The Goldman infallibility myth
AIG - the monstrously reckless US insurance and financial group - last night published a list of the "counterparties" that benefited from the $85bn emergency loan it received in September from the US central bank, the Federal Reserve.
This disclosure of which banks were on the other end of its complicated financial deals came after weeks of pressure from Congress. And it's a remarkable event: such information is typically cloaked in secrecy.
What it shows is why the US authorities felt they had to rescue AIG, while almost simultaneously (and some would say mistakenly) allowing Lehman to collapse.
If AIG had collapsed into bankruptcy, the losses for some of the world's biggest and most important banks would have been life-threatening for them and arguably lethal for the financial system as a whole.
Now, the name that leaps out for me as a leading beneficiary of the AIG bailout is Goldman Sachs.
Between 16 September and 31 December last year, Goldman received $2.6bn in collateral from AIG Financial Products - which in turn had been provided by the Federal Reserve - on credit default swaps (these are a kind of insurance against borrowers defaulting on loans, which are frequently used as a way of speculating on the health of businesses or other creditors).
There were subsequent payments to Goldman of $5.6bn, to purchase from it the securities underlying certain credit default swap contracts.
And there was a transfer to Goldman of $4.8bn to fulfil commitments under securities lending agreements.
So the gross sum received by Goldman from the US Federal Reserve, via AIG, was $13bn.
What that shows is Goldman would have been in the deepest, darkest doo-doo, if AIG hadn't been put on life support.
Which - some would say - rather explodes Goldman's fearsome reputation for controlling risk better than its rivals.
Goldman allowed itself to become deeply dependent on the health and fortunes of a business, AIG, which we now see to have been an unstable house of cards of a scale that boggles all comprehension
As it turned out, AIG was far too big to be allowed to fail by the US authorities. But few would argue that was a sound reason for Goldman - or anyone else - doing business with AIG.
If Goldman's senior executives don't wake up every morning and whisper "there but for the grace of...", well they wouldn't be quite human if they didn't.
That said, there are other fascinating conclusions to be drawn from the list of banks which received succour from the Fed, as intermediated by battered AIG.
The first is that the disclosures are something of a counterweight to the notion that French and German banks were more prudent than their US or UK rivals.
For example, the gross sum that Societe Generale of France received from the Fed via AIG was $11.9bn; and there was a gross transfer of Fed money to Deutsche Bank of $11.8bn.
It's also striking that in respect of this particularly debacle, neither Royal Bank of Scotland or HBOS - the UK's more accident-prone banks - were particularly exposed.
Of the British banks, Barclays benefited most from the lifeline given to AIG, receiving some $8.5bn (gross) of the unprecedented support given by the US central bank.
Anyway, the big point is that the losses and disruption for Goldman, Soc Gen, Deutsche and Barclays would have been hideous if AIG had imploded.
And if you were an investor in them, or a creditor to them, you'd be grateful for their luck - that they hitched their fortunes to a business, AIG, that was so enormous and complex that the US government had no other option but to put it on life support.
But if Goldman, Soc Gen, Deutsche and Barclays were to claim that they managed themselves more prudently than competitors, you might raise a querying eyebrow.
UPDATE, 17:30: Goldman Sachs has pointed out to me that it had taken out insurance against its AIG exposure - in the form of yet more credit default swaps - with other substantial financial institutions.
So in the event that AIG had collapsed, in theory its net losses would have been zero, because it would have been able to claim on these contracts.
In that sense, Goldman can probably still claim to be smarter than your average bank - except for one great imponderable.
If AIG had gone down, there would have been massive collateral damage to other financial institutions, including the insurers that had provided cover to Goldman on its AIG exposure.
Would they have been in a position to make Goldman "whole", to compensate it for its AIG losses?
We should probably be grateful that their ability to do so was never actually put to the test.