The 3.3% rupture threshold: how a small crash bankrupted a 30:1 bank
The essentials in 30 seconds
- A rupture threshold is the fall in asset values that erases a bank’s capital, and leverage sets its size.
- At 30:1 leverage the threshold is 3.3%, and Lehman Brothers sat at 30.7:1 with a threshold of 3.3%.
- The EQOS Asia 2008 back-test puts the actual decline of 2007 and 2008 at about 20%, 6 times the threshold.
- The same division gives a rough threshold for any firm that publishes its leverage.
At 30:1 leverage, a small stock market crash is enough to bankrupt a bank: a fall of only 3.3% in asset values erases its capital. This page shows the arithmetic, the banks that sat near this line in 2007, and the method you can reuse, based on the EQOS Asia 2008 back-test (EQOS Asia 2008 back-test, 2026). It is one of the pages under the EQOS score of a stock market, written as of October 2026.
On this page
Compare the EQOS Asia offers and request a 30-minute pre-diagnostic.
What is the rupture threshold of a bank in a stock market crash?
The rupture threshold is the size of fall in asset values that wipes out a bank’s own capital. Below it, the bank absorbs the loss and continues. Above it, the bank owes more than it owns, and creditors stop lending. The threshold turns a vague word, fragility, into one percentage that you can read and compare.
Leverage sets the threshold. The more a bank borrows for each unit of its own capital, the smaller the fall it can survive. The EQOS Asia study calls the result a tipping threshold, because the system moves from stable to unstable once a fall crosses it, with no gradual warning in between. That is why a stable-looking balance sheet can hide a system one small shock away from failure.
How does 30:1 leverage turn into a 3.3% rupture threshold?
Take a bank with $100 of assets at 30:1 leverage. It owns about $3.3 and borrows about $96.7. If its assets lose 3.3%, they fall to $96.7, which equals what it owes, and its capital reaches zero. The threshold is roughly 1 divided by the leverage ratio, and 1 divided by 30 gives 3.3%.
The same division works at other ratios. At 15:1, the threshold becomes about 6.7%. At 12:1, it becomes about 8.3%. Each step down in leverage buys a larger buffer, which explains why the study treats leverage as a lever that regulators and investors can read directly from a filing.
Which 2007 banks sat near the 3.3% rupture threshold?
From SEC 10-K filings, the study lists Bear Stearns at 33:1 with a threshold of 3.0%, Lehman Brothers at 30.7:1 with 3.3%, Goldman Sachs at 25.2:1 with 4.0%, and Morgan Stanley at 33.4:1 with 3.0% (EQOS Asia 2008 back-test, 2026). All 4 sat within a few points of the same fragile line.
The table shows how thin the buffer was. A fall of 3% in the value of mortgage-linked assets was enough to threaten two of the largest investment banks. Nothing here required inside information, because the leverage figures were published in annual filings that any analyst could open. Seen from the threshold, the surprise of 2008 was a surprise of attention, not of data.
How far was the 3.3% rupture threshold from the Basel II band?
The study sets a prudential reference of 12 to 15:1 under Basel II, with a rupture threshold of 7% to 8%. Against that band, the banks of 2007 carried about twice the leverage and half the buffer. The gap between the published norm and the observed practice is the first signal of the page.
A norm only protects when practice stays inside it. When the observed ratio drifts far from the reference, the reference no longer describes the system. A reader who compares the two numbers each quarter sees the drift early, before any loss appears on a balance sheet.
Why was the actual decline 6 times the 3.3% rupture threshold?
The actual decline in 2007 and 2008 was about 20%, according to the study, which is roughly 6 times the 3.3% threshold. A system with a buffer of 3.3% faces a shock of 20%, and the outcome follows from the arithmetic: capital disappears long before the fall ends.
This ratio explains why the failures came in sequence. Bear Stearns was sold for $2 a share in March 2008, against $172 a year earlier. Lehman Brothers filed with $619 billion of debt on 15 September 2008, the largest default in history according to the study.
Which leverage rule changes narrowed bank buffers before the 2008 crash?
The study lists decisions that reduced the buffer. In 2004 the leverage rules for the 5 largest banks moved from 15:1 to unlimited, and from 2003 to 2006 the Fed held rates at 1%, which encouraged borrowing and real-estate speculation. Each decision lowered the integrity of the frame. The study reads them as public facts, voted and documented, that fed the integrity dimension.
Reading rules this way connects policy to arithmetic. A change in a leverage rule moves the threshold of every bank at once, so the whole system tips together. The study therefore recommends restoring a 15:1 cap, which would place the threshold near 6.7% and raise stability from 0.08 to a target of 0.40.
How does a bank at its threshold pass losses to the wider market?
When a leveraged bank nears its threshold, it sells assets to cut debt, and the sales push prices down for every other holder. Lenders then ask for more collateral, which forces more sales. The study notes that credit default swaps of $62 trillion stood against a world output of $54 trillion, so the web of claims was large.
The loop moves from one bank to the market. A fall that begins in mortgage-linked assets reaches funding markets, then share prices, then jobs. The study counts 10 million foreclosures, 8 million jobs lost, and $22 trillion of value destroyed, by IMF estimate.
How can you estimate a rupture threshold for a firm today?
Find the firm’s total assets and its equity in the latest annual report, divide assets by equity to get leverage, then divide 1 by that leverage. The result is a rough threshold. Compare it with the typical range for similar firms and with the fall that a stress scenario would imply.
This page is general information, not investment advice, and the division gives an order of magnitude, not a verdict. The Λ simulator places the threshold inside the full φ(S) reading in 10 minutes, free and anonymous, and the method page explains how the stability dimension uses it.
See also in this guide
Frequently asked questions
What is the 3.3% rupture threshold?
It is the fall in asset values that erased the capital of a bank running at about 30:1 leverage. The EQOS Asia 2008 back-test gives Lehman Brothers a threshold of 3.3% at 30.7:1, on 2007 filings.
How do you compute a rupture threshold?
Divide 1 by the leverage ratio, where leverage is assets over equity. At 30:1 the result is about 3.3%, at 15:1 about 6.7%. The division gives an order of magnitude and does not replace a full stress test.
How large was the actual fall in 2007 and 2008?
The study puts the actual decline at about 20%, roughly 6 times the 3.3% threshold. With a buffer that small, the failures of Bear Stearns in March 2008 and Lehman Brothers in September 2008 followed from the arithmetic.
Does the threshold predict a crash date?
No. It sizes the fall that a system can absorb and leaves timing open. This page is general information, not investment advice, and a separate projection, the point of no return, gives a date range.
Test your knowledge
1. Which rupture threshold goes with 30:1 leverage?
- About 3.3%
- About 10%
- About 25%
Show the answer
About 3.3%.
2. What did the study give as the actual decline of 2007 and 2008?
- About 2%
- About 20%
- About 60%
Show the answer
About 20%.
3. How do you estimate a rough threshold?
- Multiply assets by equity
- Add debt and equity
- Divide 1 by the leverage ratio
Show the answer
Divide 1 by the leverage ratio.
4. At 33:1 leverage, what threshold did Bear Stearns show?
- 3.0%
- 8.0%
- 12.0%
Show the answer
3.0%.
5. What Basel II reference leverage did the study use?
- 30 to 1
- 12 to 15 to 1
- 50 to 1
Show the answer
12 to 15 to 1.
About the author
Stéphane Jambu
President, EQOS Forecast Asia
Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile
Sources: EQOS Asia 2008 back-test (eqos-asia.com/case-studies/crise-2008.html), EQOS Asia score page (eqos-asia.com/stock-market-crash/stock-market-crash-score.html), simulator (eqos-asia.com/simulator.html) and method page (eqos-asia.com/method.html), consulted on 22 October 2026. Page updated on 22 October 2026.