How the Bailout Was Supposed to Have Worked
The most astonishing thing about the spectacular failure of the Paulson bailout plan has been the the peculiar inability of officials to explain how it was supposed to work. The vacuum, abhorred by nature and the blogosphere alike, has given rise to all manner of conjectures, the most popular of which being that the crux of the plan is to re-capitalize banks by drastically overpaying for more-or-less worthless assets.

Inasmuch as it appears that Treasury and Federal Reserve officials are too busy at the moment to articulate a better rationale for the plan, I'll volunteer to sketch what I think is really going on.

The Goal

The goal of the plan, it seems to me, is to facilitate the re-capitalization of the banking sector with private money. That is, to help re-capitalize the banks without (a) large government purchases of bank equity leading to (b) significant direct government control of the day-to-day operations of the financial system. Now, this is similar to what I'll call "the Krugman theory" of the bailout, but not identical. In the Krugman version, the bank's new capital comes from the profits banks will post when they sell securities to the Treasury at above-book-value. This means that, to have a material effect on bank's capitalization, the Treasury needs to pay a lot over book value. In my version -- which has the advantage of being consistent with the statements Paulson and Bernanke made in their Congressional testimony -- the recapitalization comes from new private investment in the banking sector. What new private investment? To get at that, we need to ask what's keeping new private investment out under current circumstances.

The Obstacle: Or, How Banks Are Different From Everybody Else

If you've been reading about the financial crisis lately, you've probably run across the idea that businesses can fail in one of two ways: illiquidity, meaning that one day they simply don't have the cash to make payroll or meet some other payment deadline; or insolvency, meaning that the firm's liabilities exceed its assets, so it has a net value of zero or less.

This is false.

In fact, illiquidity is the only way a business can fail. An insolvent business can carry on just fine, and its employees, creditors and shareholders can all prosper, provided that it can remain liquid. Don't believe me? Check out Exhibit A, or, more specifically, Exhibit AMZN.

Amazon went public in 1997. For about three years, it managed to keep its book value in positive territory by raising new equity capital (i.e. selling more stock) to cover its operating losses. Then, in the second quarter of 2000, Amazon's book value went negative -- minus $278 million as of June 30, 2000. By the end of 2000, Amazon's book value was minus $967 million. Through 2001 - 2, Amazon held steady at around minus $1.4 billion. By 2003, it began to recover, increasing its net assets to minus $1.0 billion.

Yet, despite having been insolvent for three and a half years, Amazon was nowhere near out of business. And the market capitalization of the company (i.e. the price of its stock multiplied by the number of shares) stood at about $21 billion on December 31, 2003. Why were investors willing to pay $21 billion for a company whose own balance sheet said was worth less -- a billion dollars less -- than zero?

It's actually not that surprising. As accountants put it, a balance sheet is a snapshot of a firm's current condition. It doesn't tell you anything about the business's future. And the future -- specifically a company's future earnings -- account for more of most companies' values than the net assets currently on their books. The average publicly traded U.S. corporation is valued at more than double its book value (about 2.4 times book) and many companies are valued much more. Starbucks, for example, has a book value of $2.5 billion, but a market capitalization of $10.8 billion. Apple has a book value of $20 billion, and a market capitalization of $98 billion.

So when investors are willing to pay $21 billion for a company with a book value of minus $1 billion, they are saying, in effect, that they expect the company's future earnings to cover both the $1 billion deficit and to generate dividends (and/or stock buy-backs) worth another $21 billion. (And it looks like they made a good bet; Amazon today as a positive book value of $2.2 billion, and its market capitalization has risen to $29 billion.)

A company that is insolvent today may be solvent tomorrow, as long as it has sufficient cash to keep paying the bills.

Here's the thing: banks play by a different set of rules. In particular, banks must meet minimum capitalization standards set by regulatory bodies. (Aside: this is a good thing. The point is not that minimum capital regulation is bad, but that it is different, in ways relevant to understanding the bailout plan.) If Amazon had been a bank, it would have been seized by regulators back in 2000. Its shareholders would have lost all of their investments, and Jeff Bezos would have been fired. Fun as that may sound -- and who hasn't wanted to fire Jeff Bezos at one time or another? -- subsequent history has shown that it would have been unnecessary. And the possibility, I believe, is a serious obstacle to private re-capitalization of the financial sector today.

Consider the situation from the point of view of a potential investor in some bank. Normally, the dominant consideration would be the bank's long-term earnings potential. But it turns out that the bank has, on its books, some assets which (a) can't be valued at market prices because there currently is no market for them and (b) which are therefore valued based on statistical models tied in turn to very touchy, volatile market indexes (such as indexes of spreads on credit default swaps). Spikes and valleys in those indexes can trigger paper losses which could, in turn, cause the institution to be seized by regulators. At which point, our investment is simply gone. Poof.

The point is this: even if investors are patient and steel-nerved and prepared to wait out a long "down" period -- like investors in Amazon -- the regulators won't let them. So even a business that is in perfectly fine shape for the long haul can't possibly raise private capital if there's a significant risk that the investors could lose everything simply because the credit default swap market has a bad day.

The Fix

Hence the Paulson plan: the Federal government buys up a lot of the volatile assets, possibly in connection with individually negotiated re-capitalization plans with specific institutions, possibly more broadly. Some of the sales might be above current book value, but that's largely irrelevant -- they could also take place at or below book value, as long as the sale were part of a transaction that injects sufficient private capital into the selling institution.

The Federal government can hold these volatile assets safely because it doesn't care what their mark-to-market (or, more accurately, mark-to-model) value is on any particular day. The Treasury doesn't have to meet margin calls, doesn't have to post minimum regulatory capital, and doesn't have to worry about losing access to short-term financing. (Investors are lining up around the block to lend to the U.S. Treasury.)

The end result: in the long run, Treasury posts something between a small loss and a small gain on the assets it buys. Private investors return to the business of buying weak banks, turning them into less-weak banks. Terror in the inter-bank lending market recedes, the TED spread becomes less preposterous, and your employer can get the bank credit it needs to honor your paycheck. The big losers would be current shareholders in banks, who would lose most of their holdings to the new investors. Which is as it should be. New investors will likely insist on replacing top management, which is also as it should be.

The Train Wreck

That, I think, is more or less what Paulson and Bernanke had in mind. The question is why they didn't just say so. My guess: they think they did. I suspect that they are so immersed in this issue that they have come to believe that the chain of reasoning that I've laid out is perfectly obvious. So they employed short-hand, saying that they wanted to get the questionable assets "off the balance sheets" of financial firms to help them re-capitalize. Which sounds like saying that the asset purchases themselves were going to somehow re-capitalize the banks, thus bailing out current shareholders and management.

Hence the train wreck on Monday.