You simply must make room on your financial crisis porn bookshelf for the forthcoming All the Devils are Here, by Bethany McLean and Joe Nocera.
The worst financial crisis books, such as David Wessel's undeservedly popular tripe, focus their narrative on 2008, when the crisis broke. In contrast, Devils goes back to the 1970's and 1980's, with only the brief last chapter dealing with the events of 2008. After that, the authors give a brief epilogue, which is the only weak chapter in the book. They are journalists who handle the financial narrative deftly, but they are not in a position to step back and really draw the deeper lessons of what they so competently describe.
Given my emphasis on what I call the Suits vs. Geeks divide, Devils had me at the prologue. It describes a meeting in September of 2007 in which John Breit, a low-level, marginalized geek at Merrill Lynch, gives CEO Stan O'Neal his first clue about the depth of Merrill's exposure to subprime mortgage credit risk.
The rest of my review is below the fold.
Devils starts with the origins of mortgage securitization. It quite aptly depicts the triangular relationship among Freddie/Fannie, Wall Street, and the government. The relationship was characterized by mutual suspicion and often by one party seeking to intimidate another into submission. But always in the end, they worked together because of mutual need.
For example, on p. 17 we see Fannie Mae's Chairman David Maxwell in 1988 anxious to get a favorable ruling from the Department of Housing and Urban Development to issue REMICs, a tax-advantaged form of mortgage security that initially neither Wall Street nor the government wanted Fannie to be able to issue.
How did Fannie Mae persuade Pierce to rule in its favor? Not be sweet-talking, that's for sure; Maxwell had an iron fist inside that velvet glove of his. "We essentially gutted some of HUD's control over us in a bill that passed the House housing subcommittee," Maloni says today. In that bill, HUD's ability to approve new programs was revoked. HUD went to Fannie, and essentially pleaded for mercy. "In return for asking the Congress to drop the provision, HUD approved Fannie as issuers," says Maloni.
Maloni also called Lou Nevins and told him that if Salomon didn't back off, Fannie wouldn't do business with the bank anymore...This was a major threat. "It's like the post office saying we won't deliver your mail!" Nevins says. He remembers thinking to himself, "If they get away with this, there won't be a private company in the world that will stand up to them."
There are many more examples of the arrogance and lobbying prowess of Fannie Mae. The techniques they employed to emasculate their regulator are laid out in detail. This alone makes the book worth reading. One of my themes, that mortgage securitization emerged from rent-seeking, gets a lot of support, the way I read it.
I also think that the Suits vs. Geeks divide gets a lot of play. The geeks invent risk management tools, such as Value at Risk, and often the suits misunderstand and misuse them. On p. 57:
many Wall Street CEO's would view their daily VaR number as an expression of their firm's worst-case scenario. But it was nothing of the sort....The fact that VaR told you how much your firm might lose 95 percent of the time didn't say a thing about what might happen the other 5 percent of the time.
Devils points out that there were many indications of fragility in the financial system that were ignored. For example, on the rating agencies, p. 114
The rating agencies had missed the near default of New York City, the bankruptcy of Orange County, and the Asian and Russian meltdowns. They failed to catch Penn Central in the 1970's and Long-Term Capital Management in the 1990's. They often downgraded companies just days before bankruptcy--too late to help investors...To critics like Partnoy, the fact that ratings carried the force of law explained a troubling paradox: even as proof piled up that the agencies made mistake after mistake, their power continued to grow.
So far, I have not told you anything that came from the book that I did not already know. But in fact I gained a fresh perspective, particularly from two of the executives profiled in Devils: Angelo Mozilo, the CEO of Countrywide Funding; and Roland Arnall, the owner of a number of large subprime mortgage lenders. Each of these men were capable of saying--and apparently believing--that they were the good guys, making safe, customer-friendly loans, in a world of greedy cutthroats. About Arnall, p. 133:
Headquarters, in fact, acted as if the company were a paragon of subprime virtue, rather than a place that oozed with sleaze and fraud. In July 2000, for instance, Ameriquest publicly committed to a set of best practices...The following year...Ameriquest had been invited to testify because many in Washington considered Ameriquest a model subprime lender...
The company was constantly being hit with accusations, investigations, and lawsuits charging fraud and deceptive practices...At best, Arnall seemed to practice a kind of willful ignorance.
We also see "consumer groups" being bought off by people like Arnall. Indeed, I always thought of ACORN as shakedown artists--they would accuse a big company of racist or anti-consumer lending practices, and the main result of any settlement would be that the company would make a big grant to ACORN, as opposed to really changing its practices.
On Countrywide, p. 219:
By 2006, there was a distinct Dr. Jekyll and Mr. Hyde-like quality to Angelo Mozilo. The good Angelo had been warning for a surprisingly long time that his industry was heading into dangerous territory. "I'm deeply concerned about credit quality in the overall industry," he said in the spring of 2005...
The bad Angelo insisted that none of this would be a problem for Countrywide...Countrywide's "proprietary technology" would help it "avoid any foreclosure," Mozilo told investors
Mozilo and Arnall come across to me as pathological. They can articulate the difference between right and wrong, but they have an ability to convince themselves that they are doing right even as they are doing the opposite of what they claim. How do such people rise to the top of large companies? Once again, I say that if somebody made me the chief risk regulator and said that I had to prevent financial crises, I would change the gender of the financial CEOs. I don't think female executive minds are as capable of the same pathology.
Devils does a particularly good job explaining what happened with AIG. Here is a tidbit that was new to me (p. 203):
FP did not completely turn off the spigot at the end of 2005, even though that is what the company later told the world...under the terms of the swap contracts it wrote, CDO managers had the right to switch collateral to help maintain the yield--without having to inform AIG. As borrowers prepaid mortgages, for instance, the CDO managers would replace those earlier mortgages with mortgages that had been written in 2006 and 2007.
In various places in the book, you can see data that explain why "loan modifications" and "keeping borrowers in their homes" will not work. A large percentage of home buyers did not make even one payment and never occupied their homes. On p. 255:
Rosner's eureka moment [this is Josh Rosner, about whom Russ Roberts and I both blogged recently] comes when he sees data showing that about 35 percent of the mortgages used to purchase homes in 2004 and 2005 are not for primary residences, but for second homes and investment properties.
I should note the a Fed study found only 15 percent of mortgages for non-owner-occupied in that period. However, it would not surprise me if Rosner found a more accurate source--there is a lot of fraud in which a borrower says that the loan is for a primary residence when it's not.
Overall, I find the book to be filled with examples of what I call expert failure. Experts in the private sector as well as the public sector greatly over-estimated their knowledge and abilities. It is hardly comforting when in their epilogue, the authors of Devils write on p. 363:
These new regulations also will only be as good as the regulators themselves.