Labor's Lost Jobs
Friday brought good news on the economic front, with the Labor
Department reporting that 308,000 jobs were added last month. But
total payroll employment still appears sickly if one looks back
over the last three years. There have been two million jobs lost
since March 2001. Or have there? It depends, as usual, on which
statistics you use. And there is reason to doubt the numbers from
the payroll survey, which the Labor Department has used since 1939,
because they give a misleading picture of the 2004 economy.
The payroll survey counts jobs, not workers. But counting payroll
jobs is a questionable way of measuring America's evolving work
force, especially in light of declining job turnover. The payroll
survey's biggest problem is that it systematically double counts
workers when they change jobs. Since somewhere between 2 percent
and 3 percent of the work force changes employers every month,
payrolls tend to be noisy. The illusion of lost jobs in recent
years occurred because job turnover declined after 2000, first with
the recession, then even more sharply after 9/11. As a result, 1
million jobs have been artificially "lost" in the payroll survey
Despite last month's jobs surge, the payroll survey remains
stubbornly out of whack with other economic indicators, even other
labor indicators. Unemployment has been very low and is now near
what economists call a "natural" rate. Real earnings rose by 3
percent over the last three years. Jobless claims are 10 percent
below their historical average, and that's without adjusting for
The sharpest contrast can be seen by looking at the Labor
Department's household survey, which shows a record high level of
total employment. This survey reported an employment level of 138.3
million as of March - 600,000 more working Americans since
President Bush took office in 2001.
Since the recession ended in November 2001, the payroll survey has
reported 323,000 fewer payroll jobs, but the household survey has
found 1.9 million more overall jobs. Common sense tells us that
payroll jobs aren't the end-all, be-all of jobs in the new economy.
Economists reflexively like payroll data because it has a bigger
sample, but quantity doesn't always ensure quality.
An even bigger problem with the payroll survey is the evolution of
what constitutes work. We can think of the payroll survey as
counting all workers at traditional firms, plus some workers at
start-up companies who have payroll records. But the payroll survey
doesn't count individuals who are self-employed - despite the fact
that their ranks have surged by at least 650,000 in just two
Then there are limited liability companies, a new form of business
the Joint Tax Committee says is growing at an annual rate of 34
percent. Consider, too, the rise of consultants, like a marketing
executive who was once on the I.B.M. payroll but who has switched
to a freelance consulting role with I.B.M. None of these employees
are counted in the payroll survey.
Even so, this is not the first time the survey has been off. That's
why the Labor Department warns against using the real-time payroll
figures in the footnotes of its monthly release. In 1992, the media
proclaimed a jobless recovery based on preliminary payroll data.
Only later did benchmark revisions correct the data that the public
sees today, which show the net creation of 900,000 jobs in the year
prior to the 1992 elections. The next major payroll revision won't
occur until January 2005.
Payroll employment may well continue to surge in the next few
months. If it does, it will be a vindication of economic optimists
who have pointed to the household survey as the better jobs
indicator of the last three years. We should be prepared for the
job numbers in the payroll survey to permanently look anemic
compared to other measures. This is the brave new economy, and the
work force of 2004 can no longer be measured by an outdated
definition of a job.
Tim Kane is a research fellow in macroeconomics in the Center
for Data Analysis at The Heritage Foundation
First appeared in The New York Times.