Professor Cole's Finance Blog

This morning at 8:30 a.m. EDT, just as it does on each first friday of a new month, the Bureau of Labor Statistics released its monthly employment report entitled "The Employment Situation," for the previous month, in this case, for September 2009.

As always, the employment report summarizes the results from two surveys--the Household Survey (HS), from which the headline unemployment rate comes, and the Establishment Survey (ES), from which the headline employment number comes. From the ES, the headline jobs number was a decline of 263,000, larger than consensus estimate, which was 175,000 or less, and much larger than the decline of 213,000 reported for August. From the HS, the headline unemployment number was 9.8%, up 214,000 from the 14.93 million unemployed reported for August and up from 6.2% in September 2008.

In other words, the employment situation continues to deteriorate, not improve, in spite of the $787 billion Stimulus Package, the $700 billion TARP, the $1.6 trillion deficit for the fiscal year 2009, which ended two days ago on September 30, the $1 trillion increase in the Fed's balance sheet and the fall of the Fed's target interest rate from 5.0% essentially to zero.

Why didn't the unemployment rate rise even more? To understand, we need to delve deeper into the report. Not widely known is the fact that the HS also provides an employment number, albeit one measured with greater error than the one from the ES. That said, this second employment number can be very informative, especially if it is widely different from the headline ES jobs number. This is because the HS employment number reflects self employment and employment at smaller and newer establishments not covered by the ES. In September, the HS employment number fell by 785,000 jobs, or more than three times the 263,000 decline in the ES number.

How is this possible when unemployment rose by only 214,000? It is possible because the remaining (785,000 - 214,000 =) 571,000 formerly employed persons left the labor force. In other words, they gave up. Remember, to be unemployed, one has to have actively looked for work during the previous four weeks. If you lose your job and give up, you're not "unemployed," you're "out of the labor force."

For this reason, the BLS also reports broader measures of unemployment in Table A12. U-6 is the broadest of these measures, including officially unemployed plus discouraged workers, plus those employed part time for economic reasons, plus marginally attached workers. This measure rose to 26.2 million, or 17.0% of the civilian labor force in September, up from 16.8% in August and 11.2% in September 2008. Also, the number of long-term unemployed (jobless for 27 weeks or more) rose by 450K to 5.4 million.

Among men, the unemployment rate rose to 10.3%, up from 10.1% in August, while, among women, the unemployment rate rose to 7.8%, up from 7.6% in August. This recession discriminates against men.

The BLS also revised its employment numbers for July (down 28,000) and August (up 15,000) as establishments that missed the original survey deadlines are counted when then eventually report. The July revision was on top of a downward revision of 29,000 announced last month.

Another statistic reported in the employment report is the length of the average workweek (in hours). This number fell by 0.1 hour to 33.0. Were the average workweek to return to its 33.6 hour level of a year ago, this would be the equivalent of hiring 2.75 million new full-time workers. The implication is that employers are much more likely to increase the average workweek for existing employees before they rehire laid-off workers, so that unemployment is likely to remain high even when the economy eventually begins to recover.

Also released today by the BLS is another report on benchmark revisions to the ES jobs numbers. This report contains preliminary revisions to the monthly ES numbers released each month. Each year, the BLS benchmarks employment estimates using state unemployment insurance tax records filed by virtually all employers. Usually, these revisions are up or down no more than two-tenths of one percent, but, this year, they are likely to be three times as large. The preliminary revision for March 2009 shows that the BLS "over-estimated" employment by 824,000 workers.

These sobering numbers are more than a "lagging indicator," as dismissive politicians and pundits refer to the unemployment rate. They are a continuing reminder that it is far too early to declare "mission accomplished" in dealing with the ongoing financial crisis. Until the BLS reports that the number of jobs is growing instead of declining, we remain in a real recession, even if Bernanke, Goolsbee and other members of the adminstration have declared that we are now in "technical recovery."