Evidence-Based Economics
Talking to David and Christie Romer about whether we are or about to be in a recession. I think that the maximum-likelihood estimates of the probability for all plausible models in which "recession" is a useful and meaningful term are all equal to 1.0000. But:
- We are not sure that "recession" is a useful and meaningful term.
- We are not the kinds of people who believe in ML estimation. Instead, we are Baywsians who are ignorant of our own priors...










"Bayesians ignorant of our own priors."
Excellent line.
Posted by: Barkley Rosser | August 27, 2008 at 02:58 PM
What Barkley Rosser said.
I'd say the graph of unemployment vs an ar(2) with a fat tailed disturbance term fit to unemployment said that "recession" is a meaningful term.
Posted by: Robert Waldmann | August 27, 2008 at 03:31 PM
Somebody needs to carefully inspect the GDP and employment components and compare this slowdown to past history, including the mid-1990s hiccup. A summer 2009 exercise?
Posted by: Measure for Measure | August 27, 2008 at 05:00 PM
The concept of a national recession may be losing its usefulness.
This "recession" is by region and by business line, even if the numbers suggest there is no national recession.
What ever we want to call it, it is here.
Posted by: save_the_rustbelt | August 27, 2008 at 05:40 PM
Seems rather arbitrary. There'd be no question we were in a recession if we had a birth rate comparable to the rest of the developed world.
Faster population growth may give us statistically fewer recessions, but it doesn't make any of us any richer.
Posted by: Nietz | August 27, 2008 at 05:52 PM
Baywsians! Aewsomw!
Posted by: Noah | August 27, 2008 at 06:50 PM
Jimmy Savage died for your sins.
Posted by: Cosma | August 27, 2008 at 08:14 PM
"I think that the maximum-likelihood estimates of the probability for all plausible models in which "recession" is a useful and meaningful term are all equal to 1.0000."
Actually, I think Jim Hamilton estimates it to be 69% as of the past week.
see http://www.econbrowser.com/archives/2008/08/recession_indic_2.html
As he summarizes
" I do think there's a pretty strong case, based on the employment and unemployment numbers, that we are currently in a recession. Nevertheless, we will be making the Official Econbrowser Declaration, for what it's worth, using the GDP-based recession indicator index, ...
And with 1.9% second-quarter growth as currently reported, GDP growth doesn't meet the criteria for a recession so far. "
Posted by: SvN | August 28, 2008 at 06:50 AM