Will the Geitner plan work? Understanding Krugman's objection

It depends on what you mean by work. As Krugman noted on his blog today, the official administration line is that the Geitner put is just one piece of a larger strategy. A problem, according to Krugman, is that nobody knows what the other pieces are.

But let's back up for a second. What does the Geitner plan do, why does Krugman think that it won't work?

Geitner's plan, basically, is to create investment groups with government backing to purchase toxic assets from distressed banks. The downside risk of those investments will be guaranteed by the government, which means that buyers will have an incentive to overpay compared to current market prices. The idea is that this will get the toxic assets off of the banks' books at a price which will restore the banks to liquidity.

Krugman's critique of the plan is a little difficult to suss out. In his column this week, he wrote:

But the real problem with this plan is that it won’t work. Yes, troubled assets may be somewhat undervalued. But the fact is that financial executives literally bet their banks on the belief that there was no housing bubble, and the related belief that unprecedented levels of household debt were no problem. They lost that bet. And no amount of financial hocus-pocus — for that is what the Geithner plan amounts to — will change that fact. |Krugman|

On his blog, Krugman gives a more complete diagnosis:

Start with the question: how do banks fail? A bank, broadly defined, is any institution that borrows short and lends long. Like any leveraged investor, a bank can fail if it has made bad investments — if the value of its assets falls below the value of its liabilities, bye bye bank.

But banks can also fail even if they haven’t been bad investors: if, for some reason, many of those they’ve borrowed from (e.g., but not only, depositors) demand their money back at once, the bank can be forced to sell assets at fire sale prices, so that assets that would have been worth more than liabilities in normal conditions end up not being enough to cover the bank’s debts. And this opens up the possibility of a self-fulfilling panic: people may demand their money back, not because they think the bank has made bad investments, but simply because they think other people will demand their money back.

Bank runs can be contagious; partly that’s for psychological reasons, partly because banks tend to invest in similar assets, so one bank’s fire sale depresses another bank’s net worth.

So now we have a bank crisis. Is it the result of fundamentally bad investment, or is it because of a self-fulfilling panic?

If you think it’s just a panic, then the government can pull a magic trick: by stepping in to buy the assets banks are selling, it can make banks look solvent again, and end the run. Yippee! And sometimes that really does work.

But if you think that the banks really, really have made lousy investments, this won’t work at all; it will simply be a waste of taxpayer money. To keep the banks operating, you need to provide a real backstop — you need to guarantee their debts, and seize ownership of those banks that don’t have enough assets to cover their debts; that’s the Swedish solution, it’s what we eventually did with our own S&Ls. |Krugman|

One of Krugman's points is that the housing bubble really was a bubble and that this means that the banks investments in mortgage securities really were bad investments. And what this means is that government can't arrest the fall in the price of those securities merely by buying some of them. They can't, in other words, put a stop to the panic because there isn't a panic to put a stop to.

It follows from this that the banks are, as things stand at the moment, insolvent. This is because the asset side of their balance sheets are heavily loaded with these mortgage securities, and those securities have very little value. The banks have managed to stumble along by pretending that the securities are worth more than they are, but nobody really believes that. If those assets were fairly valued, then the banks would obviously be insolvent and would fail.

As an aside, this is the context in which to understand the Congressional GOP's push to return to "mark to market" accounting rules. The idea is to change the laws to let the banks put mortgage securities on the balance sheet for the value they expect to sell them for in the future, rather than for the value that they could earn today. This would make the balance sheets look healthy, sure, but is ostrich accounting of the worst sort. Pretending that these securities are worth something would allow current managers to justify their bonuses, but it wouldn't solve the underlying problem.

Returning to Geitner's plan, the idea is to pull the troubled assets off the books. Why won't that work? This segment from last night's NewsHour may shed some light:



Krugman's argument, I think, is this. Banks, the troubled ones at any rate, have more bad assets on their books than the Geitner plan will purchase. If the plan is to work, it must do so by establishing a price floor for those assets which is sufficient to ensure solvency. But, given the incentives of the Geitner plan, the price paid for assets is likely to be higher than the market would otherwise bear. At the same time, the plan will give us information about what those assets are worth. But once we know what those assets are worth, then banks will be forced to honestly report their value on their balance sheets. The banks are currently reporting those assets as being worth something like sixty cents on the dollar, even though the market price is more likely to be something like thirty cents. But then bob's your uncle, because at those prices many banks will turn out to be insolvent.

Which brings us full circle to the seemingly unknown other parts of the plan. The total losses on mortgage securities in the U.S. is estimated to be something like $4 trillion. The Geitner plan will cover something like $1 trillion of those losses. The Fed's announcement of quantitative easing means that they'll buy up another trillion or so of mortgage paper. Some losses have already been addressed by capital infusions. But all of that still leaves something close to $2 trillion in losses for the banks to absorb. Can they do it? I guess we'll find out.

Adding: Yes, I know that I spelled Geithner's name wrong.