Anna Schwartz on why recapitalisation is wrong, from the WSJ

"Why are they 'toxic'?" Ms. Schwartz asks. "They're toxic because you cannot sell them, you don't know what they're worth, your balance sheet is not credible and the whole market freezes up. We don't know whom to lend to because we don't know who is sound. So if you could get rid of them, that would be an improvement." The only way to "get rid of them" is to sell them, which is why Ms. Schwartz thought that Treasury Secretary Hank Paulson's original proposal to buy these assets from the banks was "a step in the right direction."

The problem with that idea was, and is, how to price "toxic" assets that nobody wants. And lurking beneath that problem is another, stickier problem: If they are priced at current market levels, selling them would be a recipe for instant insolvency at many institutions. The fears that are locking up the credit markets would be realized, and a number of banks would probably fail.

Ms. Schwartz won't say so, but this is the dirty little secret that led Secretary Paulson to shift from buying bank assets to recapitalizing them directly, as the Treasury did this week. But in doing so, he's shifted from trying to save the banking system to trying to save banks. These are not, Ms. Schwartz argues, the same thing. In fact, by keeping otherwise insolvent banks afloat, the Federal Reserve and the Treasury have actually prolonged the crisis. "They should not be recapitalizing firms that should be shut down."

Buiter on the US-SARP

'Voluntary’ is clearly used here in the sense it is used in the armed forces: “I need three volunteers: you, you and you there!”

Paul de Grauwe - temporary nationalisation is the answer

From the FT:

"The essence of what banks do in normal times is to borrow short and lend long. In doing so, they transform short-term assets into long ones, thereby creating credit and liquidity. Put differently, by borrowing short and lending long, banks become less liquid, thereby making it possible for the non-banking sector to become more liquid; that is, have assets that are shorter than their liabilities. This is essential for the non-banking sector to run smoothly."

"The generalised distrust within the banking system has led to a situation where banks do not want to lend any more. That means that they continue to borrow short but lend equally short; that is, acquire the most liquid assets."

"How to get out of this bad equilibrium? There is only one way. The governments of the big countries (US, UK, the eurozone, possibly Japan) must take over their banking systems (or at least the significant banks). Governments are the only institutions that can solve the co-ordination failure at the heart of the liquidity crisis. They can do this because once the banks are in the hands of the state, they can be ordered to trust each other and to lend to each other. The faster governments take these steps, the better."

"Government interventions have consisted of recapitalising banks. These have not worked. The main reason is that they have been triggered by bank failures as they pop up and, as a result, have only dealt with the symptoms. The liquidity crisis is pulling down asset prices in an indiscriminate way, thereby transforming the liquidity crisis into solvency problems of individual banks. The governments, then, are forced to step in and to recapitalise the bank only to find out later that when the liquidity crisis strikes again, the capital has evaporated. The governments throw fresh capital into a black hole, where it disappears quickly."

"Central bank liquidity provision, although necessary, has also failed to address the co-ordination failure and has only made it easier for banks to dispose of long assets to acquire short ones (cash). Thus central banks’ liquidity provisions do not stop the massive destruction of credit and liquidity that is going on in the economy."

Willem Buiter on Grandson of Tarp

The first fundamental problem of the US banking system is now that that there are too many banks with too little capital. The second fundamental problem is that we still don’t know what the true value of the capital shortfall is and how it is distributed. The ‘we’ in the previous sentence certainly applies to those outside the banks holding the toxic assets. It may well include the management of some of these banks also.

TARP and its descendants are ‘value revelation mechanisms’. At the very least, if trades take place under the scheme, it will reveal the reservation prices of the sellers. Without TARP there are 3 valuations for a toxic asset. From low to high they are: its fire-sale value in illiquid private markets, the value at which it is carried on the bank’s balance sheet and the hold-to-maturity value.

For a bank to offer an asset for sale under the TARP, the TARP price has to be at least equal to its fire-sale value. Most likely, it will have to be at least as high as the fair or unfair value at which it is carried on the bank’s books. Otherwise a bank would effectively decapitalise itself by selling the asset at the TARP price. The only exception would be when the bank knows it is carrying the asset at an inflated value and expects its bluff to be called soon.

If the TARP succeeds in pricing most the toxic assets on the banks’ books, we will at least gain clarity about the true financial position of the banks. Only then can the recapitalisation of the banks be considered.

Grandson of TARP is a bad way to recapitalise the banking system

TARP could help recapitalise the banking system if the Treasury were to pay more for the toxic assets that what they are valued at on the banks’ balance sheets. That would be a terrible way to inject capital into the system, however. The banks would, of course, love it. But it would be an unconditional capital injection, giving the tax payer no upside on his risky investment and no voice in the management of the banks that are benefiting. I therefore propose that the TARP value the assets it purchases as aggressively as possible (that is, as close to the fire-sale price as possible).

How to inject capital into the banks?

A capital injection by the Treasury into the banks in exchange for equity (ordinary shares or preference shares with a warrant to convert into equity at a known price) would have two obvious consequences.

First, it would continue the nationalisation of the US banking sector.

Second, it would increase the total capital of the banking sector. From a longer-term perspective, that seems unwise. The sector will have to contract its employment and balance sheet to shrink the supply of financial services and products to match a much-reduced demand.

An alternative would be for the government to set a maximum leverage ratio (or a minimum capital ratio) but to give the existing shareholders pre-emptive rights to bring the actual capital up to the target level, with the government filling in the remaining gap, if any.

I would, however, prefer a mandatory debt-to-equity conversion for the banking system. Again, the government would set a maximum leverage ratio or minimum capital ratio, but now existing bank debt would be converted (in inverse order of seniority) into stock.

Effectively, a mandatory debt-to-equity conversion would be a sector-wide high-speed version of Chapter 11.

The disappearance of money and credit markets

I propose that the authorities for the time being socialise the interbank market. They can do this in one of two ways. Either the central bank becomes the universal counterparty in all interbank transactions, secured and unsecured. Banks lend only to the central bank and borrow only from the central bank, not to and from each other directly. This is already effectively the case in the overnight market.