The government announced Friday that payroll employment rose a "robust" 262,000 in February—more than economists had expected and enough to trigger a stock market rally that took the Dow Jones Industrial Average to its highest close for four years.

But the household survey, which is the source of employment data by race and ethnicity, paints a very different picture.

According to that survey, household employment fell by 97,000 positions in February, or by 0.7 percent.

More importantly, this loss was not spread evenly: Non-Hispanics bore the entire brunt of February's household job loss, losing 110,000 positions. Hispanic employment rose by 13,000, or 0.7 percent.

February's skewed job market is not an anomaly. It mirrors conditions that have prevailed throughout the Bush Administration, and which we track in the VDARE.COM American Worker Displacement Index (VDAWDI). From the start of the Bush Administration in January 2001 through February 2005:

• Total household employment rose 2,373,000, or 1.7 percent

• Hispanic employment rose by 2,098,000, or by 13.0 percent

• Non-Hispanic employment rose by 275,000, or by 0.2 percent

Because so many Hispanics are immigrants and the children of immigrants, Hispanic employment is the best proxy we have for the impact of immigration on employment.

The ratio of Hispanic to non-Hispanic employment growth is a strong indication of how immigrants have fared relative to native-born workers in a particular month.

VDAWDI – the ratio of Hispanic to non-Hispanic job indexes – rose to 112.8 (=113.0/100.2) in February, up from 112.6 in January.  

Although our primary interest is immigration and its impact on American living standards, there are other reasons to worry about the validity of the widely cited payroll employment figures.

Figures released Friday showed that worker productivity grew faster in the last three years than in any comparable period over the last fifty years.

But the government's productivity figures are based on payroll employment, which has barely budged in recent years. Had household employment been used to calculate productivity growth, it would be anemic—and the specter of inflation and economic stagnation far more noticeable.

Both in terms of immigration policy and macroeconomic policy, it's likely that things are not as good as they appear to be—and that Washington is not as worried as it should be.

Edwin S. Rubenstein (email him) is President of ESR Research Economic Consultants in Indianapolis.