This week’s “Daily Journal” guest columnist is Rick Henderson, managing editor of Carolina Journal.
RALEIGH — This morning we get the January jobs report from the U.S. Bureau of Labor Statistics. And if this January bears any resemblance to those of the past decade, the report will offer a misleading, possibly deceptive view of the employment situation in the United States.
The data highlighted in the BLS reports also fail to reflect the dynamism of the American job market, which rises and falls in predictable cycles every year.
Between December and January of recent years, U.S. nonfarm employment has fallen. A lot. On average, since 2001 the number of jobs has dropped by 2.87 million.
Most likely, when the employment report comes out, the BLS will report a gain — perhaps by as many as 200,000 to 300,000 jobs. If the January report is anything like December’s, we may see The Associated Press recycling its coverage of the December release, which reported a gain of 200,000 jobs. AP economics writer Paul Wiseman said this suggested the nation “may finally be in an elusive pattern known as the virtuous cycle — an escalating loop of robust job growth, healthier spending, and higher demand.”
Really? The December report showed employment — in unadjusted terms — fell by 219,000 jobs.
What is going on? The BLS press releases are regurgitated by nearly all media outlets, and they focus on a “seasonally adjusted” measurement of jobs. The Census Bureau, using the same methods as BLS, describes seasonal adjustment as “the process of estimating and removing seasonal effects from a time series in order to better reveal certain nonseasonal features.” Possible seasonal effects “include natural factors (the weather), administrative measures (starting and ending dates of the school year), and social/cultural/religious traditions (fixed holidays such as Christmas).”
So what actually happened from November to December of last year is that the BLS computer models expected a seasonal decline of 419,000 jobs. When only 219,000 disappeared, we “added” 200,000 jobs!
Meantime, if the BLS January report claims an increase of, say, 300,000 jobs, in actuality the nation would have in fact lost 2.6 million. By losing fewer jobs than expected, we experienced a “gain.”
The BLS publishes both seasonally adjusted and unadjusted numbers on its website. (Look for Table B-1.) But its press releases discuss the seasonally adjusted numbers, and the media rarely report the other ones.
The concept of seasonal adjustment is sound. It helps identify trends in the economy and is useful for statisticians and investors. But it doesn’t paint an accurate picture of what’s taking place in real time in the real world. And it’s easy to manipulate for political purposes.
Channeling Dan Rather, you could say relying on seasonally adjusted job counts to the exclusion of the original, or unadjusted, numbers is reporting figures that are accurate, but fake.
For instance, Alan Krueger, chairman of the President’s Council of Economic Advisers, cherry-picked some of the sectors that showed seasonally adjusted job increases in December. He noted that construction jobs increased by 17,000, retail trade by 27,000, and leisure and hospitality by 21,000. He cited those figures alongside an appeal to pass President Obama’s American Jobs Act, and didn’t say the numbers were seasonally adjusted.
The unadjusted numbers paint a much different picture. Construction decreased by 223,000, retail trade rose by 181,600, and leisure and hospitality fell by 57,000.
But construction employment should rebound by March, and when it peaks in July or August, there may be 800,000 to 1 million more construction jobs then than now.
It would be just as accurate — and more understandable — to say (for instance) that the nation lost 2.6 million jobs in January, but that was 300,000 fewer jobs than expected. Fat chance of that.