Fisking the American Securitization Forum's Congressional Testimony
Tom Deutsch, the Executive Director of the American Securitization Forum (ASF) testified before the Senate Banking Committee this past week about chain of title problems in securitization. I was fascinated to see how much attention the ASF spent in attempting to rebut my testimony from a pair of previous hearings (here and, in a more polished form, here). My first thought was "gosh, ASF's awfully defensive. They sure seem spooked." And on looking at the details of the ASF's rebuttal, my sense is they're on very shaky ground if these are the best arguments they have.
Below I review some of the ASF's arguments and show why they're just wrong. In particular, note the PSA language that I quote that demolishes the ASF's claim that PSAs do not require an endorsement from every party in the securitization chain:
"the original Mortgage Note bearing all intervening endorsements showing a complete chain of endorsement from the originator to the last endorsee"
If there's any doubt about what that language means (more discussion below), I'd love to hear it in the comments. There's a very specific method of transfer required in securitization by PSAs and if it wasn't followed, then under New York law, the transfers are void. And it is sure looking like many deals didn't comply with the PSA terms.
- Lack of Caselaw Support
ASF takes me to task because the argument I make about PSAs is not supported by caselaw. Duh. Of course it isn't. These issues have never been litigated. The whole point I've been making is that there are a bunch of unresolved legal issues. I'm not the one who decides what the outcome is. I can only offer my semi-learned opinion. But just as my argument lacks caselaw support, so too does that of the ASF. At least I'm not the one who built a $1.2 trillion dollar private label residential mortgage securitization industry hinging on uncertain law.
- We're Too Big to Fail
The next piece of the ASF argument is that private label residential mortgage securitization is too big to fail. The ASF implies that its argument has to be right because otherwise the private mortgage securitization industry would collapse. News flash: it already has.
The creditors of some of the securitization deals of General Growth Properties tried this argument in bankruptcy court when GGP figured out that the "bankruptcy remote" securitizations could be pulled into the Chapter. The creditors claimed that permitting the filing would kill the whole securitization industry. The argument didn't fly, not least because the industry is pretty dead at the moment.
- PSAs Require Endorsements by Every Party in the Chain of Title
ASF argues that the language in many PSAs requiring a "complete" or "unbroken" chain of endorsements only means that there must be a chain of endorsements legally sufficient to effectuate the transfer of the note to the trust.
Here's what the ASF claims:
The typical language does not state, nor does it imply, that a “complete” or “unbroken” chain means that all prior owners or holders of the note must appear as part of the chain. Nor does any judicial proceeding consider or uphold this novel opinion. Nor does Professor Levitin provide any third-party support for his interpretation of a typical PSA.
The typical PSA requirement for a complete or unbroken chain of indorsements to the person signing the indorsement in blank means only that there be no gaps in the chain of indorsements, and that the chain of indorsements be sufficient to effect a transfer to the trust under applicable law.
There are a few problems with this argument. First, if the ASF is correct in its claim that the loans are transferred by sale under Article 9 of the UCC, the legal sufficiency of the endorsements should simply be irrelevant. In making this claim, ASF seems to be conceding that PSAs are the governing law for RMBS transactions.
Second, it's worth looking at the entire language used in PSAs, not the selectively quoted language referred to by the ASF. For example, consider the PSA for Securities Asset Backed Receivables LLC Trust 2005-FR3, dated July 1, 2005, § 2.01(b), July 1, 2005. It provides that the depositor will deliver to the trust:
“the original Mortgage Note bearing all intervening endorsements showing a complete chain of endorsement from the originator to the last endorsee, endorsed ‘Pay to the order of _____________, without recourse’ and signed (which may be by facsimile signature) in the name of the last endorsee by an authorized officer.”
Note the bold language (my emphasis; the italics are original). There can be no question that this language is calling for every endorsement from the originator to the trust, and cannot be satisfied with a single endorsement in blank. For deals with this language, at least, ASF's testimony is demonstrably wrong.
Now, it is important to note that not every PSA has such language. Some merely require delivery to the trustee of:
"the original Mortgage Note, endorsed in blank or with respect to any lost Mortgage Note, an original Lost Note Affidavit, together with a copy of the related Mortgage Note"
The incidence of various PSA language is unknown, but certainly there are a good number of PSAs where there has to be a complete chain of endorsements. It's really hard to say what is a "typical" PSA, but I can tell you that the language in the Fremont deal shows up in a lot of other deals and that there's a strong implication from that usage that it is also what is called for when PSAs merely require a "complete" or "unbroken" chain of endorsements, without spelling out that it must go from originator to trust.
- There's a Good Evidentiary Reason PSAs Require A Chain of Endorsements
One of the points that I made in my testimony was that there is a good business reason for requiring a complete, unbroken chain of endorsements from originator to trust: it is the evidence that documents the transfers in the securitization that are necessary to show that the assets are bankruptcy remote. ASF argues that there is no need for the complete chain of endorsements as evidence because there are mortgage purchase and sale agreements at every step of the deal prior to the PSA which provide good evidence of the transfers.
There are two problems with this claim. First, in many deals there are only mortgage loan purchase and sale agreements (MLPSAs) going from originator up to the securitization sponsor/seller. The PSA is often the document that covers the transfer from the sponsor/seller to the depositor and from the depositor to the trust. Absent endorsement, there is only a recital showing that the depositor--a wholly owned subsidiary of the sponsor/seller was ever involved in the deal. The presence of the depositor is something demanded by the rating agencies, but absent endorsement, the depositor starts to look completely fictional.
Second, a MLPSA is not conclusive evidence of a sale. MLPSAs rely on Article 9 of the UCC for a legal basis for transferring promissory notes and mortgages. Article 9 requires that the seller have rights in the property being sold for the transfer to be effective. The MLPSA itself provides no evidence that the seller had rights in the mortgages and notes being sold. This is particularly important given the occurrence of warehouse fraud, wherein the same mortgage might be sold multiple times by the same seller.
The mere representation of such rights in the MLPSA is not probative of the seller actually having rights. It merely makes the seller liable for a misstatement, if prosecuted successfully. Ultimately, the evidence needed here would be an affidavit from every party in the chain of title stating that it had good title at the time it transferred the loans. And that sort of affidavit will be hard to get from any party that's gone bankruptcy or ceased operating. The wonderful thing about negotiation as a a method of transfer is that it has evidentiary value baked in. The instrument is the obligation reified, so there is no question about whether it has been previously sold to someone else.
It's worth noting that PSA language pre-2001 (when Article 9 was revised to cover the sale of promissory notes and mortgages) is the same as PSA language post-2001. In other words, everyone contemplated doing deals the same way, irrespective of revised Article 9 because negotiation retained an evidentiary value that revised Article 9 could not supply.
- Can a NY Trust even Accept Paper in Blank?
Another trust law issue not even addressed by the ASF is whether a trustee of a NY trust can accept and hold bearer paper. It's not clear that it can. The problem with bearer paper is that there is nothing that indicates that the paper is the property of a particular trust. If a trustee, say Bank of New York Mellon, is trustee for thousands of trusts, there is nothing that indicates which trust the bearer paper belongs to. If files from different trusts were to get mixed up, there'd be no way to know which trust owned the bearer paper. There's a reason that trustees keep cash in bank accounts under individual trusts' names and don't commingle assets of multiple trusts or hold cash themselves in trusts names. This presents a potential problem for both the notes and the mortgages if one believes that endorsement in blank of a mortgage has any meaning.
- Doesn't address Kemp v. Countrywide
I found it remarkable that the ASF testimony did not address Kemp v. Countrywide. But what could ASF really say? The whole point with Kemp v. Countrywide is that it has nothing to do with the propriety of the legal structure of securitization. It's all about whether that structure was followed--it's an execution issue, and there's no reason that ASF would have any special insight into that. ASF's concern has always been the legal issues in the front-end deal design, not what happened with backroom paperwork.
To summarize: the securitization industry has a lot to be worried about if the ASF's Senate Banking testimony contains the best arguments they can marshall about chain of title problems.