Bond Insurer Notes: Preliminary Estimate of Impact on Banks; Some Monolines Close to Regulatory Requirement; ABX Overstates Losses

I don't think a day goes without something happening in the bond insurance world. A few notes on preliminary loss estimates due to the monoline dowgrades, talk of some monolines being close to breaching regulatory capital requirement, and BIS points out that the widely-followed Markit ABX overstates losses on AAA mortgage products by up to 60%!

If you are are bond insurer fan--or enemy--click on 'read more'... otherwise, nothing to see, move along... :)



Preliminary Loss Estimates Due to Monoline Downgrades

Preliminary estimates are coming out on the impact of the recent monoline downgrades. All the articles seem to concentrate on the Ambac and MBIA downgrade, but a few of the other smaller monolines were also downgraded and losses from those have to be accounted for as well. I don't have access to the S&P reports (pay only) but here is what news organizations are picking up:

"In our view, the impact of the downgrades on the structured finance market is wide spread," S&P said. Together, the downgrades affect around $350 billion of structured deals, with Ambac accounting for 60 percent of these deals, and 56 percent of the dollar amount, the rating agency said...

Many financial companies that are exposed to the bond insurance sector have already written down the value of their insurance, and current ratings on the banks have leeway to absorb incremental additional writedowns, S&P said.

The impact of the downgrade will mostly be on the structured product side. If I'm not mistaken (the quote above isn't clear,) Ambac is the largest player in the structured side, mostly through huge exposure through CDOs. The downgrades will impact all insured products including ABS of student loans, ABS of auto loans, and so forth. Everyone talks about the subprime mortgage assets but I'm not sure who owns these non-mortgage ABS assets.

Merrill Lynch has one of the largest exposures to these types of assets, with $18.8 billion in super senior asset-backed Collateral Debt Obligations (CDOs) as of March 28, S&P said. Merrill wrote down $1.34 billion of this exposure in the fourth quarter of 2007 and the first quarter of 2008.

Similarly, Citigroup has purchased $10.5 billion in protection of similar securities, and wrote down $1.5 billion of this exposure in the first quarter of 2008, S&P said.

However, "for banks and brokers, we currently view the impact of credit erosion in the monoline sector as manageable, despite the high notional amount of securities involved," S&P said.

Losses have been telegraphed for a few months now so the impact isn't likely to be large. Furthermore, the banks have been posting massive losses on mortgage assets that have nothing to do with monoline insurance so the monoline write-downs are only a small portion of them. It should also be noted that the downgrade of Ambac and MBIA is from AAA to AA so the write-downs probably won't be large (I'm not an accountant but I assume it is tied to the bond insurer's ability to pay.)

As for the municipal bond side...

The impact of the downgrades on the municipal bond market, meanwhile, will likely be limited as competing "AAA" rated insurers will easily absorb Ambac and MBIA's business, while a higher percentage of municipal debt has also been successfully sold without an insurance wrap...

Money market funds holding municipal debt are also unlikely to need to be forced sellers of securities insured by the companies, as their current "AA" rating is adequate for the credit quality guidelines.

Muni side won't see much impact for several reasons. First of all, anyone that was really desperate could have taken out insurance from one of the AAA-rated monolines. This is actually most of the business that Assured Guaranty, Berkshire Hathaway, and others were underwriting recently. Secondly, most of the underlying muni bonds have high ratings (A to AA) and those ratings haven't been downgraded. Lastly, most of the muni bond buyers seem to be individual investors and investment funds who really don't have to mark to market (even if they had to, it wouldn't impact their regulatory capital like banks.) The only dowside is likely to come from forced selling but that seems to be limited for the time being.

Some analysts like Meredith Whitney of Oppenheimer seems to think that the big Wall Street banks will post around $10 billion of losses from the monoline dowgrades:

Citigroup Inc., Merrill Lynch & Co. and UBS AG may post losses of $10 billion on bond insurance after MBIA Inc. and Ambac Financial Group Inc. lost their top credit ratings, Oppenheimer & Co. analyst Meredith Whitney said.

Ten billion sounds like a big number but it is spread across a few firms and some of that will likely be hedged. My wild guess (without doing any analysis but forming an opinion based on my knowledge of Ambac and MBIA exposure) is that you will see around $10 billion to $15 billion in write-downs due to the downgrade. This does not account for gains from hedges or prior write-downs (i.e. some firms may have written down more than they needed earlier in the year).


Some Monolines At Risk of Insolvency (Supposedly)

It's no doubt tough running a monoline these days but things are getting very dicey for some firms according to one analyst:

Bond insurers CIFG Guaranty, Financial Guaranty Insurance Co and XL Capital Assurance may breach capital requirements in this quarter, triggering a chain of events that could push them into insolvency, according to an analyst report.

The bond insurers could be taken over by regulators if capital levels fall below the required minimum and they are unable to raise more money.

That would force the insurers to make massive payouts on derivative contracts, which would likely wipe out all of their resources, CreditSights analyst Rob Haines said in a report published late on Sunday...

Under terms of these credit default swap contracts, if a regulator takes control of an insurer, the company would immediately need to pay out the full amount of the insurance. In return, it would receive the insured debt, which in the case of mortgage-backed securities may see little recovery.

The situation is far more complicated than it is portrayed. Regulatory requirements are complicated and it may be possible to satisfy the requirements by liquidating some assets (I'm not sure). I'm also not sure if the regulatory requirements are based on mark-to-market losses or on credit impairment. If it's based on actual impairment, it may be manageable.

CIFG has only $15 million surplus capital above the requirement, the smallest cushion among the bond insurers, according to the report.

FGIC has an extra $300 million of capital and Security Capital Assurance's XL Capital Assurance is $102 million above the requirement. XLCA's figures do not include reinsurance from XL Financial Assurance, which guarantees 75 percent of its exposure and had $918 million of regulatory capital at the end of first quarter.

Spokesmen the New York State Insurance Department and FGIC declined to comment. Spokesmen for the other companies could not be immediately reached for comment.

Industry leaders, MBIA Insurance Corp and Ambac Assurance Corp. that guarantee more than $1 trillion of securities, have surplus capital of $3.9 billion and $3.5 billion respectively thanks to recent capital injections, according to the report.

One should exercise caution with these estimates. As mentioned in the article above, XL Capital only has $102 million surplus but that's without taking reinsurance into account. I'm not sure if the other estimates include reinsurance or how the regulator would look upon reinsurance. It should also be noted that many of these analysts have been wrong on the way up, wrong on the way down, and likely will be wrong on any rebound as well.


Markit's ABX Index Overstates Losses

Bank for International Settlements--the central bankers' bank--recently said that the ABX index likely overstates losses by as much as 60%:

The use of prices from the ABX derivatives indexes to estimate valuations for U.S. subprime mortgage bonds may inflate likely losses at the triple-A level by over 60 percent, the Bank for International Settlements said.

The ABX indexes, each series of which is based on credit default swaps linked to a portfolio of 20 subprime-mortgage-backed securities (MBS), have been widely used as a yardstick for possible losses given the illiquidity and opacity of the underlying cash bond markets.

In my opinion, the ABX index will likely become the poster boy of what is wrong with investing when the crisis subsides. By using the index as a standard, it may end up costing investors hundreads of billions (by unnecessarily diluting their ownership at depressed prices.) Ironically, if I'm not mistaken, Markit is owned by the banks themselves and it's dissapointing to see Markit not elaborate on the flaws of its index and prevent "misuse" in valuing securities.

But the BIS said at least three sources of bias could mean that loss estimates based on the indexes have been overstated.

Firstly, subprime MBS may be held to maturity rather than as trading instruments, making them liable to a different accounting treatment that would tend to reduce writedowns and impairment charges, the BIS said.

Secondly, the ABX may not be representative of the overall market, of which each series represents just 5 percent. The BIS acknowledged that ABX deal composition in terms of credit scores and loan-to-value ratios was similar to the overall market, however, meaning this source of bias may not be large.

Finally, the deal portions referenced by the ABX triple-A indexes are not the most senior portions of the capital structure, which may tend to overstate losses, the BIS said. Instead, they reference more subordinated triple-A tranches with the longest duration.

Well, there you have it. Two of those reasons are very strong in my eyes for why the index is not what it seems. Unfortunately, I suspect accountants, who generally just want to get paid and could care less about the true meaning of what they mean, will keep using the ABX index.

Regarding the last point, the alea blog pointed out a while ago to the fact that the AAA tranche in the Markit ABX index is the lowest tranche within the AAA rating (yes, it's as confusing as it seems; you can supposedly have a AAA tranche broken out into various AAA tranches with totally different risk characteristics.) Zack Maxfield of bankstocks.com has also written in the past about the flawed ABX index (article 1 and article 2).


Maybe all this is just too much optimism on my part but whathever the hell it is, these losses on highly-rated CDOs and direct RMBS needs to stay low for Ambac, MBIA, and the other monolines to get out alive from the abyss that they have fallen into. So far, the fundamentals don't seem anywhere near as bad as they seem. From everyone falling in love with the ABX as the beacon of truth a few months ago, now we have influential bodies like the BIS saying losses may be overstated. Next positive step will be, assuming it materializes, for the accounting bodies to start distancing themselves from fair value accounting and these ABX indexes that they likely rely on. Unfortunately that won't happen until well after the fact, probably after we get mark-to-market gains on the written-down losses in an year's time (it's kind of like how the accountants realized a few years ago, well after the dot-com bust and most companies went bankrupt, that options are not free and need to be accounted for somehow. It must have been truly shocking, for a non-investor that is, to realize that issuing options like they were going out of fashion diluted the shareholders.) For the time being, the name of the game for the monolines is to lie low and keep costs down. I really hope Ambac dumps the Connie Lee idea for the time being (management trying to get too cute again :( ). Wait 'till the mark-to-market losses stabilize and then initiate some strategy!