The key point in this jumps out of the figure that Paul [Krugman] posted.
How do we know that unemployment would peak at 9 percent in the absence of any action? We don't. It's a guess. Essentially, it's a guess made by a technician with a computer.
True, the figure is well within the range of modern historical experience: The peak in October 1982 was near eleven percent (though with a more honest count of the unemployed).
But if you believe that we are in an historic crisis, then the conventional methods of forecasting--playing the averages from the postwar statistical record--are inappropriate. Honest, but inappropriate.
...
For my part, when I compare this situation to 1929, I'm not making a rhetorical point, or exaggerating for flair. It's what I think. It's what the specialists I talk to think. And have thought, since at least early last summer, long before the crisis took over the campaign.
What is happening here, is the application of routine post-war econometric modeling and budget procedures to the present situation.
If we are in a crisis, and out-of-the-postwar range, then the routine methods are going to be radically wrong. Not just off by 0.3 percent. Off by a great deal, and for much longer.