The River That Was Not on the Map
How weak mortgages became highly rated securities, and why the losses spread far beyond the homes on which the loans were made.
United States, 2006. A mortgage lender approves a loan that the borrower may struggle to repay once the initial interest rate rises. The lender expects to sell it. The buyer expects to package it with thousands of other mortgages. Investors further down the chain will see a security, a credit rating and a promised stream of payments. Each step appears to move the risk away from the institution that took the previous one.
It also makes the original loan harder to see.
This was not the fate of every mortgage written in those years. Many borrowers could afford their homes, many loans were sound and many investors understood the risks they were taking. But lending standards weakened in parts of the market as house prices rose and demand for mortgage securities grew. Some borrowers were offered loans they could afford only while introductory payments remained low or house prices kept climbing. Documentation could be thin. Brokers and lenders were paid to originate loans, while the consequences of poor lending might appear much later, in someone else’s portfolio.
At the other end of the chain, a highly rated security could look reassuring. Its expected losses had been estimated. The underlying mortgages were spread across borrowers and regions. Junior investors would absorb losses before senior investors were touched.
All of that mattered. None of it answered the question that would decide the outcome: what happens when many loans depend on the same rising housing market, and that market begins to fall?
The river was already forming. The map did not show where it could run.
FROM ONE MORTGAGE TO A SECURITY
Begin with a borrower making monthly mortgage payments. The lender can hold the loan and collect those payments over time, bearing the loss if the borrower defaults. Or it can sell the loan to another institution.
An investment bank or other arranger can assemble thousands of mortgages into a residential mortgage-backed security. Investors who buy the security receive cash generated by the borrowers’ payments, after fees and according to the contract’s rules. The mortgages do not become safer simply because they have been packaged. Their cash flows have been reorganised and sold to different people.
The arrangement commonly divides investors into tranches. A junior tranche takes losses first and is paid more for accepting that position. A senior tranche has priority and suffers losses only after the protection beneath it has been exhausted. If the pool performs as expected, that senior claim can appear very safe.
Some collateralised debt obligations, or CDOs, add another layer. Rather than owning whole mortgages directly, they may contain tranches of mortgage-backed securities. Their own cash flows are then divided into new tranches. Other CDOs use credit default swaps to gain exposure to mortgage securities without purchasing the bonds themselves.
This distinction matters. A mortgage, a mortgage-backed security and a CDO are different claims. Yet they can remain exposed to the same underlying borrowers and the same housing market. Creating another tranche changes who takes the first loss. It does not create new households able to make their payments.
Securitisation had legitimate uses. It could connect borrowers with a wider pool of capital and distribute exposure among investors willing to bear it. The weakness emerged when the demand for securities helped sustain the production of poor loans, and when the distance from borrower to final investor made loan quality harder to assess and easier to overlook.
WHAT A TRIPLE-A RATING MEANT
A triple-A rating signalled the rating agency’s view that a security had very low credit risk under its assumptions. It was not a promise that the price could not fall, that the security could always be sold or that its owner could finance it through a crisis.
That distinction became easy to miss. Ratings helped determine which securities certain institutions could buy and how much capital some regulated firms had to hold against them. A high rating therefore affected demand and financing, as well as an investor’s perception of safety.
To assess a senior mortgage claim, an analyst needed estimates of how many borrowers might default, how much lenders would recover after a foreclosure and whether losses would strike many mortgages at the same time. The last issue is often called default correlation. If problems remain isolated, losses may be absorbed by junior tranches. If a common shock reaches much of the pool, that protection can be consumed far faster.
Historical housing data offered some comfort. Local economies had often moved differently. A weak market in one city could be offset by a stronger one elsewhere. But a national expansion of easy credit could connect housing markets that previously seemed separate. When house prices stopped rising, refinancing became harder for borrowers across many regions. Falling prices and defaults could then reinforce one another.
The ratings problem cannot be reduced to a single correlation number. The estimates depended on the quality of the loans, future house prices, recovery values and the strength of the protection built into each structure. Agencies also faced pressure from the firms that paid them to rate new deals. Subsequent examinations found weaknesses in their methods, documentation and monitoring of securities after the initial rating.
The model was part of the failure. So were the decisions about what data to use, what warning signs to investigate and how much confidence to place in the result.
THE WARNING INSIDE THE LOANS
By 2006, the deterioration in some mortgage lending was visible to people prepared to examine individual loans and lending practices. Some mortgages required little proof of income. Some offered low initial payments that would later reset. Some borrowers put little of their own money into the purchase and depended on rising prices to refinance or sell.
These features did not make every loan certain to default. Together, they made the pool more vulnerable to a reversal in house prices and credit availability.
The link between them is important. When a house is worth more than the loan, a borrower in difficulty may be able to sell or refinance. If the house falls below the debt, both options become harder. A borrower with a payment about to rise may then have few ways out. Foreclosures can put additional houses on the market, weakening prices for other borrowers.
The system could therefore generate losses across regions even when earlier local housing histories suggested diversification.
Some investors recognised this. They studied the loans, examined the terms and bought protection against mortgage securities they believed were mispriced. Their work showed that the information was not wholly inaccessible. It also showed why a rating could never replace an understanding of the assets and contracts beneath it.
Being right was still difficult to turn into a trade. Protection had a cost. Losses could take time to appear. A trader who questioned the prevailing price needed enough capital and patience to stay with the position until the evidence reached the market.
That is a different problem from calculating which security should eventually fail. It is the problem of surviving until the market recognises it.
WHEN THE LOSS LEFT THE MORTGAGE POOL
US house prices began to weaken in 2006 and 2007, with the timing and severity differing across measures and regions. Mortgage delinquencies rose. Securities tied to the weaker loans lost value. Ratings were cut, sometimes sharply.
The damage did not stay inside mortgage pools.
Banks and investors held mortgage securities directly, through CDOs or through obligations to other firms. Some had borrowed short term to hold assets that would pay over years. That arrangement worked while lenders remained willing to renew the borrowing and accept the assets as collateral.
As confidence in the securities fell, lenders demanded more collateral or refused to roll over financing. Firms needing cash sold what they could. Those sales pushed prices down and made similar holdings look less valuable on other balance sheets. Falling prices then prompted more demands for cash.
A mortgage default had become one loss in a much larger chain. Funding pressure could force the sale of assets that had nothing to do with a particular borrower. Institutions that had transferred one exposure discovered that they still carried another through guarantees, derivatives, investment holdings or their dependence on counterparties.
The failure of Lehman Brothers in September 2008 intensified a crisis already under way. Confidence in short-term lending and in the ability of large firms to honour obligations deteriorated. Governments and central banks responded on an extraordinary scale to keep payments and credit flowing.
The crisis was neither a single bad mortgage multiplied by a clever equation nor a surprise that appeared from nowhere in 2008. It was a sequence of decisions and feedback effects. Weak lending, opaque structures, misplaced confidence in ratings, concentrated exposures and fragile funding allowed a housing downturn to become a systemic emergency.
The model said the river was not there. The river did not check the model.
As a metaphor, the line captures a real failure. Literally, models did not rule out every national housing decline. The problem was that too many decisions relied on estimates that made the joint losses and the consequences of a downturn appear manageable. The institutions holding those exposures often lacked the capital, information or funding to absorb what followed.
THE LESSON FROM LTCM
Article 5 ended with a fund that could identify plausible long-term value but could not finance its positions long enough for that value to be realised. The 2008 crisis involved different instruments and many more institutions, yet the funding lesson carried through.
A security may have a calculated value, while its holder has an immediate need for cash. A hedge may reduce one measured exposure, while leaving a firm dependent on a counterparty. A portfolio may look diversified by name or region, while its holdings all rely on the same source of credit and the same assumption about house prices.
The similarity does not make the two crises identical. LTCM was one highly leveraged fund whose failure threatened a disorderly unwind. In 2008, problems in lending, securitisation and funding spread through much of the financial system and into the wider economy. The scale and social cost were profoundly different.
What links them is the need to ask how a position survives the path between today’s estimate and tomorrow’s outcome. Price, collateral, liquidity and time cannot be separated when a firm must meet obligations along the way.
WHAT A TREND FOLLOWER COULD SEE
A trend follower did not need to identify a weak mortgage pool in 2006 to respond to some of the market moves that later emerged. Equity markets, government bonds, currencies and commodities all developed substantial moves at different stages of the crisis. Rules that follow sustained price movement could participate in some of them as signals formed.
The timing matters. Energy and other commodities rose before falling sharply. Equity markets had rallies within their declines. Signals could arrive late, reverse or produce losses. An individual programme’s result depended on its markets, trading speed, position sizes and ability to execute. There was no universal crisis portfolio that every trend follower held, and no guarantee of profit.
Research on diversified time series momentum has documented strong returns during some extreme market episodes, including the 2008 crisis. That finding is evidence of a possible payoff pattern across the markets and rules studied. It is not a promise that any particular trader captured the full move or that the approach protects against every crisis.
For an Outlier Hunter, the practical point is structural. Begin with modest exposure across many liquid markets. Accept small losses when signals fail. Keep predefined exits and avoid depending on one forecast of what the next shock must look like. When an unusual movement persists, allow the profitable position room to develop.
This framework has its own limits. Sudden reversals can hurt. Crowded positions and thin markets can make exits more expensive. A strategy that cannot be financed or executed when conditions change faces the same survival question as any other.
The history of the mortgage crisis offers no reason to abandon measurement. It offers a reason to use it with care. A rating, a correlation estimate or a historical test can reveal something useful. The institution still has to withstand what those measurements miss.
The map was useful in the territory it described.
Then the river crossed the blank space.
Next: Article 7, The Physicists Who Stopped Predicting
Richard Brennan writes on systematic trading, complex adaptive markets, and the philosophical foundations of trend following at atstradingsolutions.com. His books include The Fractals of Finance, Complex Adaptive Markets, Carved by Impossibility and The Aussie Turtles Trend Following Guide.
Want to explore why structure exists at all?
Carved by Impossibility: What Remains When Everything Else Is Eliminated
The book explores the architecture of constraint, emergence, and reality itself, and what it means for how we understand markets, life, and the universe.
Available now on Amazon in paperback, hardcover, and Kindle.
Want the theoretical foundation for why trend following works?
The Fractals of Finance: Determinism, Adaptation and the Geometry of Markets
The book explores the full architecture of feedback, fat tails, and fractal structure in financial markets, and what it means for how we trade, invest, and understand risk.
Available now on Amazon in paperback, hardcover, and Kindle.
Want a practical field manual for trading trends and capturing outliers?
The Aussie Turtles Trend Following Guide: A Field Manual for Hunting Outliers adapts the timeless principles of the original Turtle traders into a systematic, rules-based approach for modern markets. Co-authored with Adam Havryliv.
Available now on Amazon in paperback, hardcover, and Kindle.