Support Marketplace

Don't bet on economic forecasting

What's Next on $100 bills

TEXT OF STORY

Kai Ryssdal: If you really think about it, what the Fed's trying to do when it meets to talk about interest rates is an amazingly complicated thing: Tease out what a $14 trillion economy's going to be like six months or a year from now, and then apply the right policy.

Commentator and economist Justin Wolfers says forecasting the economic future truly is a dismal science.


Justin Wolfers: Economists and meteorologists know that forecasting is hard. But at least when the weather man is wrong, we just get wet. When economists get it wrong, we get the Great Recession. And so we economists spent much of the past decade developing a new generation of models.

But two young researchers have raised big questions for this research program. They compared the track record of various state-of-the-art models and elite forecasters. What did they find? All the leading approaches to forecasting have been terrible.

Translation: Our forecasts of GDP, inflation and interest rates have been way off.

Here's a neat way of thinking about how badly these models fail. If in 1997, you'd trained a parrot to say "GDP growth and inflation will be 3.5 percent, and interest rates won't change," then your parrot would have done as well as the leading macroeconomists, who fine-tune their numbers daily. Or if your parrot just says "Whatever happened last quarter!" whenever he's asked for a forecast, he'll still do as well as those Ph.D-toting economists hired by the leading banks. But he doesn't need a salary.

Academic economists thought we were smarter than your parrot. We've trumpeted progress in developing what we call -- wait for it -- "Dynamic Stochastic General Equilibrium models." They're not just a mouthful; they're also the result of a decade of intense research. But if you ask these models to predict the future, they're no better than your parrot.

Now, statisticians don't pretend to understand the economy, they just crunch the numbers. But their models performed even worse.

Perhaps you trust people instead of computers. But the Fed's economists were no better. And if you have greater faith in the private sector, you really shouldn't. Their prognostications were also dismal.

Now, you probably knew that most economists failed to predict the 2007 recession. What's news is that even after the Great Recession arrived, our complicated models failed to recognize the new reality.

Bob Dylan sung it best: "You don't need a weatherman/To know which way the wind blows."

And according to this new research, you don't need an economist to know which way the economy goes.

Kai Ryssdal: Justin Wolfers is a professor of business and public policy at the Wharton School of Business at the University of Pennsylvania.

Comments

I agree to American Public Media's Terms and Conditions.
With Generous Support From...