More Thoughts on the Household Survey from the Jobs Report

As we now know, the drop in the U.S. unemployment rate to 7.8% was driven by a big jump up the estimated number of employed people – 847,000 according to the Bureau of Labor Statistics. It might be worthwhile to look a couple of the factors that influence that number. It’s worth noting too that these factors might not be mutually exclusive.

1 — A big jump up in the number of people taking part-time work for economic reasons – this was up by over 500,000.

2 — An increase of over 300,000 in the seasonally adjusted employment rate for 20- to 24-year-olds (covered very thoughtfully by Catherine Rampell over at the New York Times blog Economix). Typically, from August to September the number of employed 20- to 24-year-olds drops as college kids give up their summer jobs to back to school. In fact, employment for 20- to 24-year-olds has increased only two other times since the employment series started in 1948. In addition, from July to August, in the non-seasonally adjusted data, the estimated number of employed 20- to 24-year-olds dropped by 530,000.

We would expect some give back in employment for either the young or the part-time for economic reasons in coming months just because of mean reversion.

Using our employment model: if we hold the labor force participation rate at 63.6%, we’d have to see employment grow by at least 100,000 next month to keep the unemployment rate at or below 7.8%. If employment dropped by 500k, that’d the unemployment rate at 8.19%; if the employment dropped by 300k, then the unemployment rate would wind up at 8.06%.