Yikes! Two of My Favorite Economists Go Against Each Other

One of the classic academic studies of the effects of immigration on low-skilled Americans is that of David Card, concerning the Mariel Boatlift. Card found that the sudden Cuban influx had little or no adverse work impact on native African-Americans. Lately George Borjas, one of my favorite economists, has disputed Card’s findings, only to find my UC Davis colleague Giovanni Peri disagreeing. Needless to say, Borjas has been responding to Peri. And today I was taken aback to find that another of my favorite economists, Jennifer Hunt, also disagrees with Borjas.

I have been quite open in pointing out that Peri, while very sharp and by the way very personable, generally does not cite work by those taking views opposite his — a grievous academic transgression — and is prone to contradicting his own work, as I have explained before and is also seen in the first link above. Card is highly prominent, in part because of his pro-immigration, pro-labor work, but I have seen serious indications of his bias as well.

But Borjas and Hunt are the real deal. George may have a viewpoint, but I have never seen anything other than fully professional behavior on his part. To my knowledge, no one has ever accused Jennifer of bias either.

I have not read the papers on the boatlift by any of the above, and I do not intend to, because I believe that this is a question — like many — that can’t be settled by numbers alone.

As a statistician — and all of those papers are purely statistical, nothing really drawing upon economic theory — I have always stressed to my students and my consulting clients the importance of complementing the quantitative with the qualitative. Do the numbers make sense in light of qualitative knowledge, and if not, can the discrepancy be reconciled?

What qualitative knowledge can we draw upon here? There are well-documented examples of immigrant Latinos be hired in lieu of — indeed, replacing — native blacks, such as the case of black LA janitors. There is the infamous example of Hurricane Katrina, in which African-Americans had been hired to clean up and rebuild, only to be told soon afterward that they were fired, because “The Mexicans have arrived.”

So, sorry to say this Jennifer and Giovanni, but I think it’s obvious that the arrival of the Marielitos did adversely impact local black workers. There remains the question of degree and extent of impact, and though some have described the boatlift as a “natural economic experiment,” ideal for settling the question of whether immigration harms blacks, I continue to maintain it just is not possible using just data.

But for the purpose of assessing White House adviser Stephen Miller’s claims in support of the new RAISE bill (second link above), I have to side with him. Mind you, I don’t support the bill — I don’t like points systems, as I would like to see a mix of immigrants socioeconomically, racially and so on — but there is far more to this issue than “He said, she said” academic papers.

Followup on the English Issue

Judging from some comments I’ve been receiving, I see that many still don’t get my point about the impact of requiring legal immigrants to know English. Actually, I suspect that even Stephen Miller doesn’t understand this.

To make the point most clearly, suppose that Congress adds the English requirement but makes no other changes (i.e. no point system etc.). My point is that the same people would come here, with and without an English requirement. Those of you who are thinking that this means the main people who immigrate are the college educated have it wrong. The blue collar, lesser-educated people are highly motivated to immigrate and thus WILL develop the necessary level of proficiency in order to attain a green card.

As I mentioned, long ago I was an ESL instructor in San Francisco’s Chinatown. (The story of how I got there is interesting but a tangent here.) I taught mainly ESL 50, the lowest level. Actually, the lowest had been ESL 100, but so many immigrated knowing next to nothing that the city instituted an even lower course. Most were not high school graduates, and basically knew the ABCs; some didn’t even have that. But they all learned, and by the end of the semester most were fit to work in a low-level service job in which English is spoken, and I would set this level as the requirement. Had English been a requirement for their green cards, they would have picked up their English back home before they immigrated.

By the way, refugee status is not the same as a green card. Refugees may arrive under circumstances that do not give them time to learn English before coming here. But under current law, they can apply for green cards after a year in the U.S., and again that would give them time to come up to speed in English.

Miller vs. Acosta

I seldom call for a journalist to be fired, but the behavior of Jim Acosta of CNN in today’s briefing by White House adviser Stephen Miller was disgraceful and inexcusable. Acosta was acting like an opposition party politician, not a professional journalist.

But my main point in this blog post is to explain why Acosta was wrong. First, though he may be correct in saying that a points-based immigration policy would be a radical change for the U.S., it is the approach taken by Canada and Australia, whose systems I’ve never heard criticized by CNN/Acosta. Indeed, a points-based policy was the basis of the bipartisan 2007 Comprehensive Immigration Reform bill. And to my recollection, no one at CNN opposed that approach, including Acosta. In other words, CNN/Acosta’s view of a bill depends on who supports it, not its content.

I personally have never been a fan of points-based systems, and have opposed the RAISE Act here in this blog. But Acosta is being hypocritical and, again, unprofessional.

The point I most wish to bring up here, though, is the objection Acosta had to requiring immigrants to know English as a precondition for being granted a green card. I have been advocating that for 20 years, and I am speaking on more than just armchair punditry, as I am a former ESL teacher in San Francisco’s Chinatown (I usually teach the sub-beginner course), and am someone who has been immersed in Chinese-immigrant communities for 30 years.

Let me state the obvious: English is a lot harder for Chinese people to learn than for, say, Spanish-speaking people to learn. And yet the Chinese do learn, because they know their success in the U.S. depends on it.

Moreover, if it were required that potential Chinese immigrants learn English back home before immigrating, they would learn English back home before immigrating. They would do this because of their intense desire to immigrate. There are lots of learning tools on the Web, and I guarantee you that English tutorial companies would spring up in China like mushrooms after a rain. They would not be of high quality, but they would be good enough. Even as it is, if you go into a bookstore in any Chinese city, you will see many books, tapes and so on for learning English.

THIS IS A NON-PROBLEM, FOLKS. Having this requirement would help the U.S. and help the immigrants, a win-win situation if there ever was one.

And yes, if I wasn’t clear, Acosta should be fired.

H-1B: It is NOT a Matter of Degree

As any reader should know, I consider the biggest obstacle — indeed a danger — to genuine reform of the H-1B work visa program (other than lack of political will) is what I call the Intels Good, Infosyses Bad (IGIB) myth. The presumption there is that the Indian firms such as Infosys who hire H-1Bs and “rent” them to U.S. employers are the main abusers of the visa program, while those who hire H-1Bs directly, say Intel, use the program responsibly. Actually, any legislation based on IGIB will be based on a false premise, and will likely make things worse, not better.

The latest attempt to justify IGIB is an article in today’s Quartz, which shows that the (a) Intels tend to pay more than do the Infosyses and (b) the Intel H-1Bs tend to have master’s degrees while the Infosys ones have only a bachelor’s. Both (a) and (b) are true but highly misleading, as I will explain later in this post. (The data source is USCIS.)

I have explained why IGIB is a myth many times, such as in my Huffington Post op-ed:

A popular tack taken by industry lobbyists and their congressional allies is to blame Indian firms that hire H-1Bs and “rent” them out to mainstream companies. This is pure scapegoating and a veiled appeal to xenophobia. It’s an attempt by Silicon Valley firms to distract attention from their own abuses of the system. The data show that the Silicon Valley firms do indeed underpay their H-1Bs, and individual examples of abuse by household-name firms are disturbing, to say the least.

Cisco, for instance, was exposed as routing job applications from American engineers to an immigration law office, rather than to engineering managers, apparently to gather evidence that no qualified Americans were available for the job taken by a foreign worker. The immigration lawyer was deciding who was “qualified.” A former manager at Oracle accused the firm of justifying underpayment of an H-1B by saying, “It’s good money for an Indian.”

Promoters of IGIB make a point of the fact that the Intels tend to sponsor their foreign workers for green cards. This again is correct, but it actually is the reason that the Intels’ abuse of H-1B is even worse than that of the Infosyses. Green card sponsorship gives the Intels a huge incentive to give foreign workers preference over Americans in the hiring process. Again, from my HuffPo piece:

…sponsorship renders the workers de facto indentured servants…

This immobility is of huge value to many employers, as it means that a foreign worker can’t leave them in the lurch in the midst of an urgent project. In a 2012 meeting between Google and several researchers, including myself, the firm explained the advantage of hiring foreign workers: the company can’t prevent the departure of Americans, but the foreign workers are stuck. David Swaim, an immigration lawyer who designed Texas Instruments’ immigration policy and is now in private practice, overtly urges employers to hire foreign students instead of Americans.

Now returning to the Quartz article, its point that the Intels pay more than the Infosyses is accurate but obfuscates the issue, by misleadingly giving the impression that the Intels are not underpaying their foreign workers. The problem with that thinking is that the Intels are hiring a higher class of worker. As Quartz notes, the Intels’ H-1Bs tend to have a master’s degree, but it is not just that; it is an American master’s, not an Indian bachelor’s from a three-year program, a huge difference. Thus the Intels do indeed pay more, because they are hiring better-quality people. But they are still underpaying for the level of workers they are hiring. So the Intels/Infosyses pay disparity does NOT imply that the Intels are acting responsibly. As I’ve written before, it’s like buying a car: If you have a chance to buy a car at a 20% discount, it doesn’t matter whether you are buying a Corolla or a Camry; you’re getting a bargain either way.

There is also the vital point that both the Intels and the Infosyses hire mainly YOUNG foreign workers, to the major detriment of Americans over age 35.

You might ask, “Isn’t it better to hire people with master’s degrees? Doesn’t that justify IGIB?” This is the implicit message that the Intels give Congress and the press. Let’s take a close look at this.

One can immediately see big problems with that claim. First, if master’s degrees were so important, why do the Intels routinely reject qualified Americans at the master’s level, without even a phone call? Why don’t the Intels subsidize part-time master’s work for their employees, a perk that was standard in the past? The industry’s actions simply don’t live up to their words here.

And what is so magical about a master’s degree anyway? There is no standard MSCS curriculum, and a strong bachelor’s program will always beat a mediocre master’s. Indeed, in my EPI paper, I showed that Americans with a bachelor’s degree are on average stronger than foreign workers with a U.S. master’s. And of course, Bill Gates, Steve Jobs, Mark Zuckerberg, Larry Ellison and so on never even finished their bachelor’s work.

American students know that a bachelor’s is enough, and many see no point to going further. On top of that, they are typically burdened with student loans to repay. So why forego two years of industry-level salary to pursue a master’s, even with a graduate assistantship?

Texas Instruments (see point about Swaim above) is one of the leading firms lobbying Congress for more H-1Bs. In my report on TI’s 2011 testimony to Congress, I wrote that TI’s remarks put the lie to the industry claim that not enough Americans study engineering:

Texas Instruments V.P. for HR Darla Whitaker…essentially admitted that all that “Johnnie Can’t Do Math” stuff was just slick PR…Ms. Whitaker stated that TI has plenty of engineering applicants with Bachelor’s degrees, and thus does not hire foreign workers at that level. She stated TI does hire H-1Bs, and sponsors them for green cards, at the Master’s and PhD levels, where she says there is a shortage. This naturally led one of the congresspeople on the committee to ask Whitaker, why don’t the American engineering students go on to grad school? She replied that she supposed that the American students were anxious to get out and start making money.

TI is one of the firms that, in the pre-H-1B days, used to subsidize their American engineers for master’s study.

Senator Grassley once said that “No one should be fooled” by the slick industry PR propaganda about H-1B. As shown above, slick as it is, the propaganda can be seen through with a little thought. Yet sadly, Grassley too now supports IGIB, and so do even some major H-1B critics.

The latter say that solving part of the H-1B problem, by clamping down on the Infosyses, is better than no remedy at all. But the bottom line has to be, Will any given legislation help or hurt Americans in STEM?

The fact is that the bills based on IGIB would make things worse for Americans. Many of those bills would raise the H-1B cap (on the grounds that it eliminates the abusers). Many others would grant automatic green cards for the Intels’ hired foreign students; this would be a major catastrophe, exacerbating the age discrimination issue (students are young), and adding tens of thousands of STEM workers per year as permanent fixtures to the STEM labor market, to the general detriment of the U.S. citizens and permanent residents who are already here.

Significantly, no bill features the one type of foreign worker expansion in the interests of the nation: Facilitate the immigration of the true “best and brightest.”

“No one should be fooled,” indeed.