Dr. Norm Matloff writes to his email list (I've added links):

In recent years there have been various proposals to grant automatic green cards to foreign STEM graduates of U.S. universities, a notion generally sold with under the banner "We shouldn't educate them and then send them home to compete with us." I've strongly opposed these proposals, for reasons I'll review later, and will also explain why the "send them home" argument is a red herring, but my main focus here will be to present my own counterproposal.

There is apparently a feeling in DC that some kind of auto-green bill will soon pass. For brevity, let's call it the Don't Send Them Home Act of 2011 (DSTHA2011). I'll comment on how likely DSTHA2011 is to pass, based on my very limited information, and then, more importantly, introduce a proposal of my own.

The October 22 issue of the Wall Street Journal carried an article titled, "Visas Could Aid Graduates." Here are some excerpts:

Momentum is building in Congress toward offering expedited green cards to people with advanced scientific degrees, addressing complaints from companies that say the U.S. is training highly skilled workers only to lose them to other countries.

Rep. Lamar Smith (R., Texas), chairman of the House immigration subcommittee, said he plans to introduce legislation providing up to 10,000 visas a year to foreign students graduating from U.S. universities with doctorates in engineering, information technology and the natural sciences.

Many Democrats, including President Barack Obama, support increasing the number of visas for workers with advanced training in those fields. But efforts to address the issue have stalled amid a larger dispute between the two parties over whether to provide a path to citizenship for people already in the country illegally, which Democrats favor. ...

In an interview, Mr. Smith said students who would get visas under his bill "really are going to contribute a great deal to our country and to the economy, as well." He said he would limit eligibility to those with doctorates from U.S. research universities, though others think the visas should be available to those with masters' degrees, too.

...

In another signal of momentum, another Republican, Rep. Raul Labrador of Idaho, introduced a similar bill this month. His bill took sections from legislation offered in June by the top Democrat on the immigration subcommittee, Rep. Zoe Lofgren of California, who represents Silicon Valley.

...

Ms. Lofgren added provisions to win support from the Congressional Hispanic Caucus, including one that would allow students who were in the country illegally during high school to qualify for foreign student visas if they are pursuing science or technology degrees.

...

So, will DSTHA2011 become a reality or not? The above article talks of "momentum," yet the Lofgren comments seem to say that the current proposals may fall victim to the same political impasse that blocked previous proposals. I remember two or three years ago one DC insider telling me, "This train is leaving the station. Are you going to get on it or not?"

My contact turned out to be wrong about the "train," of course (partly out of bias, as he is a strong vocal support of auto-green). And as mentioned, the same sticking points appear to be in effect now.

And yet, today a critic of foreign tech worker programs wondered aloud to me whether the activists on his side of the issue (I don't count myself as an activist, by the way) ought to endorse the Smith bill. The reasoning was that Smith would restrict DSTHA2011 to PhDs only, not Master's degree holders. Since there aren't that many PhDs produced each year in STEM, the impact would be minimal.

I replied that such an endorsement would undermine the activists' longstanding assertions, sending the message that the industry's "Johnnie can't do math" claims are valid after all. And if that is the case, the industry would say, why limit the bill to doctorates?

But more to the point, the industry would never agree to limiting to the measure to PhDs anyway. The reason is that even though the industry has usually highlighted the doctoral level in its PR—"60% of U.S. STEM doctorates go to foreign students, so we need H-1Bs!"—the fact is that they are actually interested in the Master's level; PhD is overkill for them in the vast majority of cases.

I've mentioned Intel before, for instance. I've quoted two Intel recruiters by name who told me "Intel isn't very interested in PhDs" when I said I knew a new PhD in EE who was looking for work. I've also mentioned that an Intel official (who shall go nameless here even though he did not ask for anonymity) told me in 2006 the starting salaries for new PhDs were mainly in the $95K-110K range; comparing that to the PERM (green card sponsorship) data for Intel showed that Intel does not in fact hire many foreign doctorates.

So the industry won't settle for a bill limited to doctorates. They'll trot out a plethora of poster children "innovators" who lack doctorates. In 1996, then-Senator Alan Simpson told the San Jose Mercury News what he was dealing with concerning tech worker visa reform:

 

I was working with the business community...to address their concerns, [but] each time we resolved one, they became more creative, more novel. [The lobbyists] distorted everything we were up to, everything.

Expect the same here.

Well, then, what kind of bill WOULD make sense? Let's look at the industry's own stated goals with DSTHA2011:

1. They want more STEM degree holders at the graduate level.

2. They want more innovators.

Goal 1, if the industry were sincere, would actually be quite easy to achieve. As the testimony of Texas Instruments at Smith's recent hearing revealed , there is no shortage of STEM students at the Bachelor's level. The solution is then simple: Give the students incentives to continue their education at the graduate level. As I explained in my posting at the above URL, TI by its own admission does basically nothing along these lines. I discussed this in 2006 with the Intel executive mentioned above, and he said he was working on it, but today, five years later, still NOTHING has been done, to my knowledge. To be blunt, I simply don't regard the industry's claims regarding Goal 1 to be sincere.

But what about Goal 2? Aside from the implication that Americans don't innovate—which is nothing less than a gross rewriting of tech history—the goal itself is fine.

So, instead of a blanket awarding of green cards to achieve Goal 1, if any immigration-expansionary legislation is warranted at all (which I dispute), it should focus on Goal 2.

Existing green card law includes provisions for fast tracking the "best and brightest." But such provisions could be expanded, with broader definitions of that term.

How could that term be codified? Here is my proposal: A simple, clean and reasonably tight definition would be based on salary. Deem any new foreign STEM grad degree holder as "best and brightest" if the employer's offered salary is in the top 5% of all new grads in that student's field and degree level. One might account for economic sector as well.

This idea is similar to the proposal made by some that H-1B work visas (as opposed to green cards) should essentially be auctioned to highest bidder, in the sense that employers offering the highest salaries would have priority in the awarding of visas. This market-based approach is eminently reasonable, and should (assuming sincerity on the players' part, a debatable question) appeal to the free-marketeers who take industry's side in Congress. In principle, the higher salaries should (at least roughly) correspond to higher levels of talent and greater potential for contributing to the economy.

Well, the same principle could be applied to a bill addressed at Goal 2, and I so propose. (I've made market-based proposals for H-1B in the past too; see my 2003 University of Michigan Journal of Law Reform article.)

So would my proposal send the non-best/brightest home, "to compete with us"? Well, we shouldn't care much if they're not top talents, but in any case, such people could still go through the ordinary H-1B and green card systems. And the "compete with us" sound bite is a red herring to begin with; research by UCB's AnnaLee Saxenian has shown that most immigrant engineers actively help the industry back home, through providing consulting, technical and funding—in spite of having green cards and in many cases U.S. citizenship. Indeed that situation is actually worse than people going back home after graduation, as the immigrants have more expertise to contribute, having worked in cutting-edge U.S. firms.

My proposed legislation would also impose some conditions. Here are some that come to mind for now:

1. Bills to expand the H-1B and EB-series green card programs always include a feel-good employer user fee, aimed in getting more minority students into STEM. I'd like more minorities in STEM too, but in view of Goal 1, the fee really ought to be used to fund graduate fellowships, with stipends generous enough to get American STEM Bachelor's degree recipients to continue on the graduate level.

2. Visas for foreign post docs should be limited to two years. That's long enough for the "best and brightest" to establish their research credentials and be hired into permanent jobs. As it is now, the post doc system is an absolute disgrace, creating enormous disincentives against Americans' pursuing doctorates in the lab sciences. A large part of the problem is due to the huge numbers of foreign post docs, whose presence keeps the researchers poor and with uncertain futures. We should keep the best and the brightest foreigners, which my provision would accomplish, but not continue the current outrage.

3. Make it illegal for an employer to reject a job applicant merely on the grounds that he/she is overqualified. As I've emphasized so much, the foreign worker programs are at their core designed as a mechanism for age discrimination against Americans. This provision could probably be circumvented by employers to some degree, but it would be of some help.

As I mentioned in my report at the above URL, the folly of having a broad blanket granting of green cards is illustrated perfectly by an example given by, ironically, one of the foremost advocates of auto-green: Vivek Wadhwa. He cites one Girija Subramaniam, who was hired by TI after she earned a Master's in electrical engineering at the University of Virginia in 1998. So, did TI hire Subramaniam for innovation, for R&D? No! They hired her as a test engineer, the most boring, mundane, mindless, NON-innovative job an engineer can be assigned to. The idea that TI is afraid that Ms. Subramaniam will return to India and "compete with us" there—as a test engineer!—is patently absurd.

Again, the only legislation we really need on foreign worker programs is to tighten them, not expand them. But if Congress feels that they must do something to placate the industry, then as someone who has always strongly supported bringing in the world's best and brightest, I offer my own Goal 2-oriented proposal, outlined above. Any takers?

Norm