skip to main |
skip to sidebar
As I've mentioned in the past, one of the hazards of derivatives transactions not executed through a clearinghouse is that the parties are responsible for policing each others credit. Each party must satisfy itself that its counterparty will be able to execute the transaction without defaulting.
In the 1980's parties began trying to control credit risk by taking collateral from each other. Basically, when one party is in a more tenuous financial position than the other, the shakier counterparty will pass some collateral (usually cash) to the stronger party as a guarantee. In its 1996 report, the ISDA Collateral Working Group identified three ways that collateral makes derivatives more attractive: it equalizes disparities in creditworthiness thus opening up a wider range of potential counterparties, it can reduce the interest charged to less creditworthy counterparties, and it may help regulated financial institutions reduce their capital requirements by lowering the risk weighting associated with the transaction.
OTC derivatives executed via the ISDA Master Agreement are structured using one of three ISDA credit support annexes to the Master Agreement schedule. All three establish the same framework for the operation of the collateral transfer - how collateral calls are calculated, when collateral can be substituted or exchanged, how disputes will be resolved, etc.
There are three of them because each is governed by the law of a different jurisdiction. The 1994 Credit Support Annex (security interest - New York Law) and the 2008 Credit Support Annex (Loan/Japanese pledge) create a lien on the collateral while the 1995 Credit Support Annex (Transfer - English Law) actually transfers ownership of the collateral to the other party. Each of the annexes has an explanatory "users guide" and the 1994 and 1995 annexes have standard conforming amendments to bring them out of the dark ages of the 1992 Master Agreement.
These collateral support agreements, of course, are what got AIG in trouble. Each time their rating was downgraded they found themselves facing many CDS-related collateral calls.
Subtitle C is the weirdest piece of the administration's proposed Investor Protection Act of 2009 because a lot of it is a spanking for the SEC.
It has been nearly 5 years since Congress gave the SEC explicit regulatory oversight of credit rating agencies, but the agency has had trouble imposing any control. Despite being implicated in the Enron collapse and, of course, in the current financial unpleasantness, credit rating agencies have managed to avoid any substantive oversight. They have evaded regulation of the content of their ratings by arguing that ratings are opinions and therefore protected speech under the First Amendment. They are free from public disclosure obligations, beyond form NRSRO, because they have convinced regulators that secrecy is essential to their business.
This then, is why section 932 of Subtitle C requires that, "The Commission shall establish an office that administers the rules ... with respect to the practices of nationally recognized statistical rating organizations," and why the Commission is directed to "conduct reviews required by this paragraph no less frequently than annually," and to make "[a] report summarizing the key findings of the reviews ... available to the public in a widely discernible format." Most embarrassingly, section 936 orders the Comptroller General to write a report assessing "the extent to which the rulemaking of the Securities and Exchange Commission has carried out the provisions of this Act." Ouch.
Subtitle C also imposes new obligations on credit rating agencies. For starters, they must promulgate a written conflict-of-interest policy and elect a Chief Compliance Officer to police same. For more see this one-page summary from Morrison & Forester.
The SEC has a slew of rules out that cover much of the same territory. It isn't clear whether Subtitle C and the SEC's proposal are coordinated. The SEC has "deferred" its plan to reduce the reliance placed on the NRSRO classification. Subtitle C would put this program where it belongs - with the President's Working Group.
Today, I'm going to start looking at a neglected piece of the administration's reform proposal: title IX, the vast Investor Protection Act of 2009. The IPA is a grabbag of regulatory off-cuts ranging from asset-backed securitization reform to credit rating agency reform.
The centerpiece of Subtitle E, called "Improvements to the Asset-Backed Securitization Process," is the so-called "skin-in-the-game" regulatory fix. The idea is that the present securitization process provides no incentive for deal sponsors to create good-quality securities. The way sponsors make money is by getting fees. They pay themselves a fee for putting together the pool, for underwriting, and even for servicing the underlying debt. They don't invest. The sponsor is like a chef who won't eat at his own restaurant. The thinking is that if he has to eat what he peddles, his restaurant will improve.
So, section 952 of the IPA would require "securitizers" of ABS deals to retain 5% of the "risk." Securitizers would not be allowed to hedge this retained risk and the SEC would establish standards for the risk's "permissable forms" and "minimum duration."
The new law defines "securitizer" as an issuer or an underwriter. It cleverly hops right over the amorphous matter of trying to define "sponsor" and lands right where the sponsor gets its money - underwriting.
Of course, the ABS packagers, especially the residential mortgage monsters like Lehman and Bear Stearns *were* eating in their own restaurant. In fact, they were eating the leftovers (in the form of the lowest tranches of their offerings), and so, there's a funny sort of conversation that's been going on about whether the ABS sponsors were really avoiding the securities they created. What this argument boils down to is a disagreement about how stupid they were. Were they smart enough to know their ABS deals were crappy, but too stupid to get out in time, or were they completely oblivious to how risky these deals were? Skin-in-the-Game is only an effective deterrent if the ABS packagers are just a little stupid.
Subtitle E also gives the SEC power to create a disclosure obligation for ABS issuers that could not be extinguished by de-registering under section 15 and presumably would continue until the breaking of the world.
For a complete overview, see this Cadwalader memo and wonder along with them (and me) about why the law repeals s. 4(5) of the '33 Act.
I admit that I am still agog when I look at the residential mortgage-backed securities transactions from right before the crash. I know it sounds like I am, once again, excavating things best left interred, but that is not the case. Last week, Fitch gave a triple A rating to a new security called "JP Morgan Re-securitization Trust 2009-R10." The Fitch news release mentions that among the securities being "re-securitized" is "a 25% interest in Lehman Mortgage Trust 2007-7, class 6-A-4." "And what's that?" I wondered, you know, to myself (I'm a blogger so, I am alone).
Then, I went on Westlaw Business and found the prospectus. LMT 2007-7 was an $800 million pool of about 2,000 residential mortgages. It issued 19 different classes of securities. The securities were narrowly sliced to reflect specific parts of the large pool. First,the big pool was subdivided into three smaller pools based on the quality of the underwriting standards. 77% of the mortgages in Pool 3, composed mostly of mortgages originated by Lehman's banking sub Lehman Brothers Bank, were "no-doc" loans.
The Pools were further subdivided into "collateral groups" by interest rate. Series 6-A4 was paid out of collateral group 6, a collection of 700-or-so mortgages with a weighted average interest rate of 7.5%. The "A" and the "4" indicate payment priority - "A" securities got paid first, but within "A" 1 got paid before 2, or 4. Further confusing the payment priority picture, the 6-A4 securities are also described as "super senior."
LMT 2007-7 paid right on schedule until December of 2007-7 when it filed a form 15 to withdraw its registration on the grounds that it was held by fewer then 40 people.
Virtually all the LMT 2007-7 securities were initially rated triple A by S&P. When I checked their rating last week, they were all rated B+ or lower. When I checked today, the ratings were gone.
Is this what we're going to get instead of PPIP?
Please read the gloriously deadpan Death Plays in Barron's about my favorite new securitization. "We admit" says Alan Abelson "to a nagging concern or two."
As I mentioned in the last post in this series, in 2002 ISDA revised its Master Agreement to create a better way of determining who owes what to whom when an agreement goes into default. The 1992 Master Agreement provided two methods (with the creative monikers "First Method" and "Second Method") both of which proved unsatisfactory during bad market conditions. The 2002 Master Agreement replaced these mechanisms with a much more flexible method called "close-out amount."
Unfortunately, the 2002 Master Agreement was adopted only gradually. In a March memo, Manuel Frey and Jordan Yarett of Paul Weiss observed that, "since its introduction, the 2002 ISDA Master Agreement has become increasingly common in the market place, although the 1992 ISDA Agreement continues to be widely used." In a January article in Butterworths Journal of International Banking and Finance Law, Edmund Parker and Aaron McGarry second that: "Many Master Agreements still in use are based on the 1992 version, which contains the greatest weaknesses."
In addition to an understandable unwillingness to spend money amending thousands of contracts, reluctance to adopt the new Master Agreement also stems from the need to very closely match a hedge to the transaction being hedged. The fear is that hedging a transaction written on 1992 Master Agreement with a transaction written on the 2002 Master Agreement might cancel the value of the hedge.
The collapse of Lehman Brothers, with its 8,000 Master Agreements and 67,000 outstanding transactions, pointed out the folly of being so conservative (or lazy) and since then, there as been a concerted effort to ditch the 1992 settlement procedures. In August of 2008, most of the large OTC derivatives dealers signed the Close-out Multiparty Agreement which bulk-updated their existing 1992 Agreements to conform with the 2002 Agreement's close-out protocol. In February of 2009, ISDA published the Close-Out Amount Protocol, a standard-form version of the Multiparty Agreement. ISDA maintains a list of adherents to the Protocol.
Corporate Counsel blog on the SEC Inspector General's report on La Madoff. *sigh*
From Forbes (via CompliancEx) a call for securities arbitrators to explain their decisions. Here, here!
From the New Brunswick Business Journal (with thanks to Compliance Week for pointing it out) - a Canadian SEC? Egad.
At the center of the ISDA documentation architecture is the Master Agreement. The Master Agreement began life in 1985 as the Code of Standard Wording, Assumptions and Provisions for Swaps (it spells SWAPS - how cute is that?) and matured into its present acronym-free iteration as the 1993 ISDA Master Agreement.
The Master Agreement is made up of three discrete pieces - the printed form, the schedule, and any subsequent confirmations. The printed form and the schedule lay out the mechanics of the transaction. The printed form is a standard recitation of parties, addresses and other routine information - it is not generally amended. The schedule is a mechanism for customizing the printed form. It gives the parties the option of adding customized langauge and turning on or off some of the printed form's provisions.
The confirmation is a standard practice that predates the Master Agreement. The Master Agreement sets out general operating principles for a proposed derivatives transactions, but the transactions don't actually occur until a confirmation is sent. The confirmation contains the specific monetary terms of the transaction.
The 1993 Master Agreement provides two methods for determining payment in the event of default. Parties must choose one. During the Japanese banking crisis in the 1990's both mechanisms failed to provide equitable settlements. As a result, the ISDA revised the Master Agreement to provide a broader settlement mechanism. The new mechanism is the primary difference between the 1993 Master Agreement and the 2002 Master Agreement.
For a very thorough discussion of both Master Agreements, see 1397 PLI/Corp 51, Klein, Overview of the ISDA Master Agreement Forms.
Part 3: LEHMAN
Part 1
When he planned to steal our sunlight, he crossed that line between everyday villainy and cartoonish super-villainy.
- Smithers
Before I read today's New York Times, I felt slightly superior to people who ascribed evil motives to, for instance, Goldman Sachs. Business cares about making money - it doesn't make moral judgments. But, as today's Times article proves, there's no practical difference between being amoral and being immoral.
Some Wall Street banks, apparently, are planning to securitize viatical settlements thereby creating a way to invest in the probability that other people will die. Presumably, you could even speculate on the likelihood of your own death.
What if your pension fund were to invest? The risk factors should be a hoot, too - "in the event that AIDS is cured, you could lose your entire investment."
1995 ISDA Credit Support Annex
1995 ISDA Credit Support Deed (Security Interest – English Law)
2008 ISDA Credit Support Annex (Loan/Japanese Pledge)
1994 ISDA Credit Support Annex (Security Interest - New York Law)
The first time I was asked to pull one of the components of the International Swaps and Derivatives Association's "documentation architecture" I was pretty intimidated. I spent a few minutes in that forest of teensy, similarly-named pamphlets, chose what I thought was the right one and then, shortly, found myself back at the shelf taking a closer look because I'd pulled the wrong document. Eventually, I became more adept at navigating the ISDA waters, but I always felt a little at sea.
To understand why the ISDA documentation architecture is confusing (aside from the irritatingly similar names), I think it is helpful to remember that derivatives were devised as a hedging tool. A hedge is a way of protecting an investment against a worst-case scenario. For instance, if you owned a store in a seaside town and anticipated a hot, sunny summer you might buy more sunglasses than usual. If you're wrong and it rains, you won't make any money from your investment in shades. So, to hedge against that possibility you might buy umbrellas. If the summer is as you expect, you'll make lots of money, but if it rains every day, you can limp along selling umbrellas.
To be useful as a hedge your umbrella purchase must be precisely related to your investment in sunglasses. Hedging transactions are always associated with a primary investment and must be narrowly tailored to protect against a disaster without negating the primary investment's value. Thus, hedging transactions are always customized to accommodate the needs of both counter-parties.
If you are reluctant to get into the umbrella business, there are other ways to hedge your risky sunglasses play - you could invest in an umbrella company, short a sunglasses manufacturer, or you might want to try something more exotic: like investing in the price of umbrellas. That's where derivatives come in - they were developed as a way to hedge by investing in intangibles, like the price of wheat.
The ISDA documentation architecture was developed to institutionalize and standardize the process of writing these highly customized contracts. So when, in 1980's, the ISDA started working on a set of standard forms for derivatives transactions, they were faced with a daunting task: how do you create a standard agreement for an industry where every agreement is different?
Part 2: THE ISDA MASTER AGREEMENT
