Update, September 22, 2011:Yesterday, Standard and Poor’s downgraded the credit ratings for seven Italian banks, shortly after downgrading Italy’s sovereign credit rating. As Jesse Eisinger explains in this column, Italian financial weakness could have serious consequences for the U.S. banking system.

In early 2010, top officials at the Federal Reserve began to wonder: How would United States banks hold up through the European debt crisis? Investors were fleeing Greece and Ireland and starting to get nervous about Portugal and Spain, spreading contagion.

The conclusion from the stress tests that resulted was heartening to supervisors at the regulator, according to a person who was directly involved in the exercise: American banks didn't have too much exposure to Portugal and Spain, so the contagion would not be a problem.

Unless it hit Italy.

"At the time, the results made us a bit relieved; our focus was on Ireland and Greece," said this person, who spoke on condition of anonymity because the Fed has a policy of not discussing supervisory actions. "But if Italy goes, God help us all."

American banks had not only a small exposure to Italian government bonds, but a larger one to Italian banks and companies. If the European debt crisis spread to Italy, it could cause another global financial catastrophe. Only this time, global regulators might have fewer weapons to combat it. The Fed declined to comment on its analysis.

And that is why last week was so terrifying, scarier than either the stock market drop or the Standard & Poor's downgrade of the United States credit rating: Debtholders had abandoned Spain and Italy.

At one point on Friday, Italian bonds were trading at more than 400 basis points higher than Germany's, a signal of panic. Italy has a huge debt load. If investors began to focus on that, it wasn't clear what might stop the run.

The European Central Bank intervened this week, buying Spanish and Italian government bonds. On Monday, the panic eased in Europe, with Italian and Spanish interest rates falling. The French and Germans announced that the European Financial Stability Facility (clearly named by Dr. Seuss) would be able to buy those government bonds when it was up and running in late September.

By Wednesday, the fears were back, as French banks got hit especially hard. The problem is that Europe has tried repeatedly to fence off the problem, only to have it escape again to wreak havoc. Greece and Ireland have each been through several rounds of failed bailouts and extensions. If they are bankrupt, and not simply victims of investor panic, then someone, somewhere will have to take losses. And if Spain and Italy start to go down, those losses threaten the global economy.

European banks are on the front lines, vulnerable because they are more thinly capitalized than their American counterparts. Europe has conducted stress tests, just as the Fed has, but they haven't instilled confidence, in part because they didn't subject most sovereign debt holdings to any loss estimates.

The tests did require vast disclosures, however, so that investors and analysts could delve into the numbers and conduct their own analyses.

If European banks go down, what will happen to American banks? Investors and analysts seem unconcerned. American banks disclose some of their exposure to specific countries, but the information isn't up-to-date, and the figures depend on opaque estimates of how well hedged the banks are. Analysts differ on the amounts at risk.

According to a note from the research firm CLSA on July 13, Citigroup had $12.7 billion in Italian holdings, much of it government-related, while JPMorgan Chase had $12.2 billion. According to a note from BernsteinResearch, JPMorgan had "less than $20 billion" in exposure to Portugal, Ireland, Italy, Greece and Spain combined. But that was going in the wrong direction, up from "less than $15 billion" at the end of 2010, when one might expect the banks to be paring exposure.

These aren't large numbers, less than 1 percent of these gigantic banks' balance sheets. And banks wouldn't take 100 percent losses on their investments in the event of a default.

Unfortunately, we simply don't know whether the analysts are right. Neither the Fed nor the Securities and Exchange Commission has forced United States banks to make as detailed disclosures as the European stress tests did of its banks. So, it's a matter of having to trust the banks and the regulators.

Which brings us back to the exercise the Fed undertook last year. Two Fed officials ordered up the analysis: Daniel K. Tarullo, the board member who oversees matters of bank supervision; and Patrick Parkinson, the head of banking supervision, who is reported to have undergone a conversion from an Alan Greenspan antiregulation acolyte to a believer in strong oversight. Clinton D. Lively, a number-cruncher who recently left the New York Fed, was one of the officials who played a major role.

Disturbingly, before the financial crisis of 2008, the Fed, which is the most important bank regulator and is charged with keeping the banking system safe and sound, couldn't really do this kind of analysis, according to former Fed officials. It might have asked banks for the data on their exposures to specific countries, but it couldn't play out a chain of events very easily. What if the central bank wanted to know what would happen to United States banks if the euro fell 20 percent in a short period? The Fed didn't have the tools.

Now it does, thanks in part to the efforts of Mr. Lively and others. The assessment came with its own pitfalls. At time the Fed was gathering the data, a rumor started in markets that the Fed was worried about two Spanish banks. Some European banking supervisors became nervous about the Fed's efforts and voiced those concerns.

Unfortunately, the biggest problem is whether the data truly reflects all the risks. That continues to be a matter of intense debate within the central bank. Given the complexity of the trading arrangements within the global financial system, it's far from clear that the banks have a handle on their own exposures. As we learned in 2008, hedges that seemed solid on Monday disappear on Tuesday.

Nevertheless, it's good to know that the Fed isn't flying blind.