There’s a new (preliminary) study out by Oliver Faltin-Traeger of Blackrock and Christopher Mayer of Columbia Business School that attempts to explain a key mystery of the financial crisis: why did so many collateralized debt obligations (CDOs) fail so spectacularly? Especially when many of them enjoyed AAA credit ratings from the Big 3 rating agencies?
Remember: it was AIG’s exposure to CDOs (through the naked credit insurance it sold to Goldman Sachs and others) that caused its epic fail and compelled the U.S. to nationalize what used to be the world’s largest insurance company. (And as I mentioned yesterday we are three years down the road and U.S. taxpayers are still waiting to get their $51.1 billion dollars back.)
We now know that CDOs were kind of like a Febreze for putrid mortgages. Through the miracle of financial engineering, mortgages that looked like garbage on a standalone basis were turned into bonds called asset-backed securities (ABS). The ABS might have poor credit ratings to start out with, but scoop up a bunch of them and drop them into a trust and presto! change-o! With the complicity of the major rating agencies the whole thing’s suddenly worthy of a AAA seal of credit approval.
Sounds bonkers in hindsight, doesn’t it? Well it was a $421 billion market at one point. And, like any market, there were people standing on either side of the trade. Meaning: while a bunch of smart people fell for it hook, line and sinker, another bunch of even smarter people were either the ones doing the selling or positioning themselves to profit when the whole thing sunk like a ziggurat in quicksand.
And, as it turns out, some of the folks designing and selling these structures were positioning themselves to profit from their failure without telling the buyers. Again, Goldman. To be perfectly clear: many people suspect that these CDOs were purposely designed to fail. But how can you prove this?
Well, you can start by looking at whether there’s any significant difference between the mortgage-backed bonds that the investment banks chose to package up into CDOs and those that they did not choose.
Faltin-Traeger and Mayer did exactly this in their study called Lemons and CDOs: Why did so many lenders issue poorly performing CDOs?, which was presented at a conference of the American Economic Association (tip o’ the hat to Counterparties.) They looked at 30,000 asset-backed securities underlying various CDO structures and compared them with other ABS that were similar to each other on critical metrics (rating history, initial yield, collateral type, sponsor.) Again, I emphasize that the study is preliminary (the authors would appreciate your feedback!) But they say that their results “confirm that assets included in cash and synthetic CDOs performed extraordinarily poorly relative to seemingly comparable securities that were not included in a CDO.”
As they ponder why this should be, they discuss the potential role of “costly information acquisition”: the idea that, even though all information on the securities underlying the CDOs was available to them investors were simply overwhelmed by the size and complexity of the structures.
Most CDO deals in the dataset used in this paper had at least one hundred underlying securities, each of which was tied to thousands of mortgages or other types of debt (see Figure 3). Because of this complexity, CDO ratings depend critically on a variety of modeling assumptions about the overall performance and the correlation in performance of underlying assets and the extremely complicated division of cash flows between more than a dozen securities that make up a typical cash CDO. Synthetic CDOs often had fewer tranches, but equally complex rules.
The mind-boggling complexity of CDOs has also been addressed in the past by David Fiderer:
CDOs are not like regular mortgage bonds, which may be scrutinized via their initial prospectuses registered with the S.E.C. [Regular mortgage] bonds [...] are structured so that the mortgage pool is essentially fixed at closing. What you see is what you get. Actual bond performance is available, for a price, from ABSNet.
CDOs are different. Everything is concealed. Aside from a relative handful of cases, the public has no access to the initial prospectuses. Even if a CDO prospectus were retrievable through the Irish Stock Exchange, that CDO’s investment portfolio is still likely to be kept secret. Unlike subprime mortgage bonds, these CDOs had no legitimate business purpose. They neither financed the mortgages, which had been financed through the bonds, nor did they add to liquidity in the marketplace, since the CDOs were non-tradable black box investments.
Which is why investors buying up these products were more than happy to rely on rating agencies. Because they couldn’t let a little thing like their own ignorance get in the way of earning higher investment yields when the low-interest rates prevailing at the time made it difficult to earn a satisfactory return. But the rating agencies had crazy incentives to overrate these products.
First of all, they were being paid by the issuers of the securities, OK? That’s the prevailing model in finance. You develop a product and then you pay for the seal of approval that shows the buyer how safe it is. Conflict of interest much?
Then, having rated the underlying ABS, they were usually also involved in the rating of the CDO. And as Faltin-Traeger and Mayer tell us “most agencies treated their own rated ABS in an advantageous fashion when rating a CDO that included collateral rated by competing agencies.” Meaning the agencies’ own confirmation bias is actually built in to the product!
But wait, it gets better! Why not throw a little groupthink in there for good measure:
Even more striking is the extent to which the CDOs themselves were likely to be rated by at least two agencies, and often by all three agencies. S&P and Moody’s were involved in rating the bulk of the CDOs in the database. Clearly buyers of CDOs might have been nervous about ratings shopping and required rating by multiple issuers, a requirement that in the ABS market has been shown to have provided more reliable ratings performance. However, obtaining multiple ratings did not provide stronger protection for buyers of CDOs. Table 4 [of the research report] shows that the CDOs rated by all three rating agencies were, if anything, more likely to be downgraded relative to CDOs that were rated by only two agencies.
So, there’s very good reason for regulators to try and think of ways to make the rating agencies less central to the marketing of structured finance products. Unfortunately, no one seems to have the slightest clue on how to do this.
- Related stories: Truth and Truthiness in TARP; AIG chairman says that you just don’t get it;
- Follow the Finance Addict on Twitter and like our Facebook page.