Want the news summarized?
Subscribe to The Morning Report.
The housing beneficiary sectors on which I have often focused lost a combined total of 7,300 jobs between them. Of these industries, retail trade has bounced back noticeably in the last couple of months and ended up just barely smaller than it was a year ago. The construction and financial industries, the latter of which includes real estate, fared worse and were still respectively down 5.5 percent and 2.3 percent from August 2006.
The remainder of the economy added 16,500 jobs, growing by a healthy 1.7 percent.
As in recent months, however, the weak housing-related sectors dragged overall employment growth down. In total, the San Diego economy grew by 9,200 jobs or .7 percent.
Looking closer at the individual sectors, the leisure and hospitality industry once again added by far the most jobs to the economy, followed by government and professional services. Construction, financial activities, and manufacturing were the big three losers.
Here’s a look at the overall composition of industry employment in San Diego. Please note that I did not bother to match the colors in the next chart with those of the previous one. Do I look like the kind of guy who has lots of time to play around with charts?
I’d like to end with a word about all of the above statistics. The agencies that provide them (the Bureau of Labor Statistics appears to get the data from the EDD) do so based on samplings of employer records. The resulting numbers only account for businesses that the agencies know about, however, so the agencies employ some statistical modeling in an attempt to fill in the blanks. For this they use the so-called “birth/death model.” The titular births and deaths are not those of people but of businesses which have either formed or folded too recently for the agencies to know about it. Because the agencies can’t measure such activity in real time, it is assumed that business are being born and dying at about the same rate they have in the past.
It ends up that the birth/death model has a significant impact on the numbers. For example, according to the BLS, nationwide employment in August fell by 4,000 jobs from July. This estimate includes 120,000 jobs added by the birth/death model, meaning that without this statistical adjustment, August employment would have been down by 124,000 jobs instead of 4,000. Big difference.
Because this adjustment is based on past job market behavior, the adjustments are probably sound as long as the present is acting like the past.
But sometimes, the present doesn’t quite play along — just ask the purveyors of mortgage-backed credit instruments. Should such a divergence take place, the statistical adjustments may start throwing off bad signals.
I suppose the point is that this monthly employment data, while interesting to chart and analyze, should — like so many other things in life — be viewed with a skeptical eye.
— RICH TOSCANO