April 4, 2008

Old Media’s Seasonally Ignorant Employment Reporting

Did you know that 574,000 and 1.1 million more Americans had jobs in March than in February and January, respectively?

Seriously (data can be retrieved from this BLS page; select the very first “not seasonally adjusted” table):


Now the fact remains, as you can see, that job growth during the past two months is nowhere near as great as it was during the same two months in 2006 (1.91 million) or 2007 (1.58 million). This goes a long way towards explaining why total employment, when adjusted for seasonality, fell 80,000 during March, and by 232,000 during the first quarter.

There’s no denying that the employment situation has been deteriorating for several months, and I’m not trying to minimize that. What I am saying is that the “employees were thrown out on the streets during March” narrative cooked up by Old Media today, including the Associated Press’s Jeannine Aversa, is clearly false, either because Old Media reporters and their editors don’t understand a concept as basic as seasonality, or they don’t want to.

Here are the relevant offending paragraphs from Aversa’s report today (April 8 note — link is to host copy reflecting report as originally seen; original was updated but not saved by AP at its “hosted.ap.org” site; the relevant raw-data job changes are in italics):

Employers buffeted by talk of recession slashed 80,000 jobs in March (actually, employers added a lower-than-usual 574,000 jobs in March) …..

….. Job losses were widespread in March. Construction, manufacturing, retailing, financial services and various business services all racked up losses (actually, every sector noted except manufacturing had gains). That overwhelmed gains elsewhere, including in education and health care, leisure and hospitality as well as in government (this sentence is thus false on its face).

In March, construction companies cut 51,000 jobs (actual was +49,000), factories eliminated 48,000 positions (actual was -19,000), retailers cut payrolls by more than 12,000 (actual was +53,000). Professional and businesses services lost 35,000 jobs (actual was +67,000) and temporary help firms cut nearly 22,000 jobs (actual was +20,000). Financial firms chopped 5,000 jobs (actual was +5,000).

When government hiring was removed, the numbers looked even worse. Private employers shed 98,000 jobs in March (actual was +474,000; 105,000 jobs were added in government during March, vs. the +18,000 seasonally adjusted amount inferred).

Beyond that, there’s the annoying presumption on the part of Aversa and others that job reductions always occur because employers “cut” jobs. If a person leaves his or her job voluntarily or retires and isn’t replaced, did the employer just “cut” or “chop” a job (with the heavy implication that some innocent person was thrown out of work)? Of course.

Naturally, seasonality goes in both directions. For example, in January 2008, there were 1.7 million fewer jobs than in December 2007. This is a normal post-Christmas season occurrence, and further illustrates why seasonally adjusting data is important and necessary.

Again, this isn’t to say that the economy isn’t in a rough patch. But it’s a long way from March’s easily reportable reality — unadjusted job growth occurred, but not as quickly as in previous years, causing seasonally adjusted job growth to be negative — to the patent fiction of “blood on the streets” being foisted on the public today.

Thousands of Old Media outlets already have and will continue to mindlessly relay the job-cut narratives of AP, the New York Times (“Unemployment Rate Rises After 80,000 Jobs Cut”), and others. No doubt politicians leverage it and pile on. But all the newsprint used, all the bandwidth burned, and all the uninformed rhetoric won’t change the fact that in the real world, those cuts simply did not happen.

Cross-posted at NewsBusters.org.


No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.