Friday, February 8, 2013

Fixing the H-1B immigrant worker visa program

Ross Eisenbrey of the Economic Policy Institute has an op-ed in the NYT, claiming that the US does not need to do anything to the current H-1B visa system. This is a response to Senator Orrin Hatch's bill to allow up to 300,000 more H1-B visas a year. This would be a dramatic increase over the current cap of 65,000 visas/year (excluding the first 20,000 visas issued to graduates from US universities, and those employed by US universities/government labs/non-profit research organizations.)

Eisenbrey claims that the proposed expansion would flood the market with "indentured foreign workers." That's an extreme claim, though somewhat rooted in the fact that employees are dependent on their employer, and may not be able to find more skill level-based employment till such time as they are approved for a Permanent Resident visa. For Indian and Chinese workers, this process can take over ten years. (Six years on an H-1B visa, followed by the time it takes to adjust the employee's status to PR, under the EB-2 category.) If the employee wants to switch employers, the new employer has to petition USCIS all over again, and this may or may not be subject to the afore-mentioned cap. (I don't think it is, but it is still a laborious process.) Any movement toward a Green Card is canceled by switching between employers, and the employee has to start all over again.

Among Eisenbrey's other claims are that if anybody is talented enough to get a job, they are almost guaranteed a visa. That's not quite true - I was unable to get a H-1B visa despite having a PhD in Engineering, and a job offer from a private company here in a highly-specialized field, because of the annual cap. I managed to get an O-1 visa (another painful process), only because of my research and publication record. A less-qualified individual - say, a software engineer - would not be able to get an H-1B visa after the cap has been met.
(Some of them might get an L1 - in 2010, the US issued almost 75,000 L1 visas, compared to 117k H-1B visas, and just 8589 O-1 visas. PDF. Still, it speaks to the demand for such workers.)

So the next claim is that salaries have not exploded in keeping with demand - just 4.5% between 2000 and 2011 for computer- and math-related fields for workers with a college degree. That's an interesting argument, though I'd like to see the actual trend over that period, which includes the Great Recession. Besides, one of the big inequality arguments is that wages for most workers have been largely stagnant since 1980. So an increase of 4.5% over a decade that included the Great Recession seems not too shabby. Of course, another factor to consider is that these are likely the average raises, and the composition of the work force might have changed in that time. Bottom line: I don't think this statistic is as clean-cut as it is presented.

Another claim is that the USA has "too many high-tech workers", based on the comparison of over 9 million STEM degree-holders, with 3 million who have a job in one. That is a conflation of STEM and high-tech. Not every STEM graduate can write code, and not every STEM graduate really wishes to be employed in the field of their qualification. A better comparison might be to look at graduates in fields directly relevant to high-tech fields.

And for high-tech workers, Eisenbrey finds the unemployment rate is 3.7% - low, but apparently more than twice as high as it was before the recession (references and links would be nice). There's a decent argument here, though I can't find that particular statistic in a quick FRED search - the closest I get is Bachelor's and higher degree-holders, 25 years and over: 2% pre-GR, and 3.9% at present. Still, saying "unemployment rate DOUBLED!" seems alarmist considering high-tech workers are still doing quite well at just a 3.7% unemployment rate.
(Aside: computers and software is a field where someone out of work can quickly get out-of-date. So, it may be possible for a company to look for, and not find, qualified people, even if there are American workers out of work. It's not exactly like other, conventional fields.)
(Further, does Eisenbrey/EPI also claim that Americans should not get a college education? After all, the unemployment rate for Bachelor's degree-holders is almost double that pre-GR!)

Overall, though, I am a little surprised that while Eisenbrey shows the flaws of the H-1B visa system, he doesn't propose any way to improve the plight of H-1B visa holders - an area where reforms are sorely needed. He'd apparently let the system be - let indentured status continue.

On the other hand, Senator Hatch's proposal is clearly geared toward big corporations - a GOP constituent. Given the problems with the H-1B system, I am not sure that's a great idea - but unlike Eisenbrey, I can't let the system be, either. If there's going to be comprehensive immigration reform, an attempt should be made to help workers.

An alternative proposed by some Democrats, is to "staple" a Green Card to foreigners graduating with a MS or PhD from US universities. That sure would have helped me. It is definitely a better solution than Senator Hatch's proposal. Further, it would help foreign graduates escape "indentured servitude" at the hands of American employers - a goal Eisenbrey and the EPI might be expected to support.

Now, there might be some fears that some American universities could capitalize on such a provision, and become degree-granting mills, as Matt Yglesias points out. One way might be to restrict the "stapled Green Cards" to graduates of well-known research universities. It's not simple - who decides what is well-known? The Carnegie Foundation apparently doesn't! But it would be a better system than what we have at present, and that's a start.