Is this the real deal? Yesterday’s update on new jobless claims, which fell to a 3-1/2 year low last week, suggests that the labor market will continue growing, perhaps at slightly faster rates than we’ve seen in 2011. “This is unexpectedly great news,” says Ian Shepherdson at High Frequency Economics. But skepticism abounds, and rightly so. There have been several false starts to the recovery since the Great Recession was officially declared at an end in June 2009. There’s no way to know for sure if the latest drop in new filings for unemployment benefits signals a true turning point for the better this time, but the possibility can’t be dismissed either, at least not yet.

Still, economist Brad DeLong has his doubts about the latest fall in claims. “The seasonal adjustment factor makes it shaky news,” he writes. Perhaps, but if we strip out the seasonal adjustment and look at the raw claims data on a year-over-year basis the latest numbers are in line with the trend in recent months of a roughly 10% annual decline. In fact, the latest report shows that unadjusted new claims were lower by nearly 12% vs. the year-earlier figure—the biggest drop in two months.

Some analysts worry that future revisions to the jobless claims data will wipe away some or all of the recent good news. One can never discount that possibility, but here too the trend appears friendly when we look at recent history. The next chart tracks changes in revisions for the weekly seasonally adjusted claims reports. Revisions are defined as the latest reported number for a given week less the initially reported number. As the chart shows, the revision trend has been only moderately upward yet stable for much of this year. That’s one more reason for thinking optimistically about the labor market. Moderately higher revisions for claims aren’t helpful, of course, but there’s nothing in the data that suggests that this drift is about to change. In short, relatively minimal upward revisions are no threat to a trend that’s generally moving lower.

The acid test is still ahead of us, however. The drop in claims has to prove itself by holding its ground if not moving lower. That’s a key test because this is a volatile data series on a weekly basis and so any single report has to be interpreted cautiously. More importantly, we need evidence that the drop in claims is translating into job growth. History suggests that new claims are a strong leading indicator of future economic activity, including payroll changes. Still, everything is murky in real time and there’s only so much confidence one can muster from looking at history.
Meanwhile, the uncertainty about what happens next with the euro crisis continues to hang over the global economy. The dysfunctional political bickering in Washington over the budget negotiations is another potential trouble spot. The trend is still precarious, but perhaps less precarious than the pessimists recognize. Yes, it’s still one day (and data point) at a time.