But succeed or fail, Geithner's plan illuminates a fascinating irony. "Leverage" -- borrowing -- helped create this mess. Now it's expected to get us out. How can this be? It's not as crazy as it sounds. Start with the basics on how leverage affects investment returns.

Suppose you bought a stock or bond for $100 in cash. If the price rises to $110, you make 10 percent. Not bad. Now, assume that you borrowed $90 of the purchase price at a 5 percent interest rate. Over a year, the stock or bond still increases to $110, but now you've made more than 50 percent. You pay $4.50 in interest and pocket a $5.50 gain on your $10 investment. Note, however, that if the price fell to $95, you'd be virtually wiped out ($4.50 in interest paid plus $5 lost on the security).

Economist John Geanakoplos of Yale University argues that the economy regularly experiences "leverage cycles." When credit is easy, down payment terms are loose. Investors or homeowners can borrow much of the purchase price of houses and securities. Prices of assets (stocks, bonds, real estate) rise, often to artificial levels because investment returns are so attractive. But when credit tightens -- government policy shifts or lenders get nervous -- the process reverses. Prices crash. Leveraged investors sell to repay loans. New borrowers face stiff down payment terms.

To Geanakoplos, we're suffering the harshest leverage cycle since World War II. Three years ago, he says, homebuyers could put down 5 percent or less. Now they've got to advance 20 percent or more. Hedge funds, private equity funds and investment banks could often borrow 90 percent of security purchases; now borrowing can be 10 percent or less. "Deleveraging" has caused prices to plunge to lows that may be as unrealistic as previous highs.

Grasping this, you can understand the idea behind Geithner's hedge fund. It is to inject more leverage into the economy -- not to previous giddy levels but enough to reverse the panic-driven price collapse. Details remain unsettled, but the plan would allow 6-1 leverage ratios in some cases. Here's an example. Private investors put up $5; the Treasury matches that with another $5. This equity investment could then be expanded by $60 of government-guaranteed loans. The entire $70 could be used to buy assets from banks.

Sounds simple. In practice, it won't be. Given all the deleveraging -- a record 15 percent of hedge funds closed last year -- the market prices of many securities have been driven well below prices that seem justified by long-term cash flows. Geanakoplos mentions one mortgage bond whose market value has dropped by roughly 40 percent even though all promised payments have been made and, based on the performance of the underlying mortgage borrowers, seem likely to continue.

If banks sold this and similar credits at today's market prices, they would have to record huge losses. ("Banks Face Big Writedowns in Toxic Asset Plan," headlined the Financial Times.) Their capital would be depleted, and they'd have to raise more or request more from the government. Presumably, the government-supplied leverage would enable investors to pay higher prices. After all, that's the purpose. Still, whether sellers and buyers ultimately agree on prices is unclear.

If they can't, Geithner's hedge fund will remain puny. Cautious banks will continue to constrict credit. But success also poses problems. Money managers talk about making huge annual returns of 20 percent or more from a scheme in which government puts up most of the funds and takes most of the risk. A political backlash might squash the project before it starts. Geithner treads a narrow line between financial paralysis and populist resentment.