Hung Out to Dry: The $13 Billion Debt Gamble Behind Musk’s Twitter $44 Billion Takeover

---

Executive Summary

Elon Musk’s acquisition of Twitter, Inc. (referred to as X following the rebrand) represents one of the largest leveraged take-privates in history by total consideration and one of the most structurally unusual. The $44 billion take-private – completed on 27 October 2022 at $54.20 per share – was financed through an unprecedented combination of personal equity from a single individual, co-investor commitments, and $13 billion in bank debt underwritten by a seven-bank syndicate led by Morgan Stanley.

Unlike a conventional PE-sponsored LBO, Musk’s acquisition was majority equity-financed at approximately 70:30 equity-to-debt – the inverse of typical PE practice, where equity-to-debt ratios of 40:60 or higher are standard. However, the deal’s defining feature from a debt finance perspective was not the leverage ratio, but what followed: the underwriting banks were unable to syndicate the $13 billion debt package for over two years, creating what the Wall Street Journal characterised as the worst merger-finance deal for banks since the 2008 financial crisis.

This analysis examines the sources and uses, the structure and terms of each debt tranche, the syndication failure and its consequences, and the eventual resolution of the hung debt through 2025.

---

Capital Structure Overview

Musk’s equity commitment started at $21.0 billion when the deal was signed on 25 April 2022. At the same time, banks committed a $12.5 billion margin loan – essentially a personal loan to Musk secured against his Tesla shares. Nine days later, the margin loan was halved to $6.25 billion and Musk’s equity commitment was increased to $27.25 billion to compensate. The margin loan was reportedly eliminated entirely before closing, with Musk selling approximately $15.5 billion of Tesla shares to cover the shortfall.

The total amount paid by Musk exceeds the $44 billion headline price because the acquirer also had to repay Twitter’s pre-existing corporate debt of approximately $5.3 billion and cover transaction costs. This is standard in acquisitions – the buyer doesn’t just purchase the equity; they inherit the balance sheet.

Leverage and Credit Metrics

Leverage metrics tell lenders how much debt a company is carrying relative to its ability to generate cash. The most important ratio is Total Debt / EBITDA – essentially, how many years of operating earnings it would take to repay all the debt.

Twitter’s Financials

Post-Acquisition Leverage (New Debt Only)

The leverage multiple depends on which EBITDA figure you use. The three common definitions strip out different items, producing materially different results:

Recurring EBITDA: The most conservative measure. This is calculated by taking operating earnings and adding back depreciation and amortisation but removing one-off gains – such as the $101 million in realised gains on securities that Twitter reported. This gives the cleanest picture of what the business earns on a repeatable, sustainable basis.

EBITDA unadjusted: A broader measure that includes those one-off items. The gap between this and the Recurring EBITDA largely reflects non-recurring gains and other items that flattered Twitters earnings in the relevant period but would not recur every year.

Adjusted EBITDA: The most generous measure, and the one banks typically use when marketing leveraged loans. This starts with reported EBITDA and adds back non-cash charges – most significantly stock-based compensation (SBC), which was a major expense for Twitter as a tech company. SBC is a real cost to shareholders (it dilutes their ownership), but it does not consume cash, so lenders often exclude it when assessing a companys ability to service debt. The banks would have negotiated specific add-backs in the credit agreement to arrive at this figure.

Even on the most generous basis (~10x), the leverage was roughly double the market norm. The interest coverage ratio – which measures whether a company can pay its interest bills from operating earnings – fell below 1.0x once interest rates rose. This meant X could not cover its annual interest expense from earnings alone.

Critically, Twitter’s financials were already deteriorating before the deal closed. Between March and June 2022, Recurring EBITDA halved from $667M to $331M, and the interest coverage ratio fell from 12.48x to 5.27x – all on the pre-acquisition balance sheet, before any LBO debt was added.

Credit Rating: Critically, S&P downgraded Twitter from BB+ to B- on 1 November 2022 (four days after closing), then withdrew the rating entirely on 2 December 2022. Without a credit rating, the banks’ plan to refinance the bridge loans through bond markets became impossible – a point explored further in Section 4.

---

Senior Facilities Analysis

The $13 billion was divided into four separate facilities, each with different terms, pricing, and levels of protection for lenders. Understanding the hierarchy between these facilities is essential to understanding why the debt played out as it did.

SOFR is the Secured Overnight Financing Rate – a benchmark interest rate that fluctuates with central bank policy. “+475 bps” means the borrower pays SOFR plus 4.75%. When SOFR was 0.28% (April 2022), the term loan cost roughly 5%. When SOFR hit 5.33% (July 2023), the same loan cost over 10%.

Margin and Pricing

Every facility was priced on a floating-rate basis over SOFR, meaning the interest cost moved with central bank rate decisions. This was the deal’s critical vulnerability. When the commitment letters were signed in April 2022, rates were near historic lows, which made large loans attractive for lenders. However, the Federal Reserve then embarked on its most aggressive tightening cycle in decades. SOFR rose nearly 20-fold from 0.28% to 5.33% over the next 15 months, causing the interest rate on the loans to skyrocket.

Fortunately, the bridge loans included interest rate caps – contractual ceilings that prevented rates from rising above approximately 9.75% (secured) and 11.75% (unsecured). These caps protected Musk from unlimited rate exposure, but they also made the debt unattractive to potential buyers, posing a major issue for the banks involved – investors in the secondary market could find better-yielding opportunities elsewhere for the level of credit risk involved. As a result, the capped rates made the debts essentially illiquid, effectively trapping the banks.

Both bridges also carried a 50 basis point step-up for each quarter they remained outstanding beyond their one-year maturity – a standard mechanism designed to incentivise the borrower to refinance quickly. When refinancing proved impossible, these step-ups simply increased X’s interest burden over time. No original issue discount (OID) has been publicly disclosed.

Amortisation

The term loan repaid 1% of principal per year, with the rest due at maturity (a “bullet” payment). The loan documents also required the company to use a portion of its spare cash to pay down debt: 50% of annual excess cash flow, reducing to 25% and then 0% as leverage improved. Proceeds from asset sales were also swept towards debt repayment, subject to a 540-day window allowing the company to reinvest the money instead.

Financial Covenants

Covenants are the rules lenders impose on borrowers to protect their position. They come in two forms:

Maintenance covenants are tested regularly (usually quarterly) regardless of what the borrower does. If the companys leverage rises above the covenant threshold, it is in breach – even if it took no action to cause the deterioration.

Incurrence covenants are only tested when the borrower takes a specific action, such as borrowing more money or paying a dividend. The company can deteriorate operationally without triggering a breach, as long as it doesnt take the specified actions.

The term loan was explicitly “covenant-lite” – meaning it used incurrence-based covenants only. This was standard market practice in 2022 for loans intended to be sold to institutional investors. It gave the borrower significant flexibility but offered lenders less protection.

The revolving facility had one maintenance covenant: a maximum leverage ratio, but it was only tested when the company had drawn more than 35% of the facility (~$175M). The loan documents also allowed Musk to inject cash up to five times to fix a covenant breach (known as an “equity cure”), with no more than two cures in any four consecutive quarters.

Security Package

“Security” refers to the assets that lenders can seize if the borrower fails to repay. The collateral included substantially all of Twitter’s assets – intellectual property, equipment, receivables, subsidiary shares, and other property. However, real estate was explicitly excluded, which is notable since Twitter held valuable long-term leases on its San Francisco headquarters and data centres.

The critical structural point is how the different loans ranked against each other in terms of who gets paid first:

Term loan, revolver, and secured bridge: All held equal first-priority claims on the same assets. If the company defaulted and assets were sold, these three groups would share the proceeds proportionally before anyone else received a penny. This is called a pari passu (Latin for on equal footing) arrangement.

Unsecured bridge: Had no claim on any assets whatsoever. These lenders sat behind all secured creditors and would only be repaid from whatever was left over.

This explains why the unsecured bridge carried the highest interest rate (SOFR + 10%) and was the last piece of debt to be sold (April 2025) – lenders demanded more compensation because they bore the greatest risk of losing their money.

Intercreditor Arrangements

When multiple groups of lenders have claims on the same borrower, a formal agreement called an intercreditor agreement sets out who gets paid first and who can take enforcement action.

In a default, proceeds from asset sales would go first to the three groups of first-lien lenders (term loan, revolver, and secured bridge) on a proportional basis. Only after those lenders were repaid in full would any remaining money flow to the unsecured bridge lenders. The loan documents also permitted the borrower to take on additional debt of up to the greater of $1.7 billion or 100% of annual EBITDA, plus further amounts subject to leverage tests – providing headroom for future financing needs.

---

High Yield Bond Analysis

Why the Planned Bond Issuance Never Happened

The two bridge loans were always intended to be temporary. The plan was to replace them within months with permanent bonds – fixed-rate securities sold to institutional investors through a process called a Rule 144A offering (a mechanism under U.S. securities law that allows bonds to be sold to large professional investors without the full SEC registration that a stock market listing would require).

The commitment letter set out detailed requirements for this bond placement, including a 15 business day marketing window and specified blackout periods around U.S. holidays.

The problem was that bond investors almost always require the issuer to have a credit rating – an independent assessment of creditworthiness from an agency like S&P or Moody’s. When S&P withdrew Twitter’s rating in December 2022, the bond market was effectively closed. Without a rating, no institutional investor would buy.

As a result, both bridge loans automatically converted into longer-term loans when they hit their one-year maturities in October 2023 – exactly as the loan documents provided for. The secured bridge became a 7-year loan; the unsecured bridge became an 8-year loan, both at their maximum interest rates. What was supposed to be temporary financing became permanent (and expensive) debt.

The April 2025 Refinancing

The bond market eventually played a role, though not as originally planned. In April 2025, Morgan Stanley led a bond offering to replace the last of the bridge-related debt with a fixed-rate bond at 9.5% – down from the roughly 14% floating rate that had accumulated through step-ups. X projected annual interest savings of $43 million. This achieved the refinancing the original plan contemplated, roughly two and a half years late.

---

Acquisition Agreement Interface

This section looks at how the loan documents and the acquisition agreement (the legal contract to buy Twitter) interacted. In any leveraged take-private, these two sets of documents must work together: the loans need to be available when the deal closes, and the acquisition agreement needs to address what happens if the financing falls through.

Certain Funds Provisions

“Certain funds” is a mechanism that protects the buyer by limiting the circumstances in which banks can refuse to provide the money they promised. Without it, a bank could agree to lend $13 billion and then back out at the last moment, claiming market conditions had changed.

The commitment letter was emphatic: syndication (the banks’ ability to sell the debt to other investors) was explicitly not a condition to funding. The banks were obligated to provide the full $13 billion regardless of whether they could find other investors to take it off their hands. The only conditions that could prevent funding were a narrow, specific list:

• The acquisition must close as agreed

• Musk’s equity must be contributed simultaneously (at least 40% of the total)

• No material adverse effect on Twitter since signing

• Twitter’s existing ~$5.3 billion of debt must be paid off simultaneously

• Various administrative items: fees paid, financial statements delivered, legal documents signed, anti-money laundering checks completed

Anything not on this list – including a general deterioration in Twitter’s business, a stock market crash, or the banks’ inability to find other investors – was not grounds for refusing to lend. This is what made the certain funds structure so significant: it placed the risk of being stuck with the debt squarely on the banks.

Expiry date: The banks’ commitments expired at the merger agreement’s outside date of 25 April 2023 – a year after the documents were filed. As a result, had the legal battle between Musk and Twitter dragged past that deadline, the banks’ obligations would have lapsed entirely, causing a collapse of the entire deal.

Financing Condition in SPA

The acquisition agreement contained no financing condition – meaning Musk could not refuse to complete the purchase simply because his financing fell through. This is standard for public company take-privates.

If Musk failed to close, he owed Twitter a $1 billion termination fee (a “reverse break fee”), which he personally guaranteed. While $1 billion sounds large, it was only about 2.3% of the deal value – a limited penalty for someone of Musk’s wealth.

The most important protection for Twitter was a specific performance clause – a legal mechanism that allows a court to order someone to actually do what they promised, rather than simply paying compensation for breaking their promise. Section 9.9 of the merger agreement entitled Twitter to force Musk to fund his equity and complete the purchase, provided three conditions were met: the deal conditions were satisfied, the bank debt was available, and Twitter confirmed it was ready to close. Crucially, Twitter had to choose between seeking specific performance and collecting the $1 billion fee – it could not pursue both.

This clause proved decisive. When Musk tried to cancel the deal on 8 July 2022 – claiming Twitter had lied about the number of fake accounts on the platform – Twitter immediately sued in Delaware seeking specific performance. The court scheduled a fast-track trial for October 2022, and legal commentators widely assessed that Twitter would win. Musk reversed course days before trial and closed the deal on the original terms.

For the banks, Musk’s attempt to walk away created a painful period of uncertainty. Their commitment letters obligated them to lend if the deal closed, but whether the deal would actually close was now a question for the courts.

---

Underwriting and Syndication

Morgan Stanley’s position was unique: it served as both Musk’s deal advisor and the lead bank on the debt, giving it the largest loan commitment and the greatest financial exposure when things went wrong.

What Syndication Is and Why It Failed

How it is supposed to work: When banks underwrite a leveraged buyout, they commit to lend the full amount but plan to sell most of it to other investors – mainly CLOs (vehicles that pool corporate loans) and specialist credit funds – within weeks of closing. The banks earn fees for arranging the deal and then transfer the risk. This “originate to distribute” model is the backbone of the leveraged finance market.

The commitment letter required Musk to help with this process: attending at least one lender meeting (maximum two hours), helping prepare marketing materials, and obtaining credit ratings. The banks also had the right to block certain investors from buying the debt (a “Disqualified Lenders” provision), and Musk had the right to veto specific buyers – an unusual degree of borrower control that may have further limited the banks’ options.

What went wrong: Syndication failed across all loan tranches. The term loan attracted bids of only about 60 cents on the dollar – meaning the banks would have lost roughly $2.6 billion if they had sold. They declined. The bridge loans could not be converted to bonds because the credit rating was withdrawn.

Several factors combined to make the debt unsellable: advertising revenue fell approximately 50% as major brands abandoned the platform; Musk cut 80% of staff and publicly antagonised advertisers, destroying investor confidence; the company stopped producing the regular financial reports that institutional investors need to assess credit risk; and rising interest rates simultaneously made all leveraged debt less attractive.

The major banks – Morgan Stanley, Bank of America, and Barclays – reportedly signed a joint agreement to coordinate any future sales, preventing any one bank from dumping its position in a way that would push prices down for the others.

The eventual sell-down (January–April 2025): A change in sentiment – driven by Musk’s political relationship with President Trump, the xAI merger (which valued the combined entity at $45 billion), and X’s reported EBITDA of approximately $1.5 billion – finally allowed the banks to sell:

• January 2025: $1B sold to Diameter Capital Partners (initial test of demand)

• February 2025: $5.5B sold at 97 cents on the dollar (buyers: Pimco, Citadel, Apollo, Darsana)

• March–April 2025: Further sales at 101–102 cents (above the original face value)

• April 2025: Final tranche sold at 98 cents, clearing the last debt from bank balance sheets

The banks earned approximately $3 billion in cumulative interest over the two-and-a-half-year period – more than enough to cover the small discounts on the eventual sales. However, the capital locked up on their balance sheets represented a significant opportunity cost and drew regulatory attention.

---

Market Context

This table offers a useful contrast – particularly the Walgreens/ Sycamore deal. Sycamore used far more debt proportionally (83% vs Musk’s 30%) but at a much lower multiple of earnings (~6.4x vs 10–19.5x). The consequence is that Walgreens carries a heavier repayment burden relative to its equity cushion, but the debt is serviceable – the business earns enough to cover its interest.

On the other hand, Twitter had the opposite problem: a large equity cushion protecting lenders from losses, but operating earnings that could not cover the interest bill. High leverage percentage is survivable if the multiple is low; a low debt percentage is dangerous if the multiple is extreme.

Furthermore, Walgreens also spread risk across several types of lender – asset-based lenders secured against liquid inventory, private credit funds that hold to maturity, and term loan banks taking smaller individual slices. When Twitter’s syndication failed, all seven banks were stuck simultaneously with $13 billion of homogeneous floating-rate debt and no alternative exit, because they all needed to sell to the same pool of institutional buyers.

Market Conditions

Interest rates: The loans were committed during the early stages of the most aggressive rate-hiking cycle in decades. SOFR rose from 0.28% to 5.33% over 15 months, nearly transforming the projected annual interest bill from ~$860M to ~$1.5B.

Institutional demand: The CLO market (the main source of demand for leveraged loans) remained active but became more selective. An unrated technology credit with no public financial reporting could not compete against better-documented opportunities.

Private credit: Private credit funds – firms like HPS, Ares, and Apollo that lend directly to companies and hold loans to maturity – played no role in the original Twitter financing. This contrasts with the Walgreens deal, where private credit contributed ~$4.5B. These lenders could have offered execution certainty since they don’t rely on syndication, but the speed of the deal process and the assumption that conventional syndication would work meant they were never brought in.