Tuesday, August 4, 2009

Economists Do Not Have Credibility

Economists' predictions do not have credibility. If they did, people would trust the forecasts and heed the warnings. Incentives to economists will not improve the accuracy of their forecasts.

Economists understand the relationship among the components of the economy so that, after the fact, economists are good at explaining related events or the secondary effects. For instance, economists understand that a decline in GDP will cause an increase in unemployment. Economists are just not good at predicting the original decline in GDP, its duration or severity.

When secondary effects start to appear in the economy, such as an increase in unemployment, economists know to look for an earlier causal event, such as a decline in GDP, that they missed seeing earlier.

To predict a future recession does not require any special training or expertise. Recessions are a fact of our economy and they will occur. The forecasting difficulty is to say when it will begin, its duration and its severity. If a recession will begin 6 months later than the prediction, that is 6 months more profits for some business that has the product or service to sell. A business that decreases its availability of a product or service too soon will see those lost profits go to a competitor. A competitor could gain an insurmountable edge.

Many components of GDP, and their interrelationship, move randomly around a trend line. A random movement by its very definition is short term unpredictable. Many times, random movements are offsetting, but other times they move in unison. When economic components move downward together, there is a decline in overall GDP and an economic slowdown.

The randomness of the movement of many components of the economy is what makes economic forecasting inaccurate. Inaccuracy leads to a loss of credibility as a forecaster. Without credibility, businesses and policy makers will continue to ignore the predictions of economists.

Also posted on The Economist Free Exchange Blog.

No comments:

Post a Comment