Friday, September 19, 2008

A Mystery to Me

Via CIP, this link: How Wall Street Lied to Its Computers, and from there this excerpt:
“New products, by definition, carry more risk,” she said. The models should penalize investments that are complex, hard to understand and infrequently traded, she said. They didn’t.

“One of the things that has caused great pain is complex products,” Ms. Rahl said.

That made me think back to some of the great trading debacles of the last century, such as the collapse of Askin Capital Management, a hedge fund that fell apart because of complex mortgage security investments gone bad. Wasn’t the moral of those stories that you shouldn’t put your money (or your client’s money) in something you didn’t understand? Furthermore, even if you are convinced you do understand it, you’re not going to be able to sell it when you need the money if no one else does.

This is an unheeded moral from last century's failures.

But just how does the "free market" come up with untradeable products?


CapitalistImperialistPig said...

Let me try to give a better answer than I did on my blog. The products that it comes up with, and sells, are tradeable. They are tradeable because they bear high interest yields, seem to be performing, and are highly rated by credit agencies. It's later, when they aren't performing, and the credit ratings start looking misguided, that they become untradeable. That can happen in any product - the horse you bought last month might be a lot less valuable if he's come up lame in the meantime. With financial products, its clear that there value was built on unrealistic expectations, and those unrealistic expectations were built on greed, folly, fraud, and most of all on the agent - principal problem - the disconnect between the interest of the money manager and the interest of those whose money he or she is managing.

Arun said...

You can test-drive your horse, yes, it could turn lame later. In the case of these securities, you don't have anything but somebody's say-so.

Arun said...

CIP, this seems to capture what was a mystery to me:

Thin Markets, Asymmetric Information, and Mortgage-Backed Securities*1

Edward L. Glaeserb, c, a and Hédi D. Kallale, d
a Harvard University b Hoover Institution c NBER d New York University e Salomon Brothers

Received 21 July 1993.
Available online 7 May 2002.


This paper tries to explain why the issuers of an asset would restrict what information is available about their asset. In a world where knowledge is valued, market forces should induce disclosure, but we often see markets (such as the market for mortgage-backed securities) where assets' issuers refuse to release valuable information. We present a model of market liquidity and find that market liquidity can both rise and fall with the quantity of released information. More information may increase asymmetries of information and “lemons” style breakdowns. We find that asset bundling is more advantageous when private information is more accurate, which may be the case in the mortgage-backed securities market.Journal of Economic LiteratureClassification Numbers: G14, G32.