Most collapses are stories about a company running out of money.
This one is not. Bear Stearns met its regulatory capital requirements on the day it agreed to be sold. Its collateral was good. Its assets were real. On Monday, 10 March 2008, it held roughly $18 billion in cash.
By Thursday that number was around $2 billion. By Sunday the firm had agreed to sell itself to JPMorgan Chase for $2 a share, later renegotiated to $10. At its peak in January 2007, the stock had traded above $170.
Nothing about the assets changed in those four days. What changed was that the people lending against them stopped.
THIS WEEK'S AUTOPSY
Bear Stearns funded itself in the overnight repo market. It pledged securities as collateral, borrowed cash against them, and repaid the following morning, every morning, in the tens of billions. This was not hidden, and it was not unusual. Every major broker-dealer did a version of it, and Bear's funding model was fully disclosed in its filings.
The weakness in that model is that it has to be renewed daily. A lender who simply declines to roll a loan has not made an accusation, breached a contract, or taken a loss. They have chosen not to enter a new one, and that choice costs them nothing.
That is what makes counterparty collapse different from every other mechanism covered on this channel. There was no fraud to uncover, no restatement, no concealed leverage. Bear's problem was that its survival required thousands of separate decisions to say yes each morning and only one collective decision to say no.
The signal had been visible since June 2007, when two Bear-managed hedge funds heavily exposed to subprime mortgage securities collapsed. That was nine months before the firm did. It told the market two things: that Bear's name was attached to subprime losses, and that when the losses arrived, Bear had chosen to support the funds rather than let them fail alone.
By March 2008 the rumors were enough. Hedge fund clients moved their balances elsewhere. Repo lenders declined to roll. Some counterparties refused to trade with the firm at all. None of them needed to believe Bear was insolvent. They only needed to believe that others might stop first, because the lender left holding the position when the music stops is the one who takes the loss.
The Federal Reserve extended roughly $30 billion in support to complete the JPMorgan transaction. The Madison Avenue headquarters alone was widely valued at more than the entire $2-a-share offer for the firm's equity.
One point worth stating plainly. Two Bear Stearns hedge fund managers, Ralph Cioffi and Matthew Tannin, were tried and acquitted on every count in November 2009, later settling with the SEC without admitting wrongdoing. Chief executive James Cayne and his successor Alan Schwartz were never charged with anything. Eighteen previous autopsies on this channel found a person at the center. This one finds a structure, and that is the honest finding rather than a gap.
The lesson for investors is uncomfortable, because it cannot be solved by reading harder. Solvency and funding are different questions. A company can pass every solvency test and still fail if what it owns takes longer to sell than what it owes takes to come due. The place that risk is visible is the maturity profile of the liabilities, not the quality of the assets.
THE WATCHLIST
Three companies whose position depends less on what they own than on whether their lenders keep saying yes.
Mercer International is the closest current parallel. The pulp and lumber producer added going concern disclosure to its second-quarter 2026 statements, and the reason is instructive: the disclosure stems from its revolving credit facilities being reclassified to current liabilities. Its Canadian revolver matures in January 2027, and management has said it expects to renegotiate or replace it. Aggregate liquidity fell by $37 million in the quarter to about $192 million. This is not an insolvency finding. It is an accounting consequence of a renewal date drawing closer, which is precisely the distinction this week's episode is about.
Tucows disclosed substantial doubt about its Ting Fiber subsidiary's ability to meet obligations within a year without additional financing, with strategic alternatives under review. The wider group carries a drawn revolver, roughly $300 million in term notes, and a stockholders' deficit above $200 million. Infrastructure businesses consume cash years before they return it, which makes continued lender appetite a structural requirement rather than a convenience.
BioXcel Therapeutics shows what it looks like when the lenders are already in control. Its credit agreement was amended to extend a deadline to 21 August 2026 for the company to enter a transaction that either repays the lenders in full or provides alternative capital on terms acceptable to them. The minimum liquidity covenant was cut to $3 million. When the financing options a company may pursue require the approval of its existing creditors, the creditors are setting the strategy.
None of these three is accused of any wrongdoing. They are on the list because their maturity profiles are public, and a maturity profile is the one warning sign that arrives with a date attached.
THE VULTURE'S PICK
Every issue The Vulture shares one tool or resource actually used in the research. Coming in the next issue.
NEW ON YOUTUBE
The full Bear Stearns autopsy is live on the channel. It covers the overnight repo model, the June 2007 hedge fund failures, the four days in March, the Fed-brokered sale, and why a firm that was never insolvent did not survive the week.
Watch it here: https://www.youtube.com/@thewallstreetvulture
NEXT ISSUE
March 2023. Another bank, fifteen years later, and the opposite problem. Silicon Valley Bank's depositors were not secured lenders holding good collateral. They were unsecured, they were more than ninety percent uninsured, and they all knew each other. Forty-two billion dollars was requested in a single day.
— The Vulture