In experimental physics the way I did it — I was a theorist — there were certain data to which my first response was to smack the equipment and try again. Initial claims of unemployment at 348,000 is a smack-the-equipment kind of number; it's good enough that I would actually be in a better mood with something a bit higher.

A friend who worked at the Federal Reserve told me once there was an effort to determine whether the weather had a short term impact on these numbers, perhaps pushing some claims from one week to the next. If next week's number comes in below 390,000, I promise to believe this week's number.

Anyway, productivity for Q3 preliminarily at 8.1%. If you hear people asking, "how can the economy be doing so well if it's not producing any jobs?", well, productivity tends to be the answer. Productivity above 3% is not realistically sustainable, and we won't be sustaining 7% growth without a pick-up in jobs; in fact, part of this may be due to the way numbers are calculated and the fact that inventories dropped substantially in the third quarter. That will tend to boost productivity numbers, insofar as I know what I'm talking about; as inventory numbers are bounded below, that's the kind of temporary effect that can't be sustained. It's also a hallmark of incipient economic recovery.