The U.S. Needs a Way to “Score” Immigration Bills

By David North on July 29, 2011

This week’s hearing of the Senate Immigration Subcommittee, reported in an earlier blog, suggested to me that the nation needs a neutral system for “scoring” the impact of individual immigration bills on population growth.

In other words, were we to pass a given bill, how many more people would be brought to the nation, over what period of time?

There is no such system for people, currently, but there is one for dollars.

If a member of Congress suggests a funding cut or a tax-raising proposal it goes to the bipartisan Congressional Budget Office (CBO) where experts figure out how much money it will cost, or raise. These so-called “scores” are generally regarded as more reliable measures of the fiscal impact of legislation than statements made by the sponsors of the bills.

This is a system that Congress has created, and in effect, imposed on its own members. It is very useful in the field of government finance and it would provide an excellent model for an immigration bill scoring system. Scoring is often mentioned in the current hot debate about increasing the debt limit.

During the Subcommittee’s hearing on immigration there were suggestions, made with approximately equal fervor, that:

  1. We should eliminate the country of origin restrictions that make an Indian or a Chinese scientist wait for half a dozen years to get a green card under the overall annual cap of 140,000 new employment-based (EB) visas;

  2. All aliens getting advanced high-tech degrees from U.S. universities should receive an automatic green card. Sometimes this idea is limited to people with doctorates and sometimes -- more rashly -- it includes master’s degrees. There apparently would be no numerical limits on these visas.

  3. Within the framework of the EB 140,000 limitation, we should not count family members of the workers, or perhaps that pass should be given only to spouses and kids of high-tech workers, the subject of the hearing.

What struck me as I attended the hearing (via my computer) was that no one said: “Hey, these proposals will have remarkably different impacts on our total population, and on the total flow of immigrants.” Nor is there any agreed-upon way to score them for demographic purposes. So let’s see how such a system would work, if some neutral authority had the task of scoring the proposals. Of necessity there will be some detailed calculations, based on existing data, and some speculation.

Proposal A: As background, the per country limit within the 140,000 EB annual allocation of worker and worker-family visas is 7%. This per-nation allocation notion dates back to an unattractive period in our immigration policy history. Starting in the ‘twenties, we had firm country-of-origin quotas, designed to facilitate the immigration of northern Europeans, limit the immigration of southern Europeans, and just about eliminate any Asian or African immigration. (Western Hemisphere immigration was not a concern, being relatively minor, in those days.)

The 7% provision, which also applies to numerically-limited family (or nepotism) immigration, was designed to prevent any single nation from dominating the immigration flow. What it does with high-tech workers, usually in the U.S. on H-1B visas, is to slow -- by as much as six years -- their transition from nonimmigrant to immigrant status, if they are from India or China.

Computer World’s Patrick Thibodeau, an astute reporter on immigration matters, noted, “While eliminating the cap could reduce wait times for Indian and Chinese workers . . . it could also increase the wait times for applicants from other countries . . .”

Exactly. So, unless you have (as I do not) strong convictions that high-tech Chinese workers from Taiwan and Singapore are much superior to high-tech Chinese workers from the Mainland, for instance, getting rid of the 7% per county ceiling in employment-based immigration would have no negative impacts on our immigration policy. Restrictionists might think about this as a concession to the Microsofts of the world in exchange -- hopefully -- for something really useful, like eliminating the visa lottery.

Proposal B: Let’s deal with the variations on this proposal that would grant automatic green cards to alien students securing advanced high-tech degrees in the U.S., and assume for the sake of argument that the new law would not increase the number of such degrees, which it probably would. Scoring is a more difficult operation for Proposal B than for the others.

Let’s label the proposal for the doctorates as B-1. There was, for instance, in the last session, Rep. Jeff Flake’s (R-AZ) H.R. 1791 calling for automatic green cards for aliens with new PhDs in the STEM fields of science, technology, engineering and mathematics.

We know from the 2009 Summary Report: Survey of Earned Doctorates Table 20, that there were 10,838 doctorates awarded by American universities to aliens on temporary visas; let’s round that up to 11,000 a year for our purposes.

Would H.R. 1791, sometimes called the Staple Bill, open the way for 11,000 new green cards each year, over and above the current flow? No, it is not that simple. In the first place such legislation would in all likelihood include the spouses and children of the new PhDs, and we know from data cited below that the ratio of family members to arriving PhDs is about 1.2 family members per PhD, so the 11,000 would become 24,200 (11,000 x 2.2 = 24,200).

But there are two other variables, both pushing in the opposite direction: first, many, many new high-tech alien PhDs stay in the U.S. under current rules so the legislation would not likely increase the size of this population; secondly, some of the new international PhDs would not accept the offer anyway. We have some data on the first of these variables, if not the second.

Fortunately, Michael Finn of Oak Ridge Laboratories, a federal institution, has been following the question of alien PhD retention for years. He has his hands on the lists of each year’s new PhDs, a complete set of their Social Security numbers, and access to data on their earned incomes. It is a remarkably sound data base. In his most recent report he found that in 2010, 67% of foreign PhDs in all disciplines who had graduated in 2005 were still in the U.S. two years later; further, and more pertinently to the Microsofts of the world, five years after graduation Finn found that 72% of the computer science graduates remained in the U.S.

So, for our estimation purposes let’s assume a retention rate of 70%, or a loss rate of 30%. Multiply that loss rate by the 24,200 and we have 7,260, or, say, 7,000 possible additional visas that proposal B-1 would create each year.

Bear in mind, however, that some of these brand-new PhDs would want to return, to say Taiwan or Europe, despite the prospect of a green card. So the proposal would entice some fraction of 7,000, not all of them.

Turning to the master’s plus the doctorates version of the Staple Bill, scoring produces much larger numbers. In addition to the fraction of 7,000 for the new doctorates, and using the same concepts just outlined, but using my estimate of 50,000 new STEM master’s degrees for people with temporary visas annually, we get these numbers:

  • 50,000 x 2.2 (family size) x 30% loss ratio = some fraction of 33,000

So the combination of some fraction of 7,000 and a fraction of 33,000 for this (a broader version of the Staple Bill) would produce some fraction of 40,000 more people each year. (Anyone interested in learning about my detailed estimation technique regarding the 50,000 alien STEM master’s degrees, drawn from the Institute for International Education and the National Science Foundation data on aliens and graduate education, drop me an email: [email protected])

Proposal C: This is the notion that relatives should not count against the 140,000 annual EB ceiling. Were that to apply to the first and second priorities of skilled immigrants, the classes that the open borders people were talking about, that would add about 50,000 more immigrants, and 50,000 more people to the population -- each and every year. (This is based on data from Table 7 of the 2010 Yearbook of Immigration Statistics showing the number of relatives accompanying the principals -- the workers -- in these two priorities that year. About 1.2 relatives were admitted for each worker in these classes.)

In summary, this is how I would score the proposals in terms of adding people to the U.S. population each year:

Proposal A: Eliminate country ceilings = zero

Proposal B-1: Green cards for new U.S. PhDs = some fraction of 7,000

Proposal B-2: Green cards for new PhDs and master’s degrees earned by aliens in the U.S = some fraction of 40,000

Proposal C: Do not count workers’ relatives = 50,000

Somebody, other than this lone researcher, should be doing such calculations on every immigration bill introduced in the Congress.

Footnote: Another question not asked at the Senate hearing was this: If there is such a need for more high-tech alien workers why have only 33,900 of the year’s 85,000 numerically-limited H-1B openings been requested? In some years the 85,000 slots are gobbled up the first day they are available.