A Consultant's View

Prairie Trail Software, Inc. ............................................................. January 2009

What went wrong on Wall Street?

The "risk managers" at the financial firms were supposed to keep us safe on Wall Street. These are very smart, technically astute people who had fancy computer models that were supposed to keep us from taking too much risk. What went wrong?

Some people say that the top managers at those Wall Street firms leaned on the risk managers to change the models a little. A better case can be made that the whole idea was faulty. It was faulty both in the use of risk models, and in the way that the company was structured.

Risk management is based on financial models. These models, no matter how good they are, all fail because of three reasons. First, they are based on modeling to a specific "assurance level". With any model, there is always a 1%-5% part of reality that does not fit the model. When we hit that 1%-5% of reality, the model doesn't work or doesn't predict what will happen. Secondly, disruptive events in the market are always quick events. For example, most of the damage done to a stock happens on one day out of perhaps years of data. Most of the gains that occur happen on just a few days of the year. Models are based on more steady state situations, not disruptive events. Thirdly, the models can not model systemic fraud. Human beings are very creative and that includes those who do fraud. Systemic fraud has always, and will continue to occur outside of everyone's models.

When we blindly use the models, we assume that human beings will follow mathematical rules.

The structure of the companies was at fault also. Simply put, the top managers were not being rewarded for keeping their companies safe. The pay scales, the bonuses, the keeping of the job were all based on how much return they made for their shareholders. There was no consideration of whether or not they were taking excess risk. In such an environment, they delegated the "risk management" to lowly technicians who could not stop the company from any trade. That meant that only returns were considered and not the risk of those investments.

When Toyota decided to make quality their number one priority, they gave the power to stop the production line to the lowest person on that line. That person could stop the line based on the quality level. With such power, quality skyrocketed. On Wall Street, the regulations stated that risk was supposed to be considered, but the power was not handed to those who knew the risk. That means that risk didn't count.

The lesson from this for our own operations is that when there is a significant risk, the people who know how to manage that risk have to have control over it. When something is too important, it should not be left to those who have no power.

In short, financial models are good, but they are no substitute for clear thinking. In our operations, we need to balance how things "should be" with reality and provide the proper balance between responsibility and power.