There are lots of calls to nationalize the banks and a while ago I supported to do such along the 'Swedish Model'. But I am getting less fond of this by each day. With nationalization the taxpayer takes over the risk of losses of these entities and may get a possible, but unlikely, upside.
The Swedes could do so because their authorities knew that the losses were restricted to the amounts of normal loans made to consumers and companies during a bubble. Sum all loans up and you have the maximum risk. The upside was in the continuity of normal economic activities and the downside was calculable as fraction of the annual GDP.
But since the early 1990s the banking system 'innovated' quite a bit and the possible losses are now not restricted to normal consumer and commercial loans on overvalued assets, but result from very different financial beasts.
From the Comptroller of the Currency – Administrator of National Banks this third quarter 2008 report (pdf):
• Net current credit exposure [of U.S. commercial banks] increased 7% from the second quarter to $435 billion, a level 73% more
than the $252 billion exposure of a year ago.
• The notional value of derivatives held by U.S. commercial banks decreased $6.3 trillion in the third
quarter, or 3%, to $175.8 trillion.
• Derivative contracts remain concentrated in interest rate products, which comprise 78% of total derivative notional values. The notional value of credit derivative contracts increased by 4% during the quarter to $16.1 trillion. Credit default swaps comprise 99% of credit derivatives.
Okay – derivatives in "interest rate products" are the mass behind the big number and may still be seen as somewhat reasonably. But how sure are we about future interest rates and the value of "interest rate products" when the U.S. will need to borrow $2.5 trillion this year and the rest of the world will need about the same amount? Will interest rates be negative or high in the positives for those borrowing and those who pony up the money for the governments to spend on rescuing banks?
(I for one am unlikely to buy bonds for the purpose of losing money.)
So the $176 trillions are indeed a frightening number. But the report also says:
The notional amount of a derivative contract is a reference amount from which contractual payments will be derived, but it is generally not an amount at risk.
Gosh – thanks, that's good. But then, how big is the risk?
[B]ecause the credit exposure is a function of movements in market rates, banks do not know, and can only estimate, how much the value of the derivative contract might be at various points of time in the future.
Now there you see my problem. The Swedes could estimate reasonably the maximum of the possible losses for the taxpayers when their government sized the banks.
Within today's international banking system it is impossible to even evaluate how big maximum losses for this or that bank could be. (For a wonkish example on how such derivative trades can blow up to even huger losses unexpectedly click here.) The only number we have is the hopefully top number, the notional value, and prudence demands to use that number if nothing else reliable is available.
Is $175.8 trillion is the maximum number here? Certainly a lot could be canceled out if you find and own the buyer and seller of 'insurance' and 'neutralize' their bets. Maybe $80 trillion less then. Some of these derivative contracts would turn out to be profitable contracts in the end? Subtract another $40 trillion of risk. Some contracts might be fraudulent and can be canceled in court? Subtract those too.
You will still end up with a huge amount of taxpayer money at risk that is a multiple of the annual U.S. GDP of some $14 trillion. Should the taxpayers want to take the risk responsibility for an amount of that size? For what?
Willem Buiter presents an alternative.
Found new "good banks," capitalize them with taxpayer money and let them take over the normal task of lending to consumers and commerce by granting favorable conditions. Regulate them strictly and, in a few years, privatize them. Meanwhile stop all state support for the existing banks. If they go bankrupt let deposit insurance click in for small savers, but let the rest of the mess fall in a normal court supervised bankruptcy.
I like that idea, but I fear the social/economic consequences of the short, disorderly and brutal phase that would occur when the big banks fail and believe it will be politically impossible to implement this.
My solution is a different one. I have called for all credit default swaps to be declared null and void four month ago. It may be too late for that by now to save the financial system that keeps the real economy going.
If one wants to save the real economy now, the banks will have to be nationalized in a kind of 'Swedish model'. But the risk to do so is much too high as long as those hundreds of trillions, mostly derivative swaps, stay in the books of these entities.
So my conditions for nationalizing banks is to clear them from any of these insane derivatives that are impossible, according to the Comptroller of the Currency, to be valued. Declare those obligations null and void and then nationalize.
Only then nationalize.