Recession numbers

James D. Hamilton is one of my favourite economists. He has written a great book on time series, and while in California I was lucky to sit in on his econometrics class; he is a fabulous lecturer. On his blog Econbrowser he offers analyses and comments to current economic conditions and policy. In a recent post he discusses numbers that indicate that the US economy is in a recession (industrial production fell by 33.6% (annual rate) in September). More interesting, however, are the comments to the post where a reader asks whether academic economists did (or didn’t do) their job to warn about the possibility of the severe economic breakdown that seemingly is happening in the US. Underlying this question is the much bigger question of whether (macro) economists know what they’re doing.

I find the critique against academic economists misplaced. Many of the best economists work in the financial sector and in related governmental agencies, and neither these nor the academic economists were able to forecast the current crisis. (Well, as Jim points out, several economists did warn of potential danger. The sheer size of it, however, was seemingly not predicted with enough confidence.) There is no reason to believe that academic researchers are better at forecasting than those employed in the financial sector. (Those in the finance sector would not necessarily publish their forecast, but if any of them had any reliable forecasts of the crisis we’re in, their companies would have positioned themselves relative to that, and it seems that none of the Wall Street companies had done so.) This also tells of the extreme difficult of forecasting in economics.


Tags: , , ,

One Response to “Recession numbers”

  1. 1 000 000 000 000 « Kvams Says:

    […] is a trillion? Jim asks in relation to the proposed U.S. deficits in the next few years. (Jim is one of my favourite economists, or should I say professor.) The short (and pedantic) answer would be the title of this post. […]

Comments are closed.

%d bloggers like this: