the sponsor can help with, as Elliott did at Citrix, and keep the hung position apart from the next mandate. Sponsors keep score. The 2008 contrast, when Bain Capital and Thomas H. Lee Partners sued their banks to fund Clear Channel, is traced in how sponsors choose their banks; in 2022 the banks funded, and the relationship ledger noted which did so cleanly.
The 2022 book also showed that no exit route is right in advance: Citrix's banks sold early, Twitter's held and sold near par, Brightspeed's held and were still seeking buyers two years later. Each choice came after the market had moved, when every option already carried a cost.
The decisions that set the size of those costs came earlier, at signing: how much the bank committed, how long the commitment could stay open, how wide the flex ran and whether anyone shared the risk. Those terms are negotiated when the relationship is warmest and the market calmest, which is exactly when a committee is most tempted to treat them as formalities.
