It hasn't been a great year for monolines, and Pershing Square Capital's Bill Ackman has been one of the main beneficiaries. He's famously made a kings ransom shorting MBIA, betting that losses on collateralized debt obligations and other asset-backed securities would eventually drive MBIA bankrupt.But now he's set his sights on far more conservative Financial Security Assurance (FSA). As FSA is wholy owned by European banking giant Dexia, there is no stock to short. Instead Ackman has bought credit default swap (CDS) protection against FSA defaulting on its insurance obligations. Currently FSA is rated AAA/Aaa/AAA by S&P, Moody's, and Fitch respectively with a stable outlook by all three.
Ackman's revealed his position at a conference in New York on June 18. The market paid attention. The next day CDS on FSA had moved 200bps wider to 700bps/year for protection. It rallied into the 400bps area two days later when Dexia extended a $5 billion credit line to FSA, but has since moved back into the 700 area. For context, this morning Lehman Brothers CDS was +280 offer, Washington Mutual was +585. To get into the 700 area you have to look at names like National City.
Does FSA belong in the same category as Ambac and MBIA? I'm going to explore this in two parts, first on the liability side, then on the asset side. Here are the raw facts on insurance liabilities.
The following chart details the direct residential mortgage (RMBS) exposure from MBIA and Ambac (blue bars) and FSA (red bar). Each is expressed as a percentage of the firm's total claims paying resources. So for example, Ambac has exposure to home equity lines of credit at 216%, which means that if their entire HELOC exposure went to zero, the firm would exhaust their claims paying resources two times over. The figures are from S&P and the companies themselves.
So looking at this, FSA is no better than its more troubled competitors. However, when it comes to direct RMBS exposure, the story isn't quite as dire as it would initially seem. The monolines' generally insured senior positions in RMBS deals, meaning that other securities would absorb losses first. For example, FSA states that their typical subprime RMBS transaction has 20% subordination and 7% excess spread, or excess interest collected by the trust for benefit of senior bond holders. In their first quarter earnings release, FSA estimated that more than 45% of all subprime borrowers would have to default (in a given deal) in order for FSA to pay a single dollar in claims.
Will any transaction suffer 45% defaults? Some probably will, but all of them won't. And the second lien transactions (both closed-end and HELOCs) will likely enjoy no recovery upon default, so those loss severities will be worse. On the other hand, the monolines have the luxury of making payments on RMBS losses over time, i.e., there is no large principal payment which would come due all at once. So RMBS losses will be substantial to be sure, but its probably manageable. But that's not where the really big problem is.
The big problem is in collateralized debt obligations (CDOs).
The key to the CDO creation game was to create the maximum return to the equity holder while still earning a AAA rating for the senior-most holder. In creating CDOs using RMBS as collateral, the arranger generally used subordinated securities. And by subordinated, I mean the first ones to take losses when the underlying borrowers default. Today it is widely assumed that most subordinated sub-prime securities won't receive any principal at all, and many subordinated prime RMBS will suffer significant impairments. So a CDO made up of these subordinate securities, even the senior-most piece of the CDO, is likely to incur large losses.
Which brings us to our second chart, RMBS CDO exposure, again as a percentage of claims-paying resources.
The reality is that most of these RMBS-oriented CDOs are going to take losses of 30% or more. CDOs alone will likely sap the resources of Ambac and MBIA. But FSA's exposure to this sector is nominal. Even if they were to take 100% losses on their RMBS CDO portfolio, the impact on their solvency will be minor.
Remember that FSA is a subsidiary of Dexia, and therefore Ackman's bet isn't on the common stock. In the case of MBIA, Ackman was able to short the common. Even if MBIA is somehow able to pull through as a solvent company, Ackman's gains on shorting the stock will still be substantial. With FSA, he's long CDS protection, which only pay off in the event of a default, so Ackman's bet is not on a decline in profitability or loss of AAA ratings, but on bankruptcy. In fact, FSA could lose its AAA ratings and Dexia could give up on the financial guaranty business and put FSA into run-off, and as long as FSA never runs out of cash, Ackman's CDS will expire worthless. Sure, Ackman could gain on CDS widening. Maybe he's already cashed in his gains on FSA. But given the already wide spread on FSA CDS, if FSA doesn't have liquidity problems, it will be an expensive short.
The facts on the insured side just don't support FSA going the way of MBIA or Ambac, particularly when you limit the discussion purely to solvency. Next up, a look at FSA's investment portfolio. (Hint, that doesn't look as rosy.)