Diana Furchtgott-Roth Dissembles On H-1b Visas In The New York Sun
01/03/2008
A+
|
a-
Print Friendly and PDF

The New York Sun has a guest opinion column that contains so many factual errors and distortions I doubled checked to see if the original source was of the Wall Street Journal. When you see that it was written by a former "Chief Economist" at the Department of Labor, [Diana Furchtgott-Roth , send her mail]it will reassure you of government competence. [Desperately Seeking Visas, By Diana Furchtgott-Roth, January 2, 2008

The op-ed states:

"Every year, UScIS issues 65,000 H-1b temporary visas for skilled workers certified by the Labor Department out of approximately 630,000 approved applications from employers, with applications showing no sign of abating. Immigrants who hold H-1b visas must return to their home countries when the job ends."

First of all USCIS does not issue visas, the State Department does. USCIS approves visa applications and the Labor Department approves Labor Condition Applications (LCA)

Normally, I would not pick a nit like this. When you write pieces like this, you often have to oversimplify to fit into space limitation. The problem here is that much of the article based upon this erroneous statement.

USCIS has been approving around 117,000 to 130,000 new H-1B visas a year [PDF, see below]. So not only does the author have the process mixed up, the numbers are wrong as well.

The 630,000 figure represents another interesting twist of numbers. The author here apparently wants the reader to believe this figure represents the demand for visas. The number of approved LCAs in 2006 was about 380,000. If you add the number of workers requested on all the approved LCAs you get 630,000. LCAs are not tied to a particular worker and an employer can specify any number of workers on an LCA. While the LCA give statistical evidence of what is going on in the H-1B program, there is not a one-to-one correlation between a LCAs and visa applications.

In the 2005, Infosys submitted 1,145 LCAs covering 110,000 workers. To me this indicates Infosys intended to import many H-1B workers. This author implies it means Infosys wanted to import 110,000—I don't think so.

The author then goes on to state:

"Foreign workers must be awarded labor certification from the Labor Department. This process requires the prospective employer to affirm that he has determined that no American workers are available to fill the position, and that the foreign worker will be paid the prevailing wage. "

 

This is a widely circulated myth about the H-1B program. The reality is there is no recruitment requirement as part of the H-1B labor certification process.

In addition, the law allows the employer to determine what the prevailing wage is; the employer can use nearly any source; and the law limits the approval of an LCA to checking that the form is filled out correctly.

In short, the 630,000 approvals represents the amount of paper that was pushed, not how many workers were approved for visas.

The author continues with a lobbyist talking point that come straight out of How to Lie With Statistics.

"This [65,0000] figure represents a minuscule portion of the U.S. labor force of 154 million. Even if the quota were raised to 150,000 annually, that would be less than one tenth of 1% of the labor force. "

One would hope the Chief Economist would know that the jobs for which H-1B visas are eligible represent a only a fraction of the total jobs in the U.S. Using the entire labor force as a measurement is a deliberate distortion intended to come up with a figure that dilutes H-1B's actual impact.

Between 1999 and 2005 (First year data available to last—coincidentally H-1B visas can be renewed up to six years.) the number of computing jobs in the U.S. grew by 332,660 according to the Bureau of Labor Statistics. Over that same period USCIS/INS approved 330,524 H-1B visas for computer workers.

So if you look at the occupations where H-1B visas are used, their impact is enormous. Here the Chief Economist has come up with a distortion intended to mislead.

 

Print Friendly and PDF