BLS Benchmark Revision

Last Friday the BLS released their benchmark revision with this month's nonfarm payroll number. This number is an under appreciated statistic, as the information it provides about the jobs picture is much larger than the monthly change*. The benchmark revision was -366,000, while the monthly job change in September was -95,000. This means that their 2010 estimates of the amount of people working need to be adjusted down by 0.3%.

Industry: Benchmark revision (Percent revision)

Total nonfarm: -366,000 (-0.3)

Total private: -371,000 (-.4)

Mining and logging: -20,000 (-3.0)

Construction: -62,000 (-1.2)

Manufacturing: -114,000 (-1.0)

Trade, transportation, and utilities: -144,000 (-.6)

Information: -11,000 (-.4)

Financial activities: 42,000 (.6)

Professional and business services: 14,000 (.1)

Education and health services: 6,000 (*)

Leisure and hospitality: -91,000 (-.7)

Other services: 9,000 (.2)

Government: 5,000 (*)

(*) Less than 0.05 percent.

The industry breakdown shows that there has been a lot more adjustment in the manufacturing and trade, transportation and utilities, and the leisure and hospitality industries than have previously been accounted for. The carnage in the finance industry was overestimated, and there were more jobs there than previously thought.

The BLS suggests that there revision in the benchmark survey comes primarily from nonsampling error.
Sources of nonsampling error include coverage, response, and processing errors in both data series. Additionally, the survey is potentially subject to sample design and estimator biases.
Calculated Risk collected the data of BLS benchmark revisions over time and a quick glance at the data shows that negative revisions are common during or directly after recessions.  The benchmark revision wasn't as important as last year, when it was revealed that BLS data was too optimistic by almost one million jobs, but it is still an indicator that the economic picture is not quite as rosy as many people would like it to be.

*Some people might argue that the monthly BLS numbers are more important because the direction of where the economy is going is much more important than its level.