US jobless claims continue to cast a positive glow on the outlook for the labor market. Today's weekly update shows that new filings for unemployment benefits fell a solid 13,000 to a seasonally adjusted 263,000 for the week through Oct. 3. That leaves claims only modestly higher than the four-decade low of 255,000 that was touched back in July. In short, this leading indicator is effectively telling us that the recent worries about the macro trend are excessive.
One caveat to consider: the latest estimate of job cuts by the workplace consulting firm Challenger, Gray & Christmas (CGC) suggests that layoffs are set to rise. CGC's September report notes a “surge in job cuts” last month. The firm's estimate of planned layoffs for September represents a sharp 93% rise vs. the year-earlier level.
How to account for the divergent messages in the downward trend in the government vs. the bearish sign in CGC's data? Perhaps it's an issue of announced cuts for the future via CGC vs. the Labor Department's confirmed filings for new unemployment benefits in the recent past. In theory, the two will line up… eventually. If CGC's numbers are accurate, jobless claims may be destined to rise, and perhaps sharply in the weeks and months ahead. In that case, worries that the US economy is vulnerable in a world of slowing growth will be vindicated.
But leaving aside what may or may not happen in the future, today's official claims data is effectively predicting that the labor market will continue to grow. Taking the numbers at face value implies that last month's disappointing employment report isn't a sign of rising risk for the US business cycle. Exhibit A in today's release is the continued year-over-year decline in new claims, which fell 9% last week vs. the year-earlier level. If and when claims begin to forecast a darker outlook for the labor market, we'll see the bearish signal via claims that are rising on a consistent basis in the annual comparison. For the moment, however, that degree of gloom is nowhere on the horizon, at least in the government's data.
