In his consideration of what sent the financial world spinning this month, one of my favorite newspaper columnists, David Brooks, blames the human factor, specifically, unrealistic attitudes about risk among decision makers in the financial world. He finds a tendency among these movers and shakers to get high on a little success and push their luck further with each success. This, he says, is a form of social contagion that has swept rational risk assessment away. I’m with him up to this point–even though he ignores the obvious force of deregulation in the current crisis.
But then he finds that “computer risk models” have played a role in the financial turbulence by supporting unreal attitudes toward risk in the finance sector. This is where I part company with him because I’m pretty sure he’s referring to risk analysis techniques like Monte Carlo simulation and neural network forecasting. While these are powerful techniques, any user of Monte Carlo software or a neural net program knows that these tools depend on objective premises and sound data for their effectiveness. If you are emotionally disposed toward understating an accounting construct like value-at-risk, it’s only too easy to come up with understated values.
But maybe Brooks really does understand the value of these tools and that they have a positive role to play, because he calls for a global leadership that will decide issues such as how much leverage should be allowed and whether other countries should be able to “park” huge reserves in the U.S. Obviously these decisions will have to be based on something more than emotion, and I can only assume that he assumes they will informed by statistical analysis models created by cool heads with good data.