Good question. Here's a roundabout answer.
In some sense, all of history's progress from lives that were nasty, brutish and short to today's splendiferous buffet of iPhones, nine-month courses of physical therapy, and year-round fresh broccoli can be summed up in three words: gains from trade. We live better than a tribe of chimpanzees roaming through the primordial forest because we specialize and then exchange the fruits of our skills with each other. Trade, as the ecoomists say, increases the size of the economic pie to be divided between us.
But trade introduces an element of uncertainty into our lives above and beyond the possibility that we will be eaten by something bigger than ourselves, or starve to death when the rains fail. We still have to worry about those uncertainties, although the monsters now most likely to hasten our demise have four wheels and cost entirely too much to service. But now we also have to worry about our trading partners. In an advanced economy like the United States, that means millions of other people who are somehow involved in either making the things you buy, or buying the things you make. We spend more and more of our energy trying to guess what is going on in their pointy little heads. Since we haven't even met 99.9% of them, these guesses are necessarily somewhat imperfect.
Now, to finance. Finance is, in most essentially, the way that we distribute gains from trade across space and time. Money, by giving us a universal unit of account and medium of exchange permits us to make simple trades within large networks--it takes care of the tedious yet complicated business of swapping barter commodities around between myriad players until everyone has what they (think they) want. Money has somehow, mysteriously, become the world's most sophisticated swap-meeter.
Credit allows us to time-shift those trades. I give you something you want now in exchange for something I want later. These days, that thing I want later is almost always money.
In modern America, financial institutions stand between the players and most of these transactions.
But remember, we're now not only guessing about droughts and hurricanes and monster trucks; we're also guessing about the ever-shifting desires of all the people we trade with. If I lend you money to build widgets, I need to guess how great the demand for widgets will be, what other companies might try to beat you at your own game by making cheaper and better widgets, whether the government might decide it's in the public interest to outlaw widgets, whether you're likely to be the sort of fellow who runs his widget factory into the ground and runs off to Baja with his secretary, and so forth. I also have to make guesses about the future value of money. Will the government print to much of it? How much of the stuff I actually want will this money buy when I get it?
As you probably notice when you look at your TD Waterhouse statement, sometimes we get those guesses very wrong. When too many of us guess wrong at the same time, and it turns out that America doesn't actually need a Starbucks on every corner and seventeen varieties of social networking site, we get recessions. So add to the list of things you worry about the possibility that you, and everyone around you, have guessed wrong about your future desires.
If you have ever known someone who guessed drastically wrong about their desire to spend the rest of their life with their spouse, you know that when excessive optimisim crashes, it usually overshoots on the downside. And financial institutions, which are the collective repository of all of our guesses about the future, are often the locus of the economic equivalent of a nasty, nasty divorce. We are currently enduring the Alec Baldwin and Kim Basinger of economic corrections.
All this is very interesting, I hear you cry, but what does that mean for Citibank and Bank of America?
Well, when credit markets contract, the time horizon of our trading also shrinks. We start taking economic activity like Bill W. said--one day at a time. Credit markets are already contracting because people have realized that they are not nearly as good at predicting the future as they thought they were, and had therefore better neither a borrower nor a lender be.
A major bank failure accelerates this process. It's the difference between rolling slowly into your garage, and hurtling into it with the pedal to the metal.
First, their credit disappears from the market, which shrinks the economic pie by making it more difficult to trade goods and services between our current and future selves. The economic pie shrinks.
Second, the shrinkage of the current economic pie changes peoples' estimation of the future. Much of economic forecasting is, after all, trend extrapolation. To make matters worse, we are basically hard-wired to over-weight recent events when predicting what will happen next.
Third, the changed expectations shrink even further the amount of future trade that people are willing to do between current and future selves. No one wants to defer consumption now and lend some business the money on the wan hope that Snozzleberry soda is the Next Big Thing. The economic pie shrinks further.
This is all somewhat airy-fairy; perhaps you want to know exactly what will happen if Citibank and America will fail. Will CDS markets blow up? Insurance companies in receivership? Bank runs across the land?
But as the Lehman bankruptcy illustrates, we have no idea exactly what will happen. The Fed anticipated what might go wrong as best as it could, and actually did a pretty good job preventing those problems from getting out of hand. But they didn't foresee that the bankruptcy would cause the failure of a smallish money-market fund, or that this would, in turn, cause the entire commercial paper market to lock up. Where the credit contraction will occur is much harder to predict than the near-certainty that it will happen.