• Cover Story
• Table of
  Contents
• Contents By
  Topic
• Columns
  • Brownstein
  • Cook
  • Crook
  • Rauch
  • Stokes
  • Schneider
  • Taylor Jr.
• Regular
  Features
  • Hotline Extra
  • Inside Washington
  • Insiders Poll
  • K Street Corridor
  • People
  • The Week on the Hill

Malcolm Gladwell, a New Yorker science writer, recently had this to say about the financial markets' meltdown last fall: "The roots of Wall Street's crisis were not structural or cognitive so much as they were psychological." Too many people, apparently, got too cocky. And -- kaboom! -- the whole financial system collapsed!

Sorry. That won't do. No doubt, too many people did get too cocky, but there is nothing new about cockiness ("animal spirits") on Wall Street. No doubt, too, inadequate and misguided regulation played a part. No doubt, also, expansive monetary policy and affordable-housing mandates were factors. Indeed, many, many factors were factors.

But the core of the problem remains mysterious, even when you add up the standard explanations. The story most people think they know is that the Titanic (the financial system) steamed too fast (financiers got reckless) and hit an iceberg (bad subprime lending), so the ship sank. The problem with that story is that subprime mortgage lending was a big number, big enough to cause grief for investors in subprime paper. But it should not have been nearly big enough to bring the whole financial system to the brink of collapse.

George Santayana was vindicated: Those who had forgotten history had been condemned to repeat it.

Moreover, asset-backed securities that had nothing to do with subprime lending -- paper backed by student loans, auto loans, corporate debt, credit card debt, regular mortgages, and so on -- also seized up. On the merits, the crisis should have been sectoral, not systemic. It was as if the Titanic had missed the iceberg, or was only dented, and then sank anyway. Why?

Gary Gorton, a finance professor at Yale, thinks he knows. Unwittingly, the U.S. financial system recreated the panic-prone conditions that were the norm in this country throughout the 19th century and into the first third of the 20th.

Gorton (full disclosure: a personal friend) has been thinking about financial crashes for three decades. His doctoral dissertation was titled "Banking Panics and Business Cycles." From 1996 to 2004, as a consultant to American International Group, he helped develop the math behind the financial instruments (credit default swaps) that ultimately brought AIG to its knees. "I spent my entire career studying banking crises and then happened to live through one," he says.

In a series of recent scholarly papers (one of which he co-authored with a Yale colleague, Andrew Metrick), Gorton anatomizes the collapse in fascinating detail. My summary only hints at the richness of the originals, available online here.

Banking panics are a fixture of American history. At least seven major ones occurred between 1873 and 1914. Typically, they came at the peak of a business cycle, when straws in the wind hinted at recession. Although everyone called bank runs "panics," Gorton stresses that in fact they were rational responses to an information gap. Depositors knew that in a recession some banks would fail. But they had no way to know which banks would fail. So, as a precaution, they would withdraw their money.

When a bank run began, banks rushed to raise cash by calling loans and selling assets. Distress sales drove down asset prices, which cratered banks' balance sheets and strangled the flow of credit, which induced further cash hoarding, which crushed the money supply. The typical result was a severe recession.

What really needs explaining is not why recurrent financial panics happened, but why they stopped. In the so-called Quiet Period, 1934 through 2007, systemic bank runs seemed to become relics of an unmourned past. Why? Because for about four decades, banks' activities were restricted to heavily regulated ventures that were more or less guaranteed a profit -- and, even more important, because federal deposit insurance, which began in 1934, assured depositors that their savings were safe.

Financial innovation, however, could be delayed but not denied. Around the walled garden grew a forest of new competitors and products. Money-market funds and other investment vehicles took deposits without offering federal guarantees. In a process known as securitization, investment banks converted predictable streams of income, everything from mortgage payments to health club dues, into securities that investors bought eagerly. Derivatives -- securities based on other securities--arose to spread risk and hedge against volatility. In time, shadow banking, as the new institutions and instruments were collectively called, rivaled and even eclipsed old-fashioned commercial banks.

The firms and major financial players making all these trades needed to park cash where it would hold its value and earn some interest, yet be accessible on demand. In other words, they needed the equivalent of checking and savings accounts, the "demand deposits" that banks traditionally provide and that form the backbone of the money supply. But no insured depository could begin to cope with the trillions of dollars involved. And so shadow banking developed what amounted to its own depository system, a short-term securities market called the "sale and repurchase," or "repo," market. It is immense. Gorton figures its size at perhaps $12 trillion, but he says no one knows for sure.

"It's important to see that this is a banking system," Gorton says. But it is like a 19th-century banking system, because repo "deposits" are uninsured. Unable to rely on a federal guarantee, depositors who park their holdings there require that the borrower put up something of value as collateral.

Treasury bonds, because they are safe and liquid, are the ideal form of collateral, but there were nowhere near enough of them to meet the demand. So asset-backed securities -- those packages of safe-looking income from mortgages, auto loans, and all the rest -- were pressed into service as collateral. In time, the better grades of subprime mortgage-backed securities were mixed into the blend, and they, too, won acceptance as collateral.

All of these asset-backed securities were sorted and re-sorted, combined and recombined, sold and resold, until, as Gorton writes, "looking through to the underlying mortgages and modeling the different levels of structure was not possible." Users could not independently assess the value of mortgage-backed collateral any more than your grocer can independently assess the solvency of your bank before accepting your check.

You can see, perhaps, where this leads. Repo is a form of money because it acts as a store of value and financial actors rely on it to conduct transactions. But instead of being backed by a federal guarantee, it was backed by, among other things, subprime mortgages. In this way, without anyone paying much notice, subprime mortgage debt entered the money supply. As in the 19th century, the economy had become dependent upon a form of bank-issued money that was not federally guaranteed and that was not as stable as it appeared. Unlike in the 19th century, however, no one understood how vulnerable the system was to a panic.

Calamity then struck, as it had before. First, the unexpected decline in housing prices tanked the subprime market. Repo depositors knew that most collateral was sound, but they had no way to know if their own holdings were safe; so in 2007 they began what amounted to a run on the repo system, effectively withdrawing their money. To raise cash, repo depositories dumped assets, further depressing collateral values and starting a tailspin.

In September of last year, when the failure of Lehman Brothers, the mighty investment bank, convinced investors that no one was safe, the crisis turned into a meltdown. As the repo market "virtually disappeared" (in Gorton's phrase), the money supply crashed and the economy began to suffocate. And George Santayana was vindicated: Those who had forgotten history had been condemned to repeat it.

Whether Gorton has the story right is not something I am qualified to judge. The meltdown was a seminal event, one that no one, including Gorton, foresaw and one that will take years to understand. Before the meltdown, he says, "bank regulation began to not make sense to me. But I never thought we were going to see an event this big. I'm going to spend basically the rest of my career trying to sort this out."

Assuming he is right, however, higher capital requirements for big banks, consumer-protection laws, and other commonly mooted reforms largely miss the point. The main question, Gorton says, is not how to strengthen conventional banking but how to stabilize shadow banking, which means, first and foremost, recognizing that shadow banking is banking and then asking whether it needs federal insurance, government-backed collateral, reserve requirements, regular audits, and other regulatory stabilizers.

"These are the real issues," Gorton says. "It doesn't seem to me that we're having that kind of discussion."