The data on employment and unemployment come from two different surveys, as we've often said here, and every survey is revised as missing data at the time of the release -- employment data is always released the first Friday of a month. The revisions to payroll employment this month, though, are staggering. Here's a comparison of this month and the previous month's estimates of the change in employment, in thousands (the October 6 figures as reported then are in parentheses):

Aug 06 230 (123)
Sep 06 148 (51)
Oct 06 92 (---)

Contemplate that a second: We added 107,000 jobs to the August estimate and 97,000 to the September estimate, and created 92,000 more in October. Will anyone credit the economy (or, more to the point, the Bush Administration) with 296,000 new jobs? Not bloody likely.

If you don't think this is actually good news consider this:

Average hourly earnings of production or nonsupervisory workers on private nonfarm payrolls rose by 6 cents, or 0.4 percent, in October to $16.91, seasonally adjusted. Average weekly earnings rose by 0.7 percent in October to $573.25. Over the year, average hourly earnings increased by 3.9 percent and average weekly earnings increased by 4.2 percent.

For the 12 months to September, real wages are up 2.4%; this increase will likely push that number higher. Will this get covered? Again, quite unlikely. But I think any chance of a decrease in the Fed Funds rate any time soon has diminished quite a bit with this piece of news.