Traditional conservatives (as opposed to Trump Republicans) tend to support the Austrian school of budgeting, largely based on the economic teachings of F.A. Hayek. He often pointed to the problem of economists and other scientists in predicting the future, particularly with quantitative models that presume human behavior. He called it “an ambition to imitate science in its methods rather than in its spirits.”
I remembered his quote when looking at the epidemiological forecasts of Covid-19. It was only a few months ago that Dr. Ezekiel Emanuel predicted 100,000 victims would be hospitalized in a mere four weeks. Because I have great respect for Dr.Emanuel, I wondered how he could be so wrong. Maybe, forecasts have the power of changing behaviors? Maybe, people became more careful?
Thinking about economists in general, I’ve have seen many wildly-wrong forecasts. My computer friends are always eager to remind me that “garbage in, garbage out” and I understand their point. For investment strategists, a major input is the measure of risk called “standard deviation”, which is really “garbage in”. It is probably even less reliable to next week’s weather forecast. For months now, I have been wandering through an educational desert looking for a repeatable, reliable measure of risk. Maybe, an investment forecast must also assume that human behavior is static and predictable. If not, how to model a moving target? How do you quantify the change in human behavior caused by a forecast?
It is convenient to hide in the wisdom of Casey Stengel, who famously said “never make predictions, especially about the future.” In economics, we know that supply will rise to meet demand. As long as there is a demand for forecasts, people will supply them. My long-dead but very-sweet grandmother used to say “you can’t do nothing about nothing anyway.” So, why bother with forecasts at all?
