For the second straight week, the number of new jobless benefit claims fell sharply, reported the Department of Labor.
In the week ending July 11, the advance figure for seasonally adjusted initial claims was 522,000, a decrease of 47,000 from the previous week's revised figure of 569,000. The 4-week moving average was 584,500, a decrease of 22,500 from the previous week's revised average of 607,000.
The number of continuing claims -- the total number of people drawing unemployment benefits -- fell by a record 642,000, to 6,273,000.
Echoing last week's analysis, labor market watchers warned that the numbers might be skewed because the Department of Labor normally expects auto manufacturers to idle plants and lay off workers in the summer, but that isn't happening this year, because the workers had already been laid off. But as I noted last week, the same analysis partially explains why initial jobless claims numbers did not fall as quickly earlier this year, after reaching their peak, as they normally do. So while we can't expect such sharp drops to continue indefinitely, there seems little doubt at this point that we are solidly past the peak, which implies, possibly, that unemployment rates will not rise as quickly in the second half of the year as they did in the first.
The unemployment rate won't fall, of course, until the U.S. economy is actually creating more new jobs than it is shedding.
UPDATE: The New York Times' David Leonhardt, while terming the new data "relatively good news," also includes some sobering comments from economist Joshua Shapiro:
As we have highlighted, this week's report (and last week's) were hugely distorted to the downside by the shift in the timing of automotive plant closings and a consequent inability of the seasonal adjustment process to correctly compensate for this factor. This is best illustrated by the fact that non-seasonally adjusted claims increased by a cumulative 107K in the latest two weeks, but the seasonal adjustment process turned this into a cumulative 95K decline as the unadjusted rise was less than "expected" owing to the automotive factor.
Seasonally adjusted initial claims should begin to normalize in next week's report and substantially finish that move in the following week (data for the week ended July 25). Until we see how claims have rebounded in two weeks time, it would be a big mistake to conclude anything from these numbers.