Thursday 10 December 2009

The folly of economic forecasts

In this podcast at Plant Money Russ Robert, George Mason University economist and host of EconTalk, says he has come to believe it's impossible to predict future economic conditions because good data is so hard to come by and even harder to compare. So what about all his fellow economists who seem to have an opinion on almost any topic? Roberts says they should come out and tell the truth -- that their policy recommendations are based on philosophy and ideology, not on empirical data.

3 comments:

Matt Nolan said...

"Roberts says they should come out and tell the truth -- that their policy recommendations are based on philosophy and ideology, not on empirical data"

I completely agree that even with lots of data policy recommendations are normative. And I agree that people "hide" behind useless data to avoid trying to have a decision on the issues logically (which is what using theory is for).

But, a lot of economic data is imperfect but is positively correlated to the real underlying variables - so it can at least help to inform the expected direction of things.

Paul Walker said...

Matt. I think my response to Roberts would be that the market tells me that forecasts are of, at least, some use. Otherwise why do people pay for them?

Matt Nolan said...

"Matt. I think my response to Roberts would be that the market tells me that forecasts are of, at least, some use. Otherwise why do people pay for them?"

Agreed, in so far as there is a market paying for them.

Public sector economists and economists that are payed by the public sector can't really use that line though I think ;)