- The Washington Times - Wednesday, February 11, 2004

Once again the employer-reported payroll job numbers and the household survey data on civilian employment tell different stories.

Nonfarm payroll jobs rose a modest 112,000 in January, the fifth monthly gain in a row and the first six-digit increase in more than three years. Since August, payroll jobs were up by 366,000. For 2003, new benchmark revisions to the data show a comparatively brighter — or less gloomy — picture.

From January to December on the old basis, payroll job loss was estimated at 232,000, whereas the revised data show a smaller loss of 147,000. From August to December 2003, the previous data showed a gain of 243,000 compared to an increase of 254,000 in the revised series.

According to the alternative household survey, employment started the new year with a bang. In January, total civilian employment leaped up by just short of a half-million after adjustment for new population controls — a spectacular performance.

Thus, the gap between the two employment surveys widened further, continuing a pattern that has characterized the current economic recovery and has yet to be fully explained.

Economists have noted the payroll data do not include the self-employed and miss many of the increasing number of individual contract workers who benefit from employer outsourcing, although just how much of the survey gap this might explain is uncertain. The self-employed are subject to underreporting in the household survey as well.

Since the end of the recession in November 2001, total employment as measured by the household survey (adjusted for new population controls) has risen 2.2 million while jobs as measured by the payroll survey declined by about 700,000 — a gap of 2.9 million.

Differences in definitions between the two surveys, according to the Bureau of Labor Statistics, explains only a half-million, or 17 percent of the gap. Economists in and out of government generally put more weight on the payroll survey, mainly because of its larger sample. But the truth lies somewhere between the two survey results, as Federal Reserve and other economists have noted.

If twice as much weight is given to the payroll data as the household numbers, the conclusion is that employment has risen, not declined, during the current economic expansion.

Last month, the ratio of employment to population increased to 62.4 percent, a healthy sign. This ratio can be thought of as a labor utilization rate, similar in concept to a capacity utilization rate for plant and equipment. The employment numerator represents actual usage, while the working-age population denominator is a measure of potential employment.

When employment rises faster than population growth — e.g., as the labor utilization rate increases — unemployment declines and the labor market tightens. The decline in joblessness can consist of active jobseekers, as officially measured, and/or hidden unemployment — those waiting outside the labor force for job opportunities to improve.

The unemployment rate continued to drift downward in January, and at 5.6 percent was well below last June’s cyclical peak of 6.3 percent and was the lowest jobless rate in two years.

One might have expected a bigger drop in the official unemployment rate considering the nearly half-million rise in total employment and with both numbers coming from the same household survey.

The January employment gain was about 300,000 in excess of normal population additions to the labor force, but the official unemployment count declined by only a modest 74,000. Who got the rest of the jobs? The hidden unemployed — those waiting on the sidelines for the job market to improve — whose numbers declined by some 200,000, as reflected by a rise in the labor force participation rate.

The employment-population ratio, reported monthly by the BLS, is useful in another way. Because of the way unemployment is officially measured, the unemployment number sometimes confuses by moving in the same direction as employment rather than oppositely.

A shorthand way to get around that problem is to see whether the employment-population ratio rose or fell. If it rose, it’s fair to conclude joblessness declined even if officially measured unemployment did not show a decline. Someone had to get the jobs — and if it wasn’t those in the official jobless count, it was the hidden unemployed.

Conversely, if the employment-population ratio and officially measured unemployment decline together, hidden unemployment must have risen. And when the employment ratio is unchanged, unemployment by this concept is also stable.

The employment outlook this year is promising. Nonfarm business productivity growth slowed in the final quarter of last year to a 2.7 percent rate, down from about a 4.5 percent average growth for 2002 and 2003. If productivity growth maintains its typical pattern of tapering off as the business cycle progresses, with the economy expected to grow at a 4 percent to 5 percent rate this year, employment should also resume a more normal recovery pattern, rising an average of 200,000-plus per month.

Alfred Tella is a former Georgetown University research professor of economics.

Copyright © 2019 The Washington Times, LLC. Click here for reprint permission.

The Washington Times Comment Policy

The Washington Times welcomes your comments on Spot.im, our third-party provider. Please read our Comment Policy before commenting.


Click to Read More and View Comments

Click to Hide