Summary
Models only cover certain variables and the crisis involved much more than mortgage pricing and a few macroeconomic issues. Risked based pricing as used in the mortgage industry had an underlying flaw that caused the models to fail. Greed played the other part.
Analysis
The models perdicting mortgage investments based on Risked Based pricing were flawed in that only half the equation was used. Yes the cost for a $100K mortgage to the borrower with bad credit was priced with a higher interest rate. That does not address the risk, only the yield. Some of the risk was planned for in the reserves, an increase from a 1% loss rate to a 3% loss rate. What was not addressed was the asset and its risk.
If borrower A has good credit then he deserves market rate. If borrower B has bad credit you do raise the interest but you either do not lend him the same amount, in other words you increase the LTV to say 75% in order to protect the asset, you do not give him a no money down loan. If you do want to give him a no money down loan, then you reduce the exposure to say $75K even though his data supports a $100k loan.
That was the first failure. The other failures in the models is the ability to plan for macroeconomic trends including such issues as a negative savings rate, an over building of product, etc. It amazes me that the models could not get right what several of us in the industry knew and warned against. Why, because common sense does not play a role in models.
Greed was the final factor, the product offerings got ridiculous because the industry was feeding the beast (Securities) with the need to grow volume for each quarters new securities. That combined with ultimate failure of the models to understand the "not on my balalnce sheet mentality" of the originators caused the toxic pool of loans because they became the investor's problems.
To borrow a line from the movie "National Treasure" someone needs to go to jail.
