The Book of Jobs
Employment is a campaign issue, but what do the surveys really tell us about the jobs market?
by Irwin M. Stelzer
01/22/2004 12:00:00 AM
IF YOU WANT TO KNOW why ordinary folks find it difficult to understand what economists are saying about the American economy, consider the question of jobs. We know a few things. We know that jobs are such an emotive political issue that one candidate for the Democratic presidential nomination promises that if elected his three top priorities will be, "jobs, jobs, jobs."
And we know that when there is slack in the labor market, upward pressure on wages is reduced. This allows Alan Greenspan and his monetary policy colleagues at the Federal Reserve Board to declare inflation under control and to keep a lid on interest rates.
Finally, we know that when GDP is growing, as it now is, but the number of jobs isn't, each worker must be producing more, driving up productivity. Indeed, we are being told that output is rising so fast, and new job creation is so low, that we are in the midst of "a productivity miracle," a wondrous trend that assures higher profits, continued low inflation, and a long period of economic growth.
THE TROUBLE IS that no one seems to have a clear, unambiguous picture of what is going on in the jobs market. And with good reason. The U.S. Department of Labor regularly surveys employers to ask how many people they are hiring, and how many they have laid off. It also surveys households to find how many people say they are in work. Any sensible person would expect the results of these two
surveys to be more or less the same: If employers are hiring, more households should be reporting that their members are finding jobs.
Alas, the world of economic statistics is not so straight-forward. In the past year, employers reported a net loss of over 70,000 jobs, while households reported a net gain of over two million. Believe the second figure. Here's why:
The survey of employers, which is the more widely reported of the two, is so bleak because of the way it is designed. Assume, for example, that a factory employs some 3,000 workers making widgets, and 300 workers in the on-site canteen. Management decides to outsource the food service. When this employer next responds to the employment survey, he will report a job cut of 300 and, best of all, that he is now producing all of the widgets that he once produced with a workforce of 3,300, but using only 3,000 workers--a bogus productivity miracle.
Even more misleading is that fact that the new firm formed to handle the canteen catering is not picked up in the employment survey, which does not cover either new firms or the newly self-employed.
Turn now to the household survey. The canteen worker reports that he most certainly has a job, even though it is with a new employer. And if the new catering firm upgrades the quality of the fare on offer, so that fewer workers bring their own homemade lunches, forcing the caterer to add workers, those newcomers to the job market will not be picked up in the employment survey, but will be recorded in the household survey. |
