Why the government should have nationalized AIG
This weekend’s big NYT story on Goldman Sachs and AIG does not, I think, say anything much we didn’t already know. But it does suggest, I think, that the US government made a big mistake in bailing out AIG rather than nationalizing it outright:
The government would soon settle the yearlong dispute between Goldman and A.I.G., with Goldman receiving full value for its bets. The federal bailout locked in the paper losses of those deals for A.I.G. The prices on many of those securities have since rebounded.
The dispute between Goldman and AIG was one over collateral. Goldman wanted AIG to put up ever-increasing amounts of collateral against the CDS which it had written — demands for cash which AIG was unable to meet. So the government stepped in and unwound the contracts near the bottom of the market, paying out Goldman Sachs and other AIG counterparties in full, and locking in massive losses for the insurer.
The alternative would have been to nationalize AIG outright, and imbue it with the government’s own triple-A credit rating. Since many of the largest collateral calls were a function of AIG’s own deteriorating credit rating, that alone would have helped to minimize the amount of cash needed to be put up as collateral. AIG, rather than unwinding all those CDSs, could then simply have held onto them, putting up as much collateral as it needed to, and paying out on them as and when the underlying bonds defaulted. The end result would, with hindsight, have been significantly cheaper for both AIG and US taxpayers.
Now it is true that as part of the AIG bailout, the New York Fed took possession of a lot of CDOs, putting them in portfolios with names like Maiden Lane III, and hiring Blackrock to manage them. As the value of those CDOs has risen, the US government has seen mark-to-market gains on its portfolio. But that doesn’t change the fact that Goldman Sachs bought credit insurance very cheap from AIG, and then sold it back at a very high price to the US government, locking in billions of dollars in trading gains. AIG took equal and opposite losses on those transactions, and ended up passing those losses on to the US taxpayer.
It will be many years before it becomes clear whether or not Goldman pulled off a great trade here, cashing in on its insurance assets at the height of the panic. It’s still possible that the bank would have been better off holding all that insurance to maturity, and collecting a steady income stream over the years as various instruments went into default. But Goldman will tell you until it’s blue in the face that it always marks all its positions to market, and that it doesn’t really believe in holding financial assets for very long time periods.
So at the very least, AIG and the New York Fed should have threatened to call Goldman’s bluff, and said OK, we’ll continue to put up collateral. Goldman would then have had to hand that collateral back over the course of 2009 as the credit markets rebounded, and AIG wouldn’t have locked in any losses at all. But no one did that, because AIG was only 80% owned by the government, and the government didn’t want to provide essentially unlimited liquidity support to a company which still had a relatively large number of private shareholders. Outright nationalization, then, might have been a much simpler — and cheaper, in the long run — way of addressing the situation.