In The Big Short, Michael Lewis records a strawberry picker in California with an annual income of fourteen thousand dollars who was lent the entire purchase price of a house costing more than seven hundred thousand. Somebody approved that loan. Somebody bought it and bundled it with thousands of others into a bond. Somebody stamped the bond with the highest safety grade that exists. Somebody's pension fund, possibly on the other side of the world, bought it as a conservative investment. Every one of those somebodies was paid for their step in the chain, and not one of them planned to be holding the loan when the strawberry picker stopped paying.
This is the last mania in our series, and the only one you may personally remember. Tulips, the South Sea, 1929, and the dot-coms were about buying an asset at any price. 2008 was something new: a mania in lending, built by professionals, graded safe by professionals, and detonated underneath people who never bought anything more exotic than a home or a pension.
What people believed
Three beliefs, stacked on top of each other.
First: house prices never fall nationwide. Regionally, yes, everyone knew a city could slump. But a fall across the whole United States at once had not happened in living memory, and an entire industry of risk models quietly converted "has not happened lately" into "cannot happen." Every calculation downstream inherited that assumption.
Second: a home is not just shelter, it is an ATM. As prices rose, homeowners refinanced, replacing their mortgage with a bigger one and spending the difference. The house had become a source of income, which works beautifully for exactly as long as belief number one holds.
Third, and most fatal: if a risky loan is sliced up and mixed with thousands of other risky loans, the risk somehow goes away. It does not. Risk does not disappear when you slice it. It just becomes very hard to see, and, as Edward Chancellor's history of manias keeps showing, what a crowd cannot see it happily prices at zero. Devil Take the Hindmost was published in 1999, nine years before the crash, and the pattern it describes needed no update.
The machine, in plain language
Here is the whole machine, one part at a time.
A subprime mortgage is a home loan made to someone with weak credit, someone who quite possibly cannot repay it. Brokers were paid a commission for every loan they closed, so loans got closed, documentation optional.
Securitization is the bundling of thousands of loans into a bond: a tradable IOU whose interest payments come from all those homeowners' monthly payments. The banks that built the bundles sold them to investors worldwide and collected fees on every deal.
A rating agency is a firm that grades the safety of bonds, and the agencies were paid by the very banks issuing the bonds they graded. They stamped a remarkable share of these bundles AAA, the same grade as the safest governments on earth. The Financial Crisis Inquiry Commission, the official US inquiry into the crisis, later documented how those grades were produced and how fast they collapsed.
Leverage, as a noun, means doing business with borrowed money. The inquiry found the big investment banks operating at ratios around thirty to one, and in cases beyond it: thirty euros of assets standing on one euro of their own capital. Keep that number; we will do the arithmetic below.
A credit default swap is an insurance-like contract that pays out if a bond defaults, and you can buy one without owning the bond, the way you could, if someone let you, insure a stranger's house. So on top of the real mortgages sat a second, larger tower of side bets about the mortgages, much of it insured by firms that had not set aside money to actually pay.
Now look at the chain the way the myth of the financial expert taught you to look at any adviser: by asking who pays whom, for what.
| The link | What it sold | What it was paid for |
|---|---|---|
| The mortgage broker | The loan | Closing it, not seeing it repaid |
| The lender | Loans, passed on to Wall Street | Volume sold, not loans that performed |
| The investment bank | The bundles | Fees per deal |
| The rating agency | AAA grades | By the issuer of the bond being graded |
| The insurer | Default protection | Premiums today; losses were tomorrow |
Every link in the chain was paid on volume, not on truth. Nobody needed to be a villain. Each link just needed to pass the risk one step onward before it went bad, and the incentives paid them to hurry.
How it burst
In 2006, American house prices stopped rising, and belief number one died quietly. Defaults rose. The AAA bundles had been sold as diversified, but the thousands of loans inside them were all plugged into the same national housing market, now falling everywhere at once. The safety grade turned out to describe the stamp, not the contents.
Then the sequence tightened. In March 2008, Bear Stearns, one of the five big American investment banks, collapsed into a weekend rescue sale. In September 2008, Lehman Brothers failed outright, the largest bankruptcy in American history. Within days a major money market fund, the kind of parking place for cash that is never supposed to lose a cent, fell below its sacred one-dollar value, and lending between banks, the plumbing under everything, froze. Governments stepped in with taxpayer-funded bailouts on a scale without precedent, because the alternative appeared to be the failure of the payment system itself.
And here is the human ledger: pension funds and ordinary index investors around the world, people who had never heard of a credit default swap, watched roughly half the quoted value of their savings vanish inside a year. The strawberry picker lost a house he could never have kept. A nurse in Athens or Lyon, who had done everything right, opened her pension statement and found the machine had reached her anyway.
The arithmetic of leverage
Why did the giants fall so fast? Not because they suddenly became reckless in 2008. Because they had been thin all along, and leverage decides how much bad news you can survive. The arithmetic takes one minute.
A homeowner who buys a €200,000 house with a €20,000 deposit and a €180,000 mortgage is leveraged ten to one. If the price falls 10%, to €180,000, the deposit, their entire stake, is gone. If it falls 20%, to €160,000, they owe €20,000 more than the house is worth. A 20% fall did not double the 10% fall's damage; it pushed the owner underwater.
A bank at thirty to one holds €30 of assets for every €1 of its own capital. A fall of just over 3% in the value of what it holds erases the bank's capital entirely. Three percent. That is the entire cushion the biggest institutions in the world were standing on, per the official inquiry.
Who was destroyed, and who was not
Now the part the disaster documentaries skip. The market's quoted level, cut roughly in half, recovered and went on to new highs within about five to six years. So the people permanently destroyed by 2008 were, once again, the same two groups as in every mania in this series: the leveraged, whose arithmetic killed them long before any recovery could arrive, and the forced sellers, who had to turn quoted losses into real ones at the bottom because a job disappeared, an emergency fund did not exist, or panic won.
Something else was destroyed, though, and did not recover on the same schedule: trust. The conduct in this chapter is not a cynic's fantasy; it is documented, by Lewis in narrative and by the FCIC in official findings. When the retirement accounts article argued that you should understand your own money because the industry's incentives are not your incentives, this chapter is the evidence file it was pointing at.
The lesson
Every mania teaches that prices fall. 2008 teaches something sharper: complexity is not safety. Complexity is where risk hides. A simple thing, a house, a share of a business you understand, can be examined, questioned, valued. A bundle of a bundle of side bets on bundles cannot, and the AAA stamp on it was not information. It was permission to stop asking.
If nobody in the chain can explain who ends up holding the loss, everyone is holding it.
That question, who holds the loss, is one you can ask about anything you are ever offered: a fund, a scheme, a hot opportunity, a complicated product with a reassuring grade. If the seller cannot answer it in one plain sentence, you have learned everything you needed to know.
Every mania is the first article, at scale
This series began with the psychology of money and one claim: the biggest risk to your money is you, because behavior beats knowledge. Five manias later, you have seen the proof at every scale. Tulips, the South Sea, 1929, the dot-coms, 2008: the assets change, the centuries change, the human does not. Envy of the neighbor who is getting rich, the story that this time is different, the borrowed money, and then the reckoning that transfers wealth from the forced to the patient.
Which means the quiet, unglamorous equipment this blog has spent its whole run assembling is also crash equipment. An emergency fund, so that no disaster can force you to sell at the bottom. Investing only money you will not need for years. No debt staked against your future. Knowing what you own well enough to explain it in a sentence. None of that predicted 2008, and none of it prevented the fall. It did something better: it let some people walk through 2008 poorer on paper and intact in fact. The quoted value of their savings halved. Their lives did not.
Do this now
Do an ownership audit. This week, list every place your money lives: accounts, funds, pensions, shares, property. Next to each, write two things in plain words. First: what is this, actually? Second: if its quoted value halved next year, would anything force me to sell? Every line where you cannot write the first sentence is your reading list. Every line where the second answer is yes is your real risk, and it usually has nothing to do with the asset and everything to do with the cushion around your life. In 2008, that piece of paper was the whole difference between a bad year and a lost decade.
This series ends here, but the work begins: Why value investing, where the reading of actual businesses starts, one company at a time.