Quantifying the moral hazard trade
I’ve been having an interesting email discussion with Jim Surowiecki of the New Yorker about quantifying the moral hazard trade. If big banks have lower borrowing costs than small banks, he asks, why do we automatically attribute that to moral hazard (the idea that they’re much more likely to get bailed out in extremis) rather than the simple fact that they’re less likely to default? He also makes Jamie Dimon’s point that since the debt of too-big-to-fail banks wasn’t trading at risk-free rates, there can’t have been a moral-hazard trade going on.
The second point doesn’t hold much water. Moral hazard isn’t a binary thing which either exists or doesn’t. So long as the probability of being bailed out by the government is greater than zero, a bank’s cost of funds will fall a little. And even if the probability of being bailed out was 100%, that doesn’t mean that banks could borrow at the same rate as Treasury. As David Merkel points out, even debt with explicit (rather than implicit) government guarantees sometimes trades at a significant spread over Treasuries.
As to the first point, Carrick Mollenkamp comes to the rescue today with the fascinating tale of COFI, an index off which many mortgages are priced. It suddenly spiked on New Year’s Eve, from 1.259% to 2.094%, resulting in an unexpected increase in monthly payments for people with adjustable-rate mortgages.
It turns out that most of the COFI index was based on the cost of funds at Golden West, a subsidiary of Wachovia. When Wachovia was taken over by Wells Fargo, it was no longer eligible to be a reporting member of COFI, and the COFI index went from being overwhelmingly based on Wachovia’s cost of funds to being much more reflective of what smaller banks are paying. And it turns out that smaller banks pay much more than Wachovia did for money.
This datapoint is telling, because Wachovia — largely because of the Golden West acquisition — was a very rocky bank indeed, and was sold as a highly-distressed asset to Wells Fargo. The fact that its cost of funds was so low clearly had nothing to do with its inherent safety, which means that we have to attribute it instead to the moral hazard trade. And indeed the government went to great lengths to rescue Wachovia’s bondholders by forcing a sale of the bank.
Note here that Wachovia’s shareholders were pretty much wiped out in the deal — but that was fine by anybody playing the moral-hazard trade, just so long as the bondholders remained whole. Surowiecki says that “In order to believe that the banks engaged in reckless behavior because they assumed that if they got into trouble, the government would bail them out, you have to believe that financial institutions thought it would be fine if their share prices were driven down to near-zero as long as they were rescued in the end.” Not at all: Wachovia’s executives were very unhappy about the forced sale, but they had no choice in the matter, precisely because the government bailout expected by the moral-hazard types overruled any internal management decisions. Wachovia was not really party to the negotiations surrounding its sale: those essentially happened entirely between the FDIC, Citigroup, and Wells Fargo.
The small banks which now make up the COFI index are safer institutions than Wachovia, but because they aren’t subject to the moral hazard trade, their cost of funds is higher. That’s why it’s no great hardship for Wells Fargo to be paying a 15bp bank tax: it saves much more than that on its cost of funds just by dint of being too big to fail. And there’s a good case that in fact the tax should be significantly higher.