On Bank of America’s Loan-by-Loan Fight in Putbacks

It’s more than a bit puzzling when readers get upset when once in a great while, we point out how the case against banks on a particular issue is overstated. The reaction seems to be that we’ve suddenly gone soft on financial firm miscreants, which is about as wide of the mark as you can get.

Overhyped attacks on authority (and unfortunately for us all, banks are very much part of the officialdom) backfire. They allow the defenders of the orthodoxy to paint critics as hysterical, reckless, and ill informed. As a result, the best chance for reining in the industry is to make charges that stick.

One area where the case against banks is exaggerated is non-GSE mortgage putback actions. In very simple terms, the servicer on a securitization, on behalf of the trust, has an obligation to “put back”, as in return to the originator, loans that are in violation of the representations and warranties made in the securitization agreement. There is a big difference between putback provisions on GSE paper, where Freddie and Fannie have strong rights to insist on putbacks, versus in so-called non-GSE (aka “private label”) deals, where the investors have very limited authority.

Despite this disparity, the ongoing GSE putbacks have gotten comparatively little attention in the media, while a flurry of cases filed by the monoline insurers (who have fewer obstacles to overcome in these cases than bond investors) and a letter sent to Bank of America by Pimco, BlackRock, and the New York Fed, among others, that appears to be laying the ground for a putback action have gotten a great deal of attention.

As we have stressed, our dim view of these cases is based on the fact that they are difficult and costly to win. Even though it may seem obvious that folks like Countrywide, which is now part of Bank of America, originated boatloads of bad mortgages and clearly harmed investors, that does not mean that this legal theory is a winner.

And the reason we keep harping on this issue is that there are much better legal theories that investors could be pursuing (notably ones related to the apparently widespread failure of mortgage originators to take the steps set forth in their own contracts to convey mortgage loans to the securitization trusts). So to the extent that the media is unwittingly overselling the merits of costly litigation, it may actually wind up serving the banking industry.

The latest report of the Congressional Oversight Panel sets forth the issue (boldfaced ours):

If any of the representations or warranties are breached, and the breach materially and adversely affects the value of a loan, which can be as simple as reducing its market value, the offending loan is to be “put-back” to the sponsor, meaning that the sponsor is required to repurchase the loan for the outstanding principal balance plus any accrued interest.

The problem with these cases is the plaintiff needs to show that it was the failure to adhere to the terms of the agreement regarding loan quality, and not some other cause, like borrower unemployment job loss, death, or disability, that led to investor losses. Broadly-measured unemployment at 17% and housing prices down 25% to 45% are also major causes of investor losses.

Consider this analogy: someone buys a car that unbeknownst to him has a tendency to partial brake failure which produces nasty skids. The manufacturer knew about this defect but kept selling the cars anyhow. The owner gets in an serious accident when he hit his brakes and the car started fishtailing. Seems like a slam dunk, right?

In court, the driver’s lawyer presents the information about the brake defects and the driver’s testimony. But the other side presents forensic evidence: the street was covered with wet leaves. The skid pattern is consistent with correctly performing brakes where one wheel was not getting traction due to the leaves. Oh, and the analysis of the skid marks and the impact suggests the drivers was speeding before he hit the brakes, and investigator ascertained he had had three drinks at a dinner before getting in his car. Does this lawsuit now sound so clear cut?

You have a similar process with these putback cases. One side claims it was the product defect, the rep and warranty breach, that resulted in the losses, while the other side points to other circumstances. And remember, these cases are fought on a loan-by-loan basis. Consider Bank of America CEO Brian Moynihan’s remarks yesterday as reported by Bloomberg:

Bank of America has said it would review claims “loan-by- loan” to protect shareholders as Fannie Mae, bond insurers and private investors press for so-called putbacks.

While it’s a good reflex to assume that anything said by a bank is probably untrue, it does not follow that everything they say is untrue. I’ve spoken at some length to an attorney who is on the anti-bank side of litigation and drafted one of the early putback suits (much of his language has been copied faithfully in the recent suits). While these suits do get settled, they tend to be not hugely lucrative. And he notes that the bond insurers, which have fewer procedural hurdles to overcome than securities investors, are only in the early stages of these actions, which promise to be a long haul.

Now other commentators have a more optimistic reading of these suits. For instance, a few weeks ago, when BofA last got some press for its “well fight them every inch” posture, Barry Ritholtz posted a lengthy extract from some court exchanges between Bank of America and MBIA on a putback suit. Barry read this as undercutting the Bank of America position. Barry and I are generally in agreement on the mortgage front, but I read this material quite differently than he did.

First, BofA and MBIA are still fighting over BofA’s speed (or alleged lack thereof) in getting information on 386,000 loans (the case has been in litigation for over two years). However, MBIA made a major tactical error in not only asking for the files, but asking BofA (Countrywide) to meta tag the data. This gives Countrywide more latitude in their response speed. MBIA also tried to allege that the files submitted so far weren’t all complete, but did it simply based on the length of some files, when Countrywide argued (and the judge agreed) that the issue was completeness, not length, and they hadn’t provided any proof of their beef.

MBIA also argued that they should be permitted to make their case based on a sample of the files, not on every loan. The judge seemed sympathetic, but said they needed to provide a methodology, which MBIA has yet to do. And of course, both sides will fight over what sampling methodology is appropriate.

Now consider: when these loans in these deals were reviewed by firms like Clayton Holdings prior to issuance, they’d typically look at 25%. So there is industry precedent for a sample being very large. Even if MBIA manages to whittle a sample down to 10%, how long is it going to take to work through why the defaulting borrowers in a sample of 38,000? If you assume a 30% default rate, it still adds up to an case by case analysis of 11,400 borrowers.

I hate to be the bearer of bad tidings, since I really do want the banks to get their comeuppance. But the real question is why did investors sign up for transactions that gave them such weak recourse in the case of originator fraud? Perversely, the banking industry became so careless that investors do have other avenues for recovery, as we discuss in our post on the Congressional Oversight Panel report today.

Print Friendly, PDF & Email

6 comments

  1. anon48

    Agree wholeheartedly with your premise that when making an argument on behalf of the cause, make sure the facts are rock solid and arguments are sound so that conclusions drawn become impenetrable. Otherwise, opponents can use flaws to taint the entire position as being not credible. Or as you put it so well “… Overhyped attacks on authority (and unfortunately for us all, banks are very much part of the officialdom) backfire. They allow the defenders of the orthodoxy to paint critics as hysterical, reckless, and ill informed. As a result, the best chance for reining in the industry is to make charges that stick.”

    “I hate to be the bearer of bad tidings, since I really do want the banks to get their comeuppance. But the real question is why did investors sign up for transactions that gave them such weak recourse in the case of originator fraud?…”

    Can’t prove it yet, but I bet that it was because some of the Fiduciaries responsible for overseeing the investment decisions of the various institutional investors (e.g. pension funds, university endowments, etc.) were either lazy, incompetent or incentivized to go with the flow. Before the dust completely settles on the securitization mess, these folks should be penalized. Personal experience with two separate defined benefit pension funds, the level of in-house financial sophisication was minimal. There was an over-reliance on outside managers by the governing board.

    The State AG’s should pursue the fiduciaries and investigate the level of due diligence that was observed by the institutions before making the investments. If determined to be negligent in fulfilling their responsibilities, and if they were compensated for their services, these people should be sued, fined, whatever so they have to pay back at least 100% of what they made. Negative incentives are sorely needed.

  2. dejavuagain

    Perhaps is is simply an application of Gresham’s law. How could an advisor to a pension plan not invest in CDO’s rated AAA by S&P when the return on this “junk prime” investment was higher than other investments. This is one illustration of why laissez faire regulation kills the entire goose.

    Or it could be just Greed by investors, and pension plans are not immunized from Greed. See, Bernie M.’s lack of pity as to his investors!

    Remember when ERISA rules were supposed to protect pension plans? All that law ultimately accomplished, it seems, was to permit creative judges to claim that employee sponsored benefit plans could not be regulated by states since ERISA preempted the states. But, the ERISA covered plans could invest in anything rated AAA by S&P.

  3. Fractal

    Yves, this is very, very helpful. You and Marcy Wheeler and David Dayen over at Fire Dog Lake (FDL) are the best on the net at drilling into the exact terms of the instruments that will be the downfall of the criminal banks.

    What a surprise that the GSEs protected themselves better than the fat-cat bankers did from low-quality mortgages! You emphasize the contractual protections included in agreements created by Fannie & Freddie and the absence of such contractual protections in the private-label instruments. I deduce (perhaps wrongly) that the key protection in the GSE instruments was that the GSEs have no duty to prove that “the breach materially and adversely affects the value of a loan.”

    IOW, if the GSEs prove a “breach,” they can force a put-back, but the private-label mortgage buyers must prove, IN ADDITION, that the breach was a “material” cause for the loss in value.

    If that is your analysis, it is fundamentally based on contract law principals, which can be very powerful. The GSEs simply negotiated better contract terms. (Proof of materiality is a financial contract law parallel to proximate cause in tort law.)

    In your last graf you asked “why did investors sign up for transactions that gave them such weak recourse in the case of originator fraud?” But originator “breaches” of contractual obligations are not per se “fraud,” even if the “breaches” are proved to “materially and adversely affect the value of the loans.” Fraud is a separate beast, whether civil or criminal, and is proved by state of mind (whether through circumstantial or direct evidence), not by breaches of contract obligations. Fraud is proved by extent of knowledge and failure to disclose knowledge, not by failure to perform contractual obligations. Contract breaches might RESULT from fraud, but fraud requires a separate analysis, whether at common law or under statute.

  4. Fractal

    It is shocking to see how blatantly the MSM is stroking & fluffing the criminal banks. NYT ran this story about how hard BAC is working to help homeowners facing foreclosure, replete with fulsome details of how one of its executives planned to testify to Senate Banking:

    http://www.nytimes.com/2010/11/16/business/economy/16foreclose.html?_r=1

    Then NYT’s Dealbook ran this fairly even-handed look at claims by the American Securitization Forum that MERS is lawful even though it blatantly disregards centuries of common law governing real property:

    http://dealbook.nytimes.com/2010/11/16/a-hearing-on-foreclosures-and-an-industry-at-stake/

    That particular Dealbook entry, unlike most, actually quoted a big chunk of the warning of systemic risk published by the Congressional Oversight Panel, and ended with a snarky question about the long history of bad behavior by the securitization industry, linking to Michael Lewis’s book The Big Short.

    But, still, not a word in NYT about what actually happened during the Senate Banking hearing yesterday.

    Bloomberg’s coverage is, if possible, even worse. It pretended to report on the Senate Banking hearing, but completely ignored the two witnesses who testified most cogently about the endemic fraud and homeowner abuse committed by the servicers:

    http://www.bloomberg.com/news/2010-11-17/u-s-banks-are-balancing-competing-foreclosure-interests-lawmakers-told.html

  5. kravitz

    There’s no mention of yesterday’s session, or the AG story, in The New York Times. I wonder who’s pressing them (or maybe helping them with their own debts) to keep this out of the paper?

  6. Greg

    It’s too bad that the lawyers and courts aren’t up to speed on using statistical sampling to estimate these losses. A reasonable sample size depends mostly on the variability of the losses within the files and the desired margin of error. Sampling tens of thousands of files is almost certainly overkill. Parties interested in a quick and fair settlement should be satisfied with a 10% margin of error. You should be able to make such loss estimates with sample sizes in the hundreds – regardless of the number of loans in the files.

Comments are closed.