A common refrain from the financial crisis is that poor disclosure was a big contributor, if not the cause, of the financial crisis. Buyers of even the most complicated financial instruments were misled or were not provided full information concerning their investments. The results were catastrophic when the mortgage market crashed.
The story sounds convenient: investors were deceived! That would imply that all we need to do to prevent a similar problem in the future is to provide better disclosure.
The problem is that when you actually look at the documents from some of the troubled investments during the financial crisis, in many cases the disclosure was copious. There were warnings of the risks; investors just failed to heed the warning signs that should have led them to further investigation. In other words, the disclosure failed to work.
In a new paper, “Limits of Disclosure,†Claire Hill and I examine the types of disclosure that were made be fore the financial crisis. Specifically, we examine disclosure made in connection with the sale of synthetic collateralized debt obligations, or C.D.O.'s, where the reference securities were mortgage-backed securities. These were synthetic bets on the value of mortgage securities with one party taking the long side and the other the short.
The investments had names like Timberwolf and Class V Funding III. The now infamous Abacus C.D.O. promoted by Goldman Sachs was also a synthetic C.D.O. And these products were at the epicenter of the financial crisis. One analysis estimates that asset-backed C.D.O. write-downs alone will be $420 billion, or 65 percent of the original balance, with C.D.O.'s issued in 2007 losing 84 percent of their original value.
In the wake of this colossal failure, allegations have been made that the banks promoting these financial instruments did not disclose that they also had short positions in them. Alternatively, in the Abacus cas e, the allegation was that Goldman allowed John Paulson's hedge fund to hand-select the securities to bet against, thereby creating an investment that was “doomed to fail.â€
But a review of the offering documents for these deals shows that there were ample warning signs, had buyers looked deeper. Take the Abacus C.D.O., for example. The pitch book for the deal stated specifically that Goldman Sachs “shall not have a fiduciary relationship with any investor.†That is, Goldman was not bound to see if the investment was suitable for an investor or to act in investors' best interest.
Not only that, these materials warned investors that they should do their own investigation. Again, the Abacus pitch book stated that “Goldman Sachs may, by virtue of its status as an underwriter, advisor or otherwise, possess or have access to non-publicly available information.†It continued, “Accordingly, this presentation may not contain all information that would be material to the evaluation of the merits and risks of purchasing the Notes.†In other words, Goldman told its customers to do their own investigation and not rely on the firm.
As for allegations that Goldman's trading arm was simultaneously taking a short position in the housing market, there is disclosure on that too. The Abacus offering memorandum stated that “Goldman Sachs is currently and may be from time to time in the future an active participant on both sides of the market and have long or short positions†adding that the firm may have “potential conflicts of interest.â€
Despite the warnings, the evidence is that the buyers of these synthetic collateralized debt obligations did not do a thorough investigation into the securities themselves, let alone follow up on the above disclosure.
The recent S.E.C. case against the Citigroup employee Brian Stoker shows this. The S.E.C. contends that Citigroup had sold another such investment, the Cl ass V Funding III C.D.O., while simultaneously planning to short the security, a fact it did not disclose to buyers. Citigroup settled the action, but Mr. Stoker disputed the allegations.
The largest buyer of Class V Funding III was Ambac, the mortgage-backed security insurer, which was a very sophisticated investor. When David Salz, the Ambac manager who made the decision to invest in this security, was asked at trial whether he had done an investigation of the securities underlying the C.D.O., he claimed that Ambac had not because it had relied on the work of the portfolio selection manager, Credit Suisse Alternative Asset Management.
Yet, the offering memorandum for Class V Funding III stated that the Credit Suisse unit was not acting as “advisors†or “agents†to the buyer, and that any buyer should make its investment decision determine “without reliance†on either. The memorandum further stated that not only could Citigroup and Credit Suisse have conflicts, but also that the firms' “actions may be inconsistent with or adverse to the interests of the Noteholders.†And the offering memorandum had the same disclosure as the Abacus that place the onus on the investors to do their own homework.
All these various offering memos did not even acknowledge that the mortgage market was heading downward. This disclosure taken from Timberwolf C.D.O., a residential mortgage-backed security and another Goldman deal that has resulted in litigation, began to appear in 2007: “Recently the residential mortgage market in the U.S. has experienced a variety of difficulties and changed economic conditions that may adversely affect the performance and market value of R.M.B.S.†It continued: “In addition, in recent months, housing prices and appraisal values in many states have declined or stopped appreciating. A continued decline or expected flattening of those values may result in additional increases in delinquenci es and losses on R.M.B.S. generally.â€
Yet, not only did investors ignore this disclosure, they ignored it despite reading it. At the Class V Funding III trial, Mr. Salz of Ambac was asked at trial about the risk factor disclosure in the Class V Funding III offering memo. Asked if he read it, he replied: “Yes. It's boilerplate language. . . . it was standard language.â€
In other words, Ambac felt comfortable to ignore it because it the language was commonly appearing in documents. Furthermore, Ambac's legal counsel even marked up the offering document and made comments on the offering memorandum.
Ambac lost $300 million on this deal. Mr. Stoker was acquitted by a jury of the civil charges against him.
What is so troubling about all of this is that the investors in these C.D.O.'s were the most sophisticated investors with considerable money - $100 million or more - under management. Class V Funding III's buyers included not only Ambac but als o the Koch brothers and a number of hedge funds.
These were not the “stupid†sophisticated investors that Michael Lewis depicted in his book “The Big Short.†These were investors who should have known that this disclosure should have prompted further inquiry. In particular, these investors knew that for them to take a long position on the C.D.O. there had to be someone on the short side.
So why did these investors make these investments if they did not do their due diligence or even pay real attention to the disclosure? From the testimony given at the Class V Funding III trial, it appears that these investors made macroeconomic bets on housing, following the herd, which thought housing would go up. In this regard, arguments that the securities were too complex to understand don't bear out.
This is a problem. Sophisticated investors are supposed to read the documents. We all know that retail investors don't often take the time to read disclosur e, but the securities laws are based on the idea that information is filtered into the markets through disclosure to sophisticated investors who then set the real price of the security.
This is a form of the efficient market hypothesis. If sophisticated investors can't be bothered to read the documents and act on them, then we have a real gap in the entire disclosure regime and asset pricing generally.
Unfortunately, this is what the evidence from the C.D.O. market before the financial crisis shows. And because of this, the idea that requiring still more, better or clearer disclosure is likely to be unfruitful in many cases.
I have no great solution to this. Until we better understand how sophisticated investors process and read disclosure, regulators should be wary of trying to solve the problem by simply requiring more disclosure.
This post has been revised to reflect the following correction:
Correction: November 2, 2012
An earlier version of this article misstated the name of the financial products that were in part blamed for the financial crisis. They are collateralized debt obligations, not credit-default obligations.
