The Risk Disclosure Problem

Taleb and Greenspan have gotten all the press, but my nomination for book of the year in the finance/economics space (with the important proviso that I haven't got very far into it yet) is "Plight of the Fortune Tellers" by Riccardo Rebonato. Like Paul Kedrosky, I'll come back to this later, once I've finished it. But it seems to me that Rebonato's book is a constructive and very important way of changing the way that we look at risk in the financial world.

While Paul and I are still working our way through this book, Arnold Kling has already finished it, and Rebonato seems to have sparked some extremely interesting ideas about the relationship between liquidity and transparency. I'm not at all sure that these are Rebonato's ideas, but they're fascinating all the same: Kling's big thesis is basically that the two generally work against each other.

Kling explains:

Financial intermediation is about enticing investors to buy securities backed by investments whose risks the investors cannot fully evaluate. The intermediary, such as a bank, hedge fund, or ordinary corporation, specializes in evaluating risk. The investor who buys securities from the intermediary looks to the past performance of the intermediary as well as to concise summaries of the risk of those securities. The ratings of "AA" or "A+" by bond rating agencies are just one example of these concise risk summaries.
Modern financial intermediation is multi-layered. The mortgage broker knows the specific characteristics of the house being purchased, as well as the borrower's financial data and credit history. Mortgage funders funnel funds through brokers, using only summary statistics such as the borrower's credit score, the ratio of the loan amount to the appraised value (LTV), and the broker's historical performance with the funding agency. Funders then pool loans together. Firms that buy the pools know only the general characteristics of the pool -- the rangeof credit scores, the range of LTV's, and so on. These pools may befurther carved up into "tranches," so that if loans start to default, some investors will take an immediate loss while others continue to receive full principal and interest.
At each step in the layering process, some of the detailed information about the underlying risk is ignored. Instead, investors rely on summary information. It is this use of summary information that makes these investments liquid -- that is, it enables them to be bought and sold by many investors. As an intermediary layer is added, while the amount of detailed risk information is going down, liquidity is going up.

I think this is true, and something that Wall Street is loathe to admit. Much of finance is predicated on the idea that investors, in aggregate, are in possession of all the relevant information they need to make investment decisions – and that therefore since that information is alread "priced in", there's often little point in reinventing the wheel and going out and getting all that information ourselves.

But by the time that structured products go through two or three iterations of tranching and reconfiguring, there's really no one who has a clue what the underlying risk is or how to measure it. Investors in such products are buying the reputation and history of the entity which structured the product as much as they're buying any particular risk. They also tend to place a lot of faith in statistical models which might not bear as much relation to reality as they think.

When I raised questions about the use of Value-at-Risk as a risk management tool, a commenter replied with the utmost faith in Wall Street:

I presume that [Morgan Stanley], like other investment banks, has other, more sophisticated, risk controls, but are reluctant to release detail to the public, because it is part of their IP.

Me, I presume no such thing. Morgan Stanley may or may not have other risk controls; those controls may or may not be more "sophisticated" than VaR; and that sophistication may or may not make those controls more useful as a risk management tool. But my gut feeling is that you don't have sophisticated risk managers running around Morgan Stanley telling people to take less risk: rather you have John Mack, at the top of Morgan Stanley, giving strict instructions to his bankers to take on more risk, and firing them if they don't.

There is no shortage of mathematical prowess at all levels of Morgan Stanley, I'm sure. But a lot of that is devoted to looking at securities pricing, as opposed to real-world sources of potential systemic disruption like the chance that house prices will fall by 10%. And so securities get built on top of other securities, which are built on models rather than reality. And liquidity goes up, and everybody makes lots of money, and no one, really, has the slightest clue what they're buying.