Banks and financial companies tend to die in one of two ways: from cancer or from a heart attack.

When a firm has cancer, its assets -- loans, securities and other investments -- either aren't repaid or turn out to be worth less than expected. Losses erode capital, leading to insolvency. To help avoid this, banks should hold more capital than regulations now require.

More capital, though, does little to avert a heart attack: a frenzied run on a bank that can spread and shatter the entire financial system.

Economists and academics have understood how to contain traditional bank runs for decades. Now a handful, including Morgan Ricks, Enrico Perotti and Gary Gorton, are exploring what measures might work in a financial landscape where changes have outpaced regulation.

It is no accident that the U.S. financial system experienced no systemic blowups from the mid-1930s through 2007. In a stroke, the creation of the Federal Deposit Insurance Corp. in 1934 blunted the incentive that depositors had to race to the bank for their money at the first whiff of distress.

But there was a run on the U.S. financial system in 2008. This time, though, it was investors and traders sitting at desks in front of computer screens who pulled their money. Had the Federal Reserve not injected hundreds of billions of dollars of liquidity into the financial system, something akin to the Great Depression might have occurred.

These investors and traders work in the so-called shadow banking system made up of securities firms, money-market mutual funds, hedge funds and other financial intermediaries. The firms they work for and do business with don't fund themselves with deposits. Instead, shadow banks rely on short-term funding in the form of IOUs that mature in anywhere from a day to several months. These IOUs resemble deposits in many ways.

Deposits can cause all sorts of trouble for a bank because they can be yanked so fast. But banks usually make loans and investments that mature in weeks, months or years. Borrowing short-term at low interest rates and lending longer-term at higher rates, and managing the inherent risk, is the essence of banking.

The IOUs in shadow banking -- which have found their way into the regular banking system, too -- have the same problems. The absence of something like deposit insurance means investors have a motive to reclaim their money as fast as possible if trouble is afoot. When that happens, a firm has to sell assets to come up with the necessary cash. If more than one company tries to sell assets at the same time, prices fall, and even more assets must be sold. A death spiral ensues.

This was what happened to Bear Stearns Cos., which went bust in less than a week once its funding dried up. The same was true of Lehman Brothers Holdings Inc., which had about $600 billion in assets and $200 billion in short-term funding.

Yet there may be ways to rein in the over-dependence of the financial system on unstable short-term IOUs and lessen the odds of another catastrophic run.

One idea, proposed by Ricks, a former U.S. Treasury official now at Vanderbilt University, is to place legal limits on which firms could use short-term IOUs. Banks, for example, are the only institutions allowed to fund themselves with deposits. They must pay for that privilege by purchasing insurance. In turn, there are constraints on how banks invest that money -- they can make loans and buy U.S. Treasury securities or high-grade bonds. Stocks, junk bonds, derivatives and other risky investments are generally off-limits (although compliance has been lax, which is why we need the Volcker rule that bars banks from reckless wagering with depositors' money).

As Ricks sees it, restricting who can use short-term funding will force businesses that want to speculate to rely more on equity (capital) and longer-term debt. Imagine if Lehman Brothers had used debt that matured on average in 180 days, rather than about a week: Only a bit more than a half-percent of its IOU funding could have disappeared in a day, a serious brake on panicked withdrawals.

Ricks says the equivalent of deposit insurance for IOUs might also enhance stability. It would have the added virtue of recapturing at least part of the implicit subsidy the biggest banks and securities firms get because their creditors believe they will be propped up by the government in a crisis.

But IOU insurance would be an expansion of the federal safety net, a political non-starter. There also is the moral-hazard dilemma: Insurance might cause a company to take foolish risks.

Perotti, a professor of international finance at the University of Amsterdam, favors a tax on short-term funding. That too would encourage firms to use more stable, long-term debt. The absence of an insurance guarantee also would ensure that investors remain exposed to the discipline of market risk.

Gorton, a professor at Yale University, takes a different approach: The key to stability is developing sound collateral to pledge against the IOUs. During the financial crisis, confidence in the quality of assets used to secure short-term funding -- remember all that AAA-rated junk? -- collapsed.

It took the U.S. almost 70 years, from the end of the Civil War to the Great Depression, to figure out how to contain old-fashioned bank runs. There's no reason to wait that long to head off another run on shadow banking's IOUs.

(James Greiff is a member of the Bloomberg View editorial board. Follow him on Twitter.)