Posted by: Dirk | April 17, 2009

Dodgy forecasting

Some months ago, it became apparent that US monetary policy would not be available as a policy instrument in this recession. The interest rate hit the zero bound without any significant increase in investment. That is no surprise, since a lower interest rate works mainly through an increase in housing starts. The whole story reveals that Black Swans do exist: extrapolating history is not a good idea to forecast the future. But this is what most forecasters do these days. The models of the DIW don’t work since they are based on past observations. Also, predictions of the oil price have never predicted any swings. When the oil price was rising above $100, most economists predicted that it would reach $200. It turned out that the oil price is partly demand driven. If you cannot predict global demand, you cannot predict the oil price. So everybody predicting the oil price must have a (working) model of the global economy. Nobody has that.

I admit that it is very difficult for social scientists to predict anything, since there is non-linear behavior in many aspects. That is why most macroeconomists predict only tendencies of exchange rates, given some imbalance or other. But history is no guide to exchange rate movements. The US dollar has always lost against European currencies since WWII, but that does not mean that it will continue to do so. If the Fed would decide to stop printing that much money (although I cannot see that coming), the decline would certainly stop. A declining dollar is no natural law but the outcome of decisions made by human beings. And sometimes we change our opininion about policy.

Therefore, so-called experts of forecasting must know something more about their field than just the past movements of the variable(s) they are tracking. They must have a model of how the variable “works”, how it relates to the rest of the universe. In other words: economists should rely on econometrics, but they should have an idea about an underlying model also. Just extrapolating past data leads to wrong predictions every time the economy turns recessionary, and then during the crisis nothing can be said. But this is exactly the time when policy-makers need input from economists.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: