I've just finished Diary of a Very Bad Year: Confessions of an Anonymous Hedge Fund Manager, which I highly recommend. It's a detailed and mostly accessible discussion of the financial crisis, the reasons it happened, and the fallout, as told by an anonymous hedge fund manager. Along the way, you pick up lots of information about how our financial system works.
People tend to be interested in stories about unseen large things in our society. Sex slave trade, for instance, is a topic of countless pieces that I've seen. Well, the financial sector is probably the hugest thing that ever existed on the Earth, in monetary terms, and people are mostly unaware of it. Derivatives trading alone reached, in 2010, something like 700 trillion dollars (the entire GDP of the US is a mere 13 trillion).
Some things I was struck by:
Part of what was amazing about the financial collapse was that it was happening because of computers. On days when the markets crashed in late 2007, no actual humans were involved. Instead, many financial organizations, banks and hedge funds alike, have "black boxes", computers that are programmed to make trades in split seconds. As the anonymous hedge fund manager [HFM] explains, the machines represent "guys with a lot of physics and hard-core statistics backgrounds who come up with ideas about models that might lead to excess return, and then they test them, and then basically all these models get incorporated into a bigger system that trades stocks in automated way. ... The problem is that the DNA of a lot of these models is very, very similar, it's like an ecosystem with no biodiversity, because most of the people who do stat arb can trace their ... intellectual lineage back to four or five guys who really started the whole black box trading discipline." This causes a cascade: one system starts selling in response to a real event, and the other black boxes look at the aggregate trades and sell because there are tons of other people (actually just one black box) are selling. This makes it VERY easy for a market crash to happen.
Another thing that was familiar to a scientist was the idea of a dominant paradigm in an organization (the way everyone thinks). Nowhere was this more apparent that in subprime mortgages.
[T]he person who was an expert, the person who ran the subprime business, who traded subprime paper and issued CDOs, he was a true believer in the paradigm ... If you have somebody who's really trained in the mortgage business, he's been in the mortgage business for 15 yeras, in equilibrium, he'll do a great job. But in terms of detecting the paradigm shift ... he's not going to catch [it]. ... I see the remittance reports every month, I've been involved in the 2003 subprime issuance ... but it's performed very well. And I have all the details. You have anecdotes? I have details.
Basically, the subprime guys had been doing subprime for a decade. They looked at quarterly reports, and did not see anything that would indicate a collapse. Yet the people who weren't mortgage guys thought something was wrong. Who wins? The guy with the most experience, of course! But they were wrong.
Conventional wisdom, at least among people who I talk to, is that the SEC is a really sophisticated organization. If you did a deal that was inside, like Martha Stewart, they got you, they always knew. But they didn't catch any of the stuff that happened in the run-up to the collapse.
The SEC is an organization of lawyers---they understand the law very well---but they're at a loss when it comes to understanding actual market behavior. The Madoff thing is a good example. The SEC went and looked at Madoff Securities, and a lot of people have been very critical of the woman at the New York office who was responsible for the investigation ... and I'm sure she knew the securities laws back and forth, but what was needed there was not someone who understood the securities laws, it was somebody who really understood how brokerages work.
Lots of people at banks and funds knew Madoff was dirty, but the SEC as a watchdog is preoccupied with law; they don't have the expertise to detect stuff going on off the books. There are no "hunches" that police can go on, because these cops weren't out on the street. If you let the DAs do the investigation, you might not expect them to catch too many people---they're too divorced from what's going on.
In all of this discussion of the evils of novel financial instruments and over-leveraged funds, I think it's easy to lose sight of the fact that hedge funds perform some really important functions in the markets. Suppose there are two companies in an industry, of roughly equal quality, but one is very overvalued compared to the other. This could be because of reputation of the brand, or the fact that some of the management wasn't well known, or whatever. This is actually really bad for the underdog company; it significantly influences their ability to raise money and do their business. And hedge funds actually help to balance this out. They bet against (short) the big company and bet in favor (long) the smaller one. And this influences the market; things tend to even out. The hedge fund has to do a ton of research into the industry to make this kind of informed bet, which is logical, since the value inequality was caused by lack of information in the first place.
Hedge funds also allow businesses like airlines to stabilize their business in the event of shocks. If Jet Blue is worried about a sudden shock in fuel prices, they have a legitimate interest in buying insurance against that. They hedge their risk with somebody who is willing to accept their money now in return for a possible payout later.
The financial organizations forgot the lessons of the crisis very quickly. When a deal is done, a certain amount of collateral is required from the counterparty. Before the crisis, "deals where if we wrote a million dollars in protection they used to take $30,000 or $40,000 in margin". But following the crisis "they were now asking for $200,000 or $250,000 in margin---in other words 25 percent of the exposure". Good---this makes sense, because the banks started to realize that the risk of a given security going belly-up was much higher than their old evaluations suggested. Except, 7 months after the crisis, the rates went back to where they were! Risk was already being wrongly assessed again. "It's funny how quickly people forget."
Finally, the idea that investment banks and hedge funds operate in a largely lawless environment is basically indisputable. Contracts are broken all the time, because there is no ability to litigate when somebody decides to screw you. One deal "was an arrangement where the terms of that financing couldn't be changed except with six months' notice, and they just called up one day and said, 'We're changing the requirements, we're changing our margin requirements, we're multiplying them by 3 or 4'." When asked if he could sue them, the HFM replied "By the time this thing wends its way through the courts, you're out of business." HFM also talks about many other times where people who had liability simply didn't pay. And there's no recourse.