So by now you’ve heard about the SEC fraud case against Goldman involving a synthetic CDO. How weird: once again, as with Climategate, a scandal emerges into the spotlight on which – unlike most who are doing so – I’m actually somewhat qualified to opine. (After all, I’m sort of, um, in that business, technically.) So, it’s a weird feeling.
My initial reaction to reading the surprisingly readable SEC complaint was (1) they probably have the goods on the one Goldman employee (Fabrice Tourre) but (2) Goldman will probably deflect the bulk of the blame towards him and away from them as an institution, and thus (3) Goldman will pay a fine (=Obama 2012 campaign contribution?) that represents a tiny fraction of what they’ve already earned in Q1 2010, and that will be the end of that.
To me the only real uncertainty here is how much commentary full of ignorance and half-truths we are likely to be saddled with in connection with this story. Some people have made going-short into some sort of evil thing (usually these are the same people who don’t like ‘bubbles’, ironically.) Some people think CDS in particular are bad, it’s CDS that are the culprit. Some people think it’s CDOs. Most people don’t know the difference. Some people think CDOs are ok but there’s something evil about ‘synthetic’ CDOs (as if there’s a huge and meaningful difference). Most people (i.e. everyone but me) think there is a huge and meaningful difference. Most people are convinced that this is just another illustration of the (supposed) ‘free market’ needing more ‘reining in’ by the government.
Those people are all wrong. The problem is the government.
Ok, I’m slightly joking. But only slightly so. Let’s get one thing clear: CDOs (especially synthetic ABS CDOs) would not exist if not for the government actions and regulations that incentivized them.
Why does the financial sector create new ‘financial products’? In all cases, it’s because they think those products represent legal ways to make more money for someone (in particular, them, but fundamentally they’re in sales so they only make money if they find ways to make money for other, “real-money” people). Smart people on Wall Street read all the rules, read the fine print, ran the numbers, and realized that if they do XYZ they can achieve more value than people currently get. CDOs are no different and they are nothing special or intrinsically evil. They are a solution to a problem.
Here’s the problem CDOs solve(d): “How do I get a high return from a AAA-rated (or at least, investment-grade) fixed-income instrument?” But the only reason people really wanted (want) to solve this problem is: the government, in one form or another, for one reason or another.
Let’s back up a bit to the first “CDOs”. I claim that the first “CDOs” (at least, in the modern sense…) were actually the mortgage bonds you may have read about in Liar’s Poker. CDO just stands for ‘collateralized debt obligation’, which I scan as something like: repackaged IOU. So just think of a CDO as an IOU based on (collateralized by) other IOU’s. It is just a way to buy a cut of some portfolio of IOUs.
A mortgage is an IOU. You borrow money from the bank and give them a piece of paper with your signature on it. The piece of paper says, in effect, IOU. The bank, which has lots of these pieces of paper, could sit around and just wait for all of them to pay off via everyone’s monthly mortgage checks, but that has risk/uncertainty. If someone came along and offered to pay cash today for all those IOUs, it might be worth it to banks. Problem: individual mortgages themselves aren’t “AAA-rated securities”. Certain types of buyers (who would, otherwise, want those returns) therefore can’t buy them – either aren’t allowed to (because they are, say, an insurance company required by law – i.e. the government – to hold a certain amount of AAA securities), or can’t afford the capital charges they’d be hit with on non-AAA securities (due to government regulations).
Solution: turn mortgages into AAA-rated securities. Pool up mortgage IOUs, and sell off the top 90% as a “AAA” bond. This is what was done in the late ’70s/early ’80s. It was also a “CDO”, technically. It solves the problem of how to sell off mortgage debt to buyers who can’t buy things that aren’t “AAA”. Let me emphasize: this is a problem that would not have existed if not for government actions/regulations. Imagine that no buyers had any legal or regulatory reason to care about or pay attention to a bond’s “rating” in the first place; then the act of cleverly repackaging IOUs to achieve a certain “rating” wouldn’t make any sense. There would be no reason – no financial incentive – to do it at all. It would be a waste of time and money, legal fees and man-hours.
There would be no CDOs or even mortgage bonds, at least, not as we know them. They would not have been overvalued. So there would have been no ‘bubble’ to burst.
Let’s pause for a second to discuss what “AAA” is. “AAA”, the “rating” of a bond, is a thing slapped onto it by some or another private company called a “rating agency”. Another way to describe this, I claim, is that it’s an opinion. A private company has come up with an opinion regarding how risky a bond is, and if their opinion is that it’s not very risky at all, they will call it “AAA”.
So far so good. I mean everyone has the right to their opinion. Who cares? If I, personally came out with a blog post stating my opinion that the bond ABAC 2007 AC1-A1 is “money-good”, would you rush out to buy it as a result? Maybe you would, maybe you wouldn’t, it would depend on how much you valued my opinion. That would be your call. (You probably wouldn’t.)
The problem is that when it comes to rating agencies, the government has elevated their opinion to carry the force of law. Suddenly a “rating” is not just an advisory, like a movie review or some article in Consumer Reports; it has legal and regulatory ramifications for the buyer. Imagine how this would warp any market. Imagine that, by law, you could get better financing on a car that Consumer Reports gave four-stars. Two effects would predictably occur: (1) the price of “four-star” cars would inflate, (2) auto manufacturers would work very hard to try to achieve at minimal cost a rating of “four-stars” for all new models (especially those that would otherwise be on the fence between three and four “stars”…).
Exactly the same thing happened in the debt market. Who the hell is a “rating agency” (some private company) to declare that a bond is “AAA”, and why should anyone care (unless they trusted and believed that rating agency’s analysis, i.e. unless they had done their own analysis too and confirmed that opinion)? Well, the government forces people to care, because the government made “ratings” matter, by law. As a result, buyers (like the German bank who bought ABAC07) end up paying more attention to a “rating” than to anything fundamental. This is a natural, normal, predictable result of the incentives that were in place. The fact that these incentives were in place is entirely the government’s doing and the government’s fault, yet nobody seems to want to blame the government, instead they want to blame rating agencies, or short sellers, or financial firms, all for doing what they are designed to do – read the rules, and figure out the best way to work within them.
Nothing I’ve just written above changes one bit if you move away from mortgage bonds themselves and start talking about “CDOs”. A cash CDO is the same thing (repackaged IOU), except the IOUs inside it are other bonds rather than mortgage loans. Nor does anything change if you speak about ‘synthetic’ CDOs. A ‘synthetic’ CDO just means the IOUs inside it are side bets on bonds rather than the bonds themselves. None of these details have anything to do with the fundamental reason for the bubble, the conflicts of interest, the abuse people are now complaining about, which all stem from one thing: the government incentivizing the achievement of certain “ratings” in the debt market, by making a bond’s “rating” matter in a legal way. This distorted the market by incentivizing people to eschew genuine risk analysis in favor of merely looking at a bond’s “rating”. It created the incentive to game the “rating” system by chasing ways to construct ever-higher-spread (thus, higher-priced and more appealing to the less discerning, rating-sensitive buyers) bonds that were still “AAA”.
Incentives matter. It was the government who designed (intentionally or not) all these incentives. In a very real sense: the government made CDOs. If you don’t like them, and think they’re bad and evil, you now know who to blame.