More on the Peculiar Pimco, BlackRock, New York Fed Putback Letter to Countrywide

Readers may find it odd that I keep returning to the matter of the widely touted letter last week signed by investors Pimco et al pushing Countrywide as servicer to put back loans on some 115 mortgage securitizations totaling $47 billion, of which the letter-writers holdings represent roughly $16.5 billion. The big reason is that this letter, which is a procedural first step which may be laying the ground for litigation, is likely to mean far less than the media reaction would lead you to believe. We did a quick and dirty analysis last week that showed that even if this effort succeeds, the recovery amount is likely to be far less than is widely anticipated.

Aside from the overhyping of this particular situation, we have suddenly seen a shift in sentiment where banks and servicers are suddenly seen as vulnerable, and all sorts of litigation is being filed. The problem is that a lot of reporters tend to treat all cases as having equally good potential, when some have good foundations in underlying legal theories and fact sets, while others are a stretch. This particular possible future legal action falls into the latter category.

Let’s dig into a couple of issues. First is that the letter claims that Countrywide (as servicer) has been remiss in not putting back loans. The argument is that there have been underwriting breaches, that loans were put into the trust that were in violation of the applicable representations and warranties when they were purchased by the securitization entities, and that Countrywide had a duty both to notify the investors of such breaches and to put back the loans to the originator.

Now let’s look at simple way this might play out: what if Countrywide simply denies their are breaches, or ignores the letter? What happens next?

The parties would then need to sue. They would argue that their have been breaches, that Countrywide should have put back loans, that it is in violation of its servicing agreement. The intent would presumably to force Countrywide to comply or to fire Countrywide. But how meaningful a threat is that, really? From MBSGuy:

If holders of the certificates representing a majority or supermajority, depending on the provision, vote together, they can potentially terminate a servicer or subservicer or a trustee.

Yves here. Note this group does not constitute a majority across all the bonds, but it might on selected issues, so the ability to terminate is questionable. But let’s go to the next part from him:

But what would be the point? Removing either of these entities would not fix the issue of conveyance of the mortgage loans and would not remove any liability that these parties might have. In addition, what replacement would be better? All of the servicers and trustees are implicated in this mess. As a result, it is really an empty threat.

On deals I was involved in, we would use the threat of terminating the servicer to try to compel them to correct their actions. It was very hard to get a new servicer to step in and take over a severely distressed transaction because it is a negative cash flow business. They would only do so if they could be paid in excess of the contractual servicing fee. In that event – where will the money come from? The only way to get it from the trust is to amend the agreement, which requires the consent of a super majority of the bond holders or the bond insurer. Even if you accomplished it, you would likely face litigation from adversely affected bond holders. Despite all of this, bond insurers have special contractual rights and can terminate servicing without bond holder consent.

The one time we did this, the new servicer really didn’t improve anything. The only way it really works is if you have a third party servicer on board at closing who contractually agrees to take over servicing at a set fee.

Yves here. So the only way to remove the servicer, if you aren’t a bond insurer and don’t have a majority or supermajority, as the case may be, is to sue for contract violation. But here the problem is circular: the plaintiff say there has been breaches, the servicer says no. This is a he-said, she said, unless the plaintiffs can provide make a credible enough case that there must have been breaches ex having access to the loan files. If the plaintiffs can’t allege specific breaches, the odds are decent a judge will not permit discovery and will dismiss the case.

But some readers will say, aha, but didn’t Clayton just tell the FCIC that there were all sorts of violations of reps and warranties in the deals it reviewed in 2006 and 2007? ,True, but were any of these reports on Countrywide deals? There is reason to suspect not. Clayton was used most often on deals packaged by investment banks, where they were acquiring many of their mortgages from smaller banks and mortgage brokers. So the check by Clayton provided some reassurance to rating agencies. By contrast, Countrywide sourced all its paper from its own operation, and many investors looked more favorably upon loans that were sourced by banks than by diffuse networks. So it isn’t clear that Clayton would have reviewed any Countrywide deals.

In addition, as we have stressed, it isn’t clear whether the investors will be able to prove that delinquencies were the result of underwriting failings, as opposed to job loss, death, or disability. As we indicated, this is normally argued on a loan-by-loan basis, which makes these cases absurdly expensive to pursue, hence they are usually settled for not very juicy amounts. Barry Ritholtz points to some testimony which suggests in the MBIA lawsuit against Countrywide on similar issues, the judge seemed inclined to look at a sample rather than all 384,000 loans, but the two sides have yet to agree on a sampling methodology. And the loans within that sample will be reviewed individually.

Before you argue, “Surely, it’s common sense, the fact that the loans were bad had to have made a difference,” let’s consider a hypothetical conversation. Countrywide calls the risk manager for one of the investors to the stand:

Laywer: You were in charge of risk management during 2006 and 2007, is that correct?

RM: Yes

Lawyer: Did that include the development and use of models to estimate losses on subprime loan pools and bonds?

RM: [Some detail about nature of job and who did what, but answer is effectively yes]

Laywer: What was the worst case scenario that you modeled for housing price declines and unemployment?

RM: [After much hemming and hawing] Housing prices flat and unemployment at 6.5%

Lawyer: What level of losses do you think your model would have shown with unemployment at 10% and housing prices falling 25-45%, with the most severe declines in the biggest markets?

RM: [After vigorous efforts not to answer the question] Over 35%, maybe over 40%

Lawyer: And what losses are projected for subprime mortgage pools?

RM: 40%.

Let’s consider a few more factors that will complicate the plaintiffs’ case (assuming things get that far).

The mortgage securitization pipeline was screaming for product from 2005 onward. In particular, the demand was for “spready” loans, meaning bad ones to feed the demand for CDOs, which in turn was driven by subprime short like Magnetar (see ECONNED for details).

There may be incriminating evidence at the investors re the quality of these deals. If the investors bought the paper knowing the loans were worse than advertised, how can they claim to be victims of the servicer?

The bond insurers had the right to review the loan files at the time of the deal. It’s a bit hard for them to call foul now when they dropped the ball then. Although the Pimco/BlackRock/Fed action does not depend on the bond insurer outcome, the bond insurer agreements give them stronger grounds for litigation than investors.

There are also procedural hurdles to overcome. Per reader Fred:

Not only is standing a huge issue, but class certification is very problematic. The defense will argue that each underlying putback should be reviewed individually on its own and that it is not a class action. This could well be true since each loan needs to be looked at individually to determine the FICO score and individual misrepresentation.

Also major problem with choice of law in which state. Can’t use same foreclosure law uniformly throughout the United States since the laws are different in each state.

Certainly venue will be a key factor. It will be Federal Court but where is not yet known.

If case gets beyond standing, class issues and choice of law issue the case will drag on for years through the appellate courts.

Of interesting note is the firm that is representing the plaintiffs is a Houston firm with little mortgage experience. Not your typical NY Blue Blood firm, that may understand the huge procedural hurdles.

We are no fans of Countrywide, but that should not stand in the way of recognizing that not every legal case against them is necessarily a slam dunk.

We’ll discuss soon, hopefully this evening, possible reasons for the NY Fed signing up for this adventure.


Originally published at naked capitalism and reproduced here with permission.