Sharing upside and downside risk ...

Sovereign wealth funds have invested about $35b in US financial institutions. Adding in Qatar’s investment in Barclays and Singapore’s investment (through the GIC) in UBS brings the total up sovereign funds have invested in firms with a large US presence to around $55b.*

The US taxpayer is now being asked to invest $700b to help recapitalize the global financial system – a sum that is more than 10 times as much as the world’s sovereign funds put in.

But, at least as I read Paulson’s initial proposal, the US taxpayer would not get any equity in the world’s large financial institutions in exchange for this help.

Now the US isn’t making a pure equity investment, though some – like Doug Elmendorf and Sebastian Mallaby– think it should.

It is buying the banks’ illiquid assets.

But there is at least the possibility that it will “overpay” for those assets, and in the process effectively contribute equity capital to the US and global banking system. Indeed, there is a real probability it will overpay by more than the $55b sovereign funds have put into the global financial system.

There are broadly speaking two ways a government can recapitalize a banking system.

One is to put in a lot of equity. That equity can be put in to allow the banks to write down (and eventually sell) their bad assets – or it can be put after the banks’ have down the write down, providing the funds both to make the banks’ creditors whole and to supply the banks with new equity. It works out to the same thing. Such an equity infusion is good for the holders of the financial systems debts (depositors, money market funds, bond holders) and bad for the holders of the banks’ equity.

The other is to buy the banks bad debt. That clearly generates a bit of liquidity for the banks – they have more cash on hand, and thus more capacity to make new loans. But it also is risks providing a large gift to the banks equity investors – as they get to move an illiquid asset off the banks books at what may well prove to be an above market price.

Most bank recapitalizations have elements of both. The FT – in its usual sensible way – suggests that the US recapitalization should too. I agree.

To give a concrete example, the Chinese recapitalized their state banks earlier this decade both by buying bad assets off the banks books (usually by exchanging bad assets for the bond of an asset management company, with bad assets shifted to various asset management companies at book value) and by injecting new public funds (whether new funds from the Finance Ministry or some of the PBoC’s foreign exchange reserves) in exchange for the banks’ equity. Now the state banks were already owned by the government, so it didn’t matter too much – apart from internal accounting -- exactly how the system was recapitalized: Chinese taxpayers stood to gain on their “equity” in the state banks if China’s taxpayers overpaid for the banks bad assets. But the Chinese example still illustrates the range of choices available to the government.

Some banks still seem to have faith in the illiquid assets they hold and worry that the government might be getting too good a deal on their MBS and complex securities; it will buy distressed assets at a discount from illiquid institutions, and could end up with a significant financial gain if it holds those assets to maturity. Maybe. The mysterious knzn at least initially believed that this scenario wasn’t entirely unrealistic. The obvious solution to this concern is to give the banks a bit of the government’s upside: if the taxpayers make a profit, some of it could be “given back” to the banks.

On the other hand, the government might end up overpaying – perhaps significantly – on the banks’ assets. That helps the banks and hurts the government. The fairly obvious solution here is to give the taxpayers some of the banks upside.

Apparently one objection to an “equity” component of a recapitalization is "a gut Congressional reaction against the government taking equity stakes in a broad array of American corporations.”

Alas, that cart has already left the barn.

The Fed effectively bought AIG last Wednesday – and the Treasury took over Fannie and Freddie the previous weekend.

And unless foreign governments’ do not count, governments (though not the government) already have an equity stake in a lot of US financial firms. The governments of Singapore, Abu Dhabi, China, Korea, Kuwait and Qatar (through Barclay’s investment in Lehman) all hold equity stakes in the US financial system.

Is it really better to reward the banks’ existing equity investors -- remember, they own the financial institutions that made the bad bets that led the financial system to seize up – to avoid a US government equity stake?

* My accounting is as follows:

Morgan Stanley: $5.6b (CIC)

Citi: $17.4b ( $7.5b from ADIA, $6.88b from the GIC, $3 from the KIA)

Merill: $12.5b ($5.9b from Temasek and, per Craig Karmin and Carolyn Cui of the Wall Street Journal, $6.6b from KIA and the KIC)

Total: $35.5b

UBS: $9.54b from the GIC and $1.8 from an unnamed investor in the Gulf widely thought to be a member of one of the region’s royal families.

Barclays: around $9b from a rights issue with substantial participation from both the QIA and Sheik Hamad’s private "Challenger" fund (Sheik Hamad also runs the QIA)

Combined total: $55.84b

This leaves out some pre-crisis investments – like the CDB’s investment in Barclay’s. And I am not totally confident of the accounting for the increased stake various funds took in Merrill when the initial deal was reopened, or for the Barclay’s rights issue. It is an approximation.

Sovereign funds may have provided additional funds through private equity funds or participation in public rights/ convertible issues. I have no way of tracking those investments.