The Bankruptcy Threat

-What keeps you up at night, Mr Dillinger?
-Coffee.

Being rather lightly caffeinated, there isn't much keeping this blogger wide-eyed through the night. Still if we were at the helm of a bond insurer, pondering the recapitalization of our firm and industry the cost of buying credit insurance on Countrywide Financial Corp would seem to be a bargain with a spread ask price under 110 bps on the 2 yr contract. That implies a 3% chance of default with a 20% recovery rate.

Countrywide Financial Corp is the holding company acquired by Bank of America and it is also a defendant in the guarantors' suits.

If one supposes that BofA offered a settlement to MBIA at approximately 80% of incurred losses last week, Jay Brown could buy this hedge and drag Moynihan to court to face strong and validated claims. Of course, MBIA also has BofA on the hook for successor liability but there appears to be at least some threat of this being reversed if Delaware law is applied instead of New York law.

Other guarantors have also had BofA added to their respective suits. With CFC CDS priced below that of BofA, even a remote chance of the subsidiary being severed makes this a compelling hedge. If nothing else, it removes BofA's best negotiating tool for a very reasonable price. In fact, CFC 2yr contracts are pricing about 5 bps lower than comparable BofA contracts.

For those with access to the market, this is an equally effective hedge to a long guarantor position - similar to the original long MBIA, short BofA positions suggested by Manal Mehta and Branch Hill Capital over a year ago. A long (protection selling) MBIA Inc. (Hold Co.) CDS position could finance multiples of notional in a short (protection buying) CFC position.

The market says that this is not a substantial threat. This threat doesn't keep me awake at night either. But that is partly because Mr. Market has offered such a cost effective manner to eliminate the risk.

BofA Discusses Repurchases

Reports suggest that Bank of America is indicating further reserve expenses for private label and monoline reps and warranties. Monoline and bank stocks have outperformed over the past two days.

A presentation by BofA Legacy Asset Servicing Executive Terry Laughlin contains the bullet point on non-GSE repurchase losses:

“Estimate of upper range of possible loss could be up to $7B to $10B over existing accruals - this does not represent a probable loss.”

Bank of America ended 2010 with $5.4B in total reps and warranties loss reserves. This number includes GSE exposure, but BofA did settle much of its exposure with the GSE’s prior to year end.

Another presentation by Chuck Noski, the bank's CFO indicated that the company’s “Rep and warranties costs expected to be lumpy in 2011.”

None of these presentations or any of the comments reportedly coming from the investor day is inconsistent with very high monoline recoveries. The past two days of stock performance suggest that there is plenty of room for both sides of this mess to benefit from these developments – a continuing theme at The Blue Dragon.

An Actual Case of Reps and Warranties - CWHEQ 2006-S10

We have just published excerpts from the CWHEQ 2006-S10 deal, one of fifteen securitizations that are the subjects of the flagship repurchase case MBIA Insurance Corp. v. Countrywide Home Loans et al. We chose that deal because we thought it likely to be loaded to the brim with the stinkiest of crappy home equity. For future reference, these will be available under the label "CWHEQ 2006-S10."

The purpose for pulling these excerpts was to identify the points that we feel are the crux of MBIA's allegations. Mark Tapley may have been born in the 1800's, but not even experience compares with the certainty obtained from primary documents. We are all blessed with varying degrees of obsession, so we provide a summary here, what we consider key excerpts on the relevant posts, and links to the corresponding original documents on each excerpt post. There are also other neat documents like the insurance policy** on the SEC website. So dig as deep as you please!

In this examination, it is made plain that Countrywide made representations and warranties as to the characteristics of individual loans and the overall loan pool. Specifically in the Pooling and Servicing Agreement, Countrywide reps and warrants that "(7) the information set forth on the Mortgage Loan Schedule with respect to each Initial Mortgage Loan is true and correct in all material respects" and "(52) the Mortgage Loans, individually and in the aggregate, conform in all material respects to the descriptions thereof in the Prospectus Supplement." 

Because of their relevancy and specific citation, we have included excerpts from and links to the final Collateral Tables and Prospectus Supplement. Countrywide is responsible for the loans, individually and in aggregate, having the characteristics described. Most notably, this includes any representations of FICO, LTV, structure type, debt-to-income*, and origination type (Refi, Cash out, Purchase.) Any misrepresentation by the mortgagor not caught by Countrywide is Countrywide's problem, not MBIA's.

It is important to note that Countrywide specifically identifies residence type (primary, secondary, investment) data to be "based upon representations of the related borrowers at the time of origination." This is an important impediment against MBIA on this data point. However, in specifying such a qualification here, Countrywide strengthens MBIA's claim that Countrywide is responsible for the veracity of all other loan characteristics - regardless of mortgagor fraud. Some may consider this note irrelevant due to Countrywide's explicit representations and warranties of the loan characteristics, and we would agree.

And yes, Countrywide specifically reps and warrants that the loans it originated met its own guidelines.

"(45) The Mortgage Loans originated by CHL were underwritten in all material respects in accordance with CHL's underwriting guidelines for closed-end second lien mortgage loans or, with respect to Mortgage Loans purchased by CHL were underwritten in all material respects in accordance with customary and prudent underwriting guidelines generally used by originators of closed-end second lien mortgage loans."

Those underwriting guidelines were described in the prospectus supplement:

"Countrywide Home Loans will conduct a further credit investigation of the applicant. This investigation includes obtaining and reviewing an independent credit bureau report on the credit history of the applicant to evaluate the applicant’s ability and willingness to repay."

"Appraisals are generally performed for all closed end second mortgages" over $100k. 

Other significant reps and warranties that we think are the basis for MBIA recoveries include but are not limited to:

(10) No LTV over 100%;

(46) Appraisal by a qualified, independent appraiser;

(64) No predatory loans as defined by federal, state, or local law. 

Finally, The Pool and Servicing agreement goes on to describe the repurchase and replacement mechanism to be the contractual cure for any breach of the reps and warranties listed. The time frame in which such cure must take place is 90 days.

So if it pleases you to dig deeper see the posts published just prior to these. 

*Update: Or clarification really. We have not found debt-to-income to be an R&W in public documents - most notably the prospectus supplement - other than as referenced in underwriting guidelines which we consider weaker than the item 52 reps and warrants. All the other items have been listed publicly in the aggregate insomuch as they apply to the Statistical Calculation Pool which is basically the main piece of Initial Mortgage Loans but not the Subsequent Mortgage Loans which must not alter certain characteristics outside of a given band.
**Update again: we erroneously stated that the insurance agreement was available on through the SEC system. That document is not public as far as we know. We meant to say insurance policy.

The latest settlement... until the next one.

If you're reading this, you know about it so I'll spare you the review.

My take: it's good for B of A and has good implications for Financial Guarantors too. How can that be?

Well it shows BofA's GSE reserves were about half of what they should have been but BofA will not be wiped off the face of the earth. And this is pleasing to both sides. Look at the stocks today.

Every settlement that doesn't point to exploding investment banks is good for both sides. For the banks it's obvious. For the guarantors it's deeper than just ability to pay. The more the GSE's and uninsured private labels take care of themselves or are otherwise distinguished from the guarantors, the less danger of a rush to the courthouse.

This is the case because the the court system enforces the law with economic incentives in mind. This applies to both judicial and the real economics. Most literature on the subject is focused on long-term incentives. With that in mind, if a ruling would cause one of the biggest investment banks to explode and confidence in all others to evaporate, judge Bransten might still be given pause. Talcott Franklin's innovation was actually a negative to the bond insurers. If Bransten looks out West she can see Bill Gross and his gang perched on the grapevine or the Fed down the street staring right back at her.

Now for the uninsured private labels, securities law provides some distinction from the suits brought by the guarantors. Every 10B-5 dismissal is a win not only for the investment bank/securities underwriters but also the guarantors. If the FG industry can pull in 30B without creating a rush to the courthouse from their winnings, the result is intuitively pleasing in the limited sense of having punished the banks significantly while leaving them standing and restoring the FG industry. If there is little chance of more suits following, exemplary damages could come into play. However, compensatory damages could easily be made large enough to create sufficient deterrence. Start with claims paid and go to legal fees to unearned profits to the cost of financial distress. What will restore a bankrupt Ambac to what it would have been? (Those dollars would flow to the sr. unsecureds if your curious. Quote 13/15 on the 11's. A post on Ambac and implications of ACA Holdings is currently past due. In short, a bet on Ambac Sr. Unsecured is a bet that the ACA/Goldman ruling was (or could be with a probability of about 0.5%) a reliable precedent for CDO fraud.) But as is my way, I have meandered into unbridled speculation.

The point I started to make is that there is room for the costs to grow to investment banks without hurting their prospects. Now, this was acute in especial with BofA. They were under-reserved by roughly half with regard to the GSE's and only (as far as I can make out) have about $500M reserved for insured and uninsured private label (the lion's share of which is probably for Assured.) The $100B discount to book value says the market doesn't buy it, but that is plenty of ground for guarantors to win huge and BofA to prove out. It is also ground in which any greater traffic to the courthouse is manageable for banks and Bransten alike.

GMAC/Ally Settlement

Looking at the Ally/GMAC Mortgage settlement last night, my first impression was very negative for the financial guarantors. A $462 million settlement for $280 B of mortgage exposure didn't seem like much, especially when some journalists incorrectly claimed that the 280B number was outstanding repurchase requests. But a little familiarity with the MBIA and Bank of America situation helps us understand that such a thing could not be the case. Actually, Ally cited in a November 17th presentation

(Ally Financial Inc., 2010 Citi North American Credit Conference, November 17, 2010) that their outstanding repurchase requests were less than their reserves. Bottom line this is bad for B of A, good for financial guarantors... but not as good as it might become. Here's a key slide from Ally's presentation: 

Update on 12/31: it is important to note that the settlement was for more than the reserves. At the end of Q3 Ally had only $218M in outstanding claims from GSE repurchases vs. $632M from Monolines. As a % of UPB (Unpaid Principal Balance) the numbers are not as flattering. For a good report on the (seemingly consensus) viewpoint that this settlement is good for BoA et al. see seeking alpha blog here: http://seekingalpha.com/article/243930-ally-s-fannie-mae-settlement-scales-back-potential-bofa-jpmorgan-putback-costs. However UPB doesn't take into account the quality of the loans sold while reserves are at least supposed to. There is also much greater variability of loan quality and misrepresentation on the private label side. Based off the above slide and the settlement we can surmise that Ally has about $800M in reserves after settling with Fannie and Freddie (in March).

Also, I accidentally deleted the original post in editing it. Still learning. Cheerio.