
Is immigrant teachers such as Flora Gaskin helping school districts bring down their labor costs? Or do they actually fulfill the need for highly qualified teachers this nation needs. (Photo courtesy of the Washington Post)
Norm Matloff — he of the eponymous Immigration Forum – didn’t take too well to my latest report in The American Spectator on the teacher quality reasons for expanding the H-1B visa program and sensibly reforming immigration overall. As far as he was concerned, the piece was lacking, largely because it allegedly failed to admit evidence that somehow the skilled visa program was merely a tool for school districts and tech firms “to save on labor costs.”
The problem with Matloff’s assertion is that ultimately, he didn’t bring the evidence. Especially on the education front, an area I have covered for the past four years. As yours truly will make clear on Free Trade Nation, anyone can say anything about any subject. What matters most is the facts of the case. One may disagree with conclusions derived from facts, but the facts speak for itself.
When a study doesn’t say what you say it says: Let’s start with one of the two “Congress-commissioned reports” Matloff mentions, a Government Accountability Office study, released five years ago during another round of battling over expanding the program. At best, the GAO report simply concluded that more study was needed. The report couldn’t conclude whether H-1B visas had a negative impact on salaries, as opponents of H-1B proclaim. As GAO concluded: “Much of the information policymakers need to effectively oversee the H-1B program is not available because of limitations of [the Department of Homeland Security's] current tracking systems.”
The report itself has problems, largely because GAO couldn’t get a hold of most of the 135 employers it attempted to contact for its research. The employers they did manage to speak to said that they used H-1B to fill engineering positions. As a result, GAO admitted its results on this front may have been affected by a self-selection bias inherent in any such survey.
Certainly the GAO report did note that many H-1B hires were made by firms outside of the tech sector. But, as noted by the report, tech-related positions accounted for 65 percent of all approved H-1B petitions in 2000 and 40 percent of hires in 2002, as the nation recovered from recessions in the tech industry (and the overall economy). Essentially the evidence shows that tech firms do have a need for high-skilled labor that cannot always be easily filled by native-born Americans. And the fact that tech firms aren’t the only users of H-1B visas merely proves that there are numerous beneficiaries from the program itself.
As for reductions in science- and engineering-related positions mentioned in the report at the time? One must also remember that the recession of the early 2000s adversely affected the tech sector the most; after all, it was the epicenter for growth throughout the late 1990s as companies such as Cisco Systems and dotcom outfits were hiring. It also shows another reality that has been the target of those opposed to lowering trade barriers and globalization: The outsourcing of tech jobs during this decade as firms such as India’s Infosys began handling such operations from their offices in Asia. Companies who cannot hire skilled workers they need will find a way to do so — even if it means moving some of their operations offshore.
And a study that hardly merits attention: Then there is the Urban Institute report, also cited by Matloff as a refutation of the arguments made by myself and others favoring H-1B expansion. Let’s just say that some of the data used by Urban in the report is, well, not worth the paper upon which it is printed doesn’t stand up to scrutiny when closely examined; its overall argument, therefore, begins to fall apart.
[By the way: This is a general problem with data cited by opponents of immigration; the information itself may be ostensibly credible, but doesn't hold up when more closely scrutinized and then compared with other data that can stand up to scrutiny. You can cite anything you want, but it better stand up to the harsh light that is criticism.]
For example, the report proclaims that more 18-24 year-olds have graduated with a high school diploma then during the 1970s, based on the infamous Current Population Survey conducted by the U.S. Census Bureau. Infamous? Yes, infamous, because the survey has numerous holes. For one, institutional populations — the military on one side (which has a large number of GED recipients and thus, a large population of high school dropouts) and prisons (another haven for dropouts and GED recipients) — are not counted in the survey.
As Nobel laureate James Heckman points out, CPS also has a low coverage of population problem; that is, it leaves out large numbers of population groups. For example, it leaves a large portion of the young black and latino male population — groups whose graduation rates are, in general, lower than the overall population. As a result, notes Heckman: “As a result, the overall graduation rate based on the CPS data is
14 nearly 2 percentage points higher than a Census-based estimate” the same group. Essentially, CPS leaves out a large pool of dropouts, thus skewing the percentage of Americans leaving the K-12 education system with a sheepskin; therefore it also skews how many people attend and complete college.
This is why the leading scholars on graduation rates and high school completion — Jay Greene of the Manhattan Institute, Chris Swanson of Education Week (a former Urban Institute scholar), Robert Balfanz of Johns Hopkins University and Michael Holzman of the Schott Foundation for Public Education — never use CPS in their research. They look at actual scholastic enrollment, which is more accurate because schools are actually paid based on the number of children in school.
And based on that data, no one is truly sure what the dropout rate really was in 1972. At best, we know that graduation rates have been in decline since the 1980s. And if one wants to use available data, as Heckman has done by using decennial Census data in his studies of high school completion ( I wouldn’t recommend it), it has declined from 80 percent to 75 percent over the past five decades. As a result, college completion rates have remained essentially flat.
[By the way, Heckman isn't exactly a fan of the Greene/Swanson/Balfanz/Holzman camp; he argues that both the Greene camp and those using CPS data (which also includes noted graduation rate skeptic Lawrence Mishel, who also uses the even less useful NREL data) overstate their case. Having covered the nation's dropout crisis for quite a while, I would disagree. But the fact that his own study essentially counters the argument made by Urban speaks for itself.]
CPS is also not reliable because of the self-reporting bias inherent in it. Those being surveyed must be willing to own up to being dropouts or graduates. Since those being surveyed may fudge that bit of data, the surveys are of questionable accuracy on that front.
This self-reporting bias becomes a major factor depending on what part of the nation in which one lives. There is less of a stigma against dropping out in Indiana — a state long-dependent on traditional manufacturing and less-focused on education — than in states with traditions of college completion such as a Wisconsin. So those in such states may have more incentive to conceal their lack of a sheepskin.
Meanwhile the other assertion made by Urban — that students are doing better on math and science tests — hardly stands up. The very chart it uses in its own report — of data from the National Assessment for Educational Progress — speaks for itself. Average scale scores for 17-year olds scoring in the 90th percentile — essentially students in the “advanced” category — have remained flat between 1978 and 2003; average scale scores for those scoring in the 25th percentile increased by just ten points during that same period and have been flat for other percentiles. Average scale scores for 13 year olds have increased by only 6 percent during that same period.
The best test for comparison, however, may be between American students and those of other countries, whose academic performance has improved steadily over time to the point that in most tests, America’s best students trail its rivals. On the 2006 Programme for International Student Assessment, for example, ten countries — including Chinese Taipei, Korea, Hong Kong, Switzerland, Finland, Belgium, the Czech Republic and Liechtenstein — have high-performing students with better math test scores than our country’s best students.
Meanwhile Matloff simply suggests that teacher positions — especially in math and science — would be filled if salaries for those posiions were increased. On this front, I would have to partly agree. One of the biggest challenges in education lies in the archaic and inefficient system of teacher compensation, in which teachers are generally paid based on seniority and the number of graduate course work they accumulate over time. Such a system not only fails to reward teachers for improving academic performance, but doesn’t even allow for pay to be differentiated based on scarcity. As a result, science and math collegians pursue positions in better-paying fields.
The system doesn’t save money for school districts at all; they still must pay the teachers the salaries given to native-born colleagues along with the 10 days of sick leave that can be used at any time. If anything, the system benefits, borne during the emergence of the NEA and AFT in the 1960s – when teacher performance couldn’t be measured at all (versus not so easily measured today) — benefits rank-and-file teachers and unions at the expense of school districts (which benefit from the labor peace the system brings, but lose out on the ability to improve academic performance in a meaningful way), parents and ultimately, students. This is why the system still exists.
Enacting teacher pay differentials rewarding those teaching math and science would help alleviate the shortages. Allowing mid-career professionals to enter teaching through alternative certification programs would also help. But it may not fully relieve those shortages. After all, one assumes that there is a large pool of math and science students who want to teach; given the high attrition among teachers with less than two-to-three years experience (especially in light of the fact that they can usually achieve tenure — and near-lifetime job security) within that time, there may never be enough of such skilled native-born workers willing to go into — or stay in — the profession.
Teaching isn’t for everyone; you must not only have a wide and deep range of subject-matter competency and strong instructional method, you must also care for children that aren’t your own and be patient with them in getting them up to speed on a subject. Not everyone can hack it.
Ultimately, in any case, the evidence speaks for itself.
Which comes back to my original argument: Expanding the H-1B visa pool would help expand the pool of high-quality teachers — especially in those scarcity positions needed in America’s woeful public schools. Just as importantly, the marketplace for talent is no longer just a nation-based affair; talented workers need freedom of movement in order to exploit their “means of production” for their own gain. Organizations need the ability to seek out the best talent available to meet their needs.
This means a full overhaul of an immigration system that is Byzantine, antiquated, originated from mistaken (and sometimes, bigoted) notions of what being American means and built on the fears of a few instead of the realities of the modern world.
Not to say that H-1B itself is perfect: The rules, crafted in response to opponents of immigration who don’t want any new citizens here of any kind, is too much like a guest worker system; the rules should give high-skilled workers the ability to decide whether they want to become full-fledged Americans as previous emigres have done or not, as well as transfer freely from one employer to another. Such flexibility, however, is unlikely given the opposition — from the rather calm views expressed by Matloff to the demagoguery of the Center for Immigration Studies — to any pro-free market version of immigration reform.
But the current system doesn’t work; the overall immigration system is a mess and needs reform. Of the right kind.