First, spam e-mail of the day. "Open jars in seconds!" is the message header. Now that's a product I need, because I've always thought it was taking too damn long to open jars. I mean, I opened a jar of olives the other day and it took me at least... well... seconds to open. Unacceptable! Aaaaaaanyway...
The concept of "too big to fail" has been bandied about quite a bit lately. For most people, the idea stems back to Continental Illinois, which was at one time the 7th largest U.S. bank, and was bailed out by the Federal Government in 1984. One might argue that the Chrysler bailout in 1979 was a similar theory, albeit for different reasons.
Does the too big to fail concept still hold today? I think it depends on what you mean. For the sake of discussion, let's frame the question this way: would the Federal government conduct a full scale bailout, either by taking an equity position (as in Continental) or by guaranteeing debt (as in Chrysler) of any financial institutions should they fail due to mortgage-related losses?
Before I move on, please note what we're talking about a direct bailout only. Not the Fed cutting interest rates to help banks, or the Treasury trying to facilitate a merger between two banks to prevent one of them from liquidating. I'm not even talking about the Fed agreeing to take unusual collateral from a bank at the Discount Window. I'm thinking strictly of a direct bailout. This is the difference between "we'd rather it didn't fail" and actually "too big to fail." (2B2F from here on, unless that sounds too much like an Astrodroid?)
For the sake of this discussion, I'll use three actual companies as examples. I've heard professional investors and/or analysts suggest a Federal bailout as a possibility in all three cases should it become necessary. The three are Countrywide, Citigroup, and FGIC. We'll consider what the consequences might be if one of these firms failed in isolation, and whether the Feds would likely get involved.
Now let's look at some of the common arguments for 2B2F and whether they might apply to any of these three firms. If you have your own 2B2F rationale, please write in the comments. If we get enough good ideas, I'll write a follow-up.
Failure would induce panic in the markets.
You'd assume the Federal government only gets involved if the "panic" would have wider economic reverberations. Not just a steep stock market sell-off, but a real lock-up of the credit markets. As we've discussed before on this blog, expensive credit isn't a big problem. Unavailable credit is.
In order to argue for a Federal bailout, it would have to a situation where the Fed's normal liquidity operations wouldn't be effective, and that the bailout itself wouldn't cause just as much panic as the bankruptcy.
I can't imagine a situation where a Federal bailout of any of the three companies wouldn't induce panic in and of itself. So in terms of calming the market, I don't know what a Federal bailout would accomplish. In terms of stemming a "run" on Citibank, perhaps. But Countrywide's bank is too small, and FGIC wouldn't be subject to anything similar to a run.
Liquidation of assets would create an unacceptable contagion.
This is a better argument for bailing out Citigroup, as they have over $2 trillion in assets, most of which are financial assets. A fire sale of Citi's assets would certainly have a major impact on financial markets. However, the overwhelming majority of Citi's assets are not in mortgage-related securities. So under a scenario where Citi is insolvent because they took giant losses in ABS CDOs, you'd think there would be a buyer of the rest of their assets. It would seem as though the Fed could orchestrate a merger or at least a buyout of various Citi units. I'll note that Continental Illinois was far less diversified compared with Citigroup, the former having large exposure to the then sinking oil market. By all accounts, the Fed sought a willing buyer for Continental for a couple months before the FDIC eventually took an equity position. It would seem that Citi would have valuable parts, even if their CDO/ABS positions had sunk the whole.
Countrywide is drastically smaller than Citi, but their assets are more concentrated in the home loan market. It may be fair to say that Countrywide may have more assets in "problem" markets, like sub-prime HELOCs, at least as a percentage of assets. On the other hand, Countrywide with about $200 billion in assets is considerably smaller than Northern Rock was, and the BoE was able to orchestrate a buyout there.
FGIC wouldn't actually have to liquidate, they'd just run off the insured portfolio until there was nothing left. Whether or not there might be a buyer for FGIC would depend on various factors, but there doesn't seem to be any contagion risk to FGIC liquidating. Right now there are some large municipal buyers who hold FGIC-insured paper and have contacted other insurers about replacement insurance. If FGIC went bankrupt, there would likely be more of this.
The firm is an integral part of vital systems (e.g. check processing). Failure would cause a shut down of these systems.
This was a concern with Continental. And while Citi may be bigger and more deeply ingrained in our financial processing system, I'd like to think that technological advances since 1984 have eliminated this risk. Or at least made it such that the processing could be easily passed to another firm at a low cost. I don't know that much about bank's back offices, but it would seem like this could be accomplished.
Both Citi and Countrywide are very large mortgage services. Given the level of delinquencies and foreclosures right now, mortgage servicing is a critical system for the U.S. economy. We cannot have a situation where a large number of loans effectively have no service all of a sudden. My sense is that this too could be passed to another firm, although I admit I don't know enough about the servicing business to say this for certain. I'd just think that given how many loans are bought and sold routinely in the mortgage market, that there are good systems for passing servicing rights from one firm to another, which would be effective even if the service was very large.
FGIC is indeed an important part of the municipal market. However, in almost all cases, municipalities have paid for insurance up front, and if the bond rate is fixed at issuance, then the subsequent bankruptcy of the insurer is of no moment to the municipal issuer. Put another way, if the City of New York sold a municipal bond with FGIC insurance with a 4% coupon in 2005, it
will have a 4% coupon until the bond matures, no matter what happens with FGIC. New York doesn't care. Hence there would be no reason for municipalities to create political pressure to help out FGIC.
Now, its likely that a FGIC bankruptcy would cause some weird trading in the muni market. Recently Radian-insured bonds have traded worse than the underlying rating, implying that the Radian insurance makes the credit worse than had there been no insurance at all. Radian isn't even bankrupt, and in fact have had their credit rating affirmed recently. If FGIC actually went
bankrupt, all hell would break loose in secondary muni trading. But I can't see the Federal government seeing this as a good reason to bail out FGIC. And few municipalities, if any, would feel the pain directly. New issuers would just go to one of the other insurers, or issue without insurance. The muni market would eventually normalize and we'd all move on.
For what its worth, I can imagine various street firms getting together and buying FGIC, since the perceived value of muni insurance makes muni underwriting easier and hence more profitable. It may be that the street has their own capital problems and such a thing never materializes, but its worth thinking about.
Job losses would be too great.
While Citi certainly employs over 300,000 people (not all in the U.S.), I don't think this would be viewed the same as Chrysler was in 1979. First of all, government's attitude toward interference industry is quite different today vs. the late 1970's. There were also other pressures on the government, including the rise of Japanese auto makers, public fear that U.S. industry was becoming irrelevant, as well as union political clout. Citigroup wouldn't have any of this putting pressure on the government to bail them out. In fact, if I'm right that Citi's pieces could be sold, then the job losses would be small in the grand scheme of the U.S. economy.
Neither FGIC nor Countrywide employs enough people for this to be a serious consideration.
In sum, it seems like 2B2F isn't relevant to any of the big names where insolvency has been thrown around recently. The only two companies where I think 2B2F would hold would be Freddie Mac and Fannie Mae. Particularly now, when the disappearance of either would decrease liquidity in the mortgage market at a time when liquidity is sorely lacking. Over the years, commentators have greatly overestimated the cost (or implied that the cost would be greater) of bailing out either of the GSEs. The cost wouldn't have anything to do with the size of their mortgage holdings. It would be the size of their losses. So if you hear dollar figures starting with a "T" stop reading and find another source to read.
Let's hope it doesn't come to that.