Jewish World Review April 9, 2004 / 19 Adar, 5764
The jobless recovery may not have been jobless
You may ask, doesn't the Bureau of Labor Statistics' establishment survey make it clear that there has been job loss? The establishment survey has long been taken by economists of all political stripes to be the best indicator of job gains and losses. And the BLS is a highly competent organization, impervious to political influence.
But the employer survey is not the only measure of jobs. The BLS also conducts a household survey, which shows a net job gain of 612,000 from January 2001 to March 2004.
The contrast is even sharper if you disaggregate the numbers and look at job gains and losses in the recession, officially defined by the National Bureau of Economic Research as occurring from March 2001 to November 2001, and at the postrecession period from November 2001 to March 2004.
This discrepancy between the population and household surveys has persisted since the late 1990s and economists have not been able to explain it. But economist Tim Kane of the conservative Heritage Foundation gives several reasons for doubting that the establishment survey provides the more accurate picture of job gain or loss:
Kane concludes that "the best measure of job growth now comes from the smoothed total employment recorded in the household survey." Many economists will still disagree. But most will probably agree when he says, "Policymakers and analysts should treat payroll data with caution when making comparisons to employment levels in 2001 and earlier years."
The BLS and other government statistical agencies do a superb job. But the economy they are attempting to measure changes over time, and changes in the underlying reality tend to make any statistical measure over time less accurate. The establishment and population surveys make it clear that there was significant job loss during the recession of March to November 2001 and particularly after the September 11 attacks. But they are at variance of what has happened since November 2001: a small job loss, says the establishment survey; a much bigger job gain, says the population survey.
Disinterested observers might be wise to assume it was something in between and to keep in mind both surveys when they are trying to understand whether the economy is producing a net gain in jobs. Meanwhile, the BLS has been conducting new surveys to try to understand the reasons for the discrepancy between the two surveys and, perhaps, to come up with new methods that will provide more accurate measures.
All this may be moot politically. The establishment survey has long been considered the more accurate measure, and that verdict has recently been reaffirmed by the Congressional Budget Office and Federal Reserve Chairman Alan Greenspan. The Bush administration and the Bush-Cheney '04 campaign missed any chance they had last year to argue that the household survey was more accurate; it is too late in the political season now to do that convincingly. And if the April and May establishment surveys show job gains anything like the 308,000 in the March survey, the administration and Bush-Cheney '04 will have plenty to crow about, and John Kerry and the Democrats will have to revise their economic talking points. The jobless recovery, it seems, may not have been jobless at all.
Every weekday JewishWorldReview.com publishes what many in Washington and in the media consider "must reading." Sign up for the daily JWR update. It's free. Just click here.