Read other letters about this article
That's the difference between 6,750,000 and 6,735,000. Seems not too far removed from statistical noise. I mean, if you were standing in the middle of a sea of six and three quarter million people and fifteen thousand of them vanished, unless one of the vanished was standing next to you, would you even notice?
I'll raise the same question that others have raised: How much of the decrease in continuing claims for benefits (which is where this "good" number comes from) are due to people who have exhausted their benefits? Jobs are still being lost at a pretty astounding clip and unemployment is still rising.
Finally, why would the layoff of more than 620,000 be reason to think that the worst is over? That number may be down from the truly horrendous peak, but it's still way, way high. In fact, it wasn't until the end of January of this year that initial claims for UI exceeded 600k in this downturn, and whatever "good" news we are supposed to get from 620k sort of gets blunted by the fact that initial claims were 605k in back on May 2. So we're higher than that now.
The bottom line here is that you can't micro-parse the data. Week to week results don't mean much. It's the bigger picture you need to look at, and in that regard, the moving average actually went up by 4k. Go figure.
(You can run the numbers on claims here: http://www.ows.doleta.gov/unemploy/claims.asp , or click my sig.)