People who don't believe that unauthorized immigrants in the US should be given legal status tend to emphasize that it would be unfair to immigrants who are in the US legally — because they deserve a reward for settling in the US "the right way."
But what if the legal immigration process is itself unfair?
A new study by researchers at MIT and Brown University suggest that might be what's going on — unintentionally. They looked at applications for employment-based green cards among immigrants who were already in the US. What they found was that, in the standard approval process, Latin American immigrants were much less likely than average to get approved — and Asian immigrants were more likely. But when the government went through a slower but more complete approval process, the disparities disappeared.
From engineers to restaurant cooks, Latin Americans are at a disadvantage
For this study, the researchers (Emilio J. Castilla of MIT, and Ben A. Rissing of Brown) evaluated the phase in an application for an employment-based green card where the US Department of Labor has to approve or deny the immigrant's "labor certification" for a particular job. These applications get filed when a company decides to take an immigrant who's (almost always) already in the country on a temporary visa — like a work visa or student visa — and sponsor him or her for a green card, which would let him or her stay in the US permanently and eventually apply for citizenship. So this is the phase in the process where temporary immigrants can get approved to become permanent ones.
Technically, this is supposed to be evaluating the immigrant's would-be employer, not the immigrant him- or herself. In order to get the immigrant "labor certified" for a green card — or any work visa — the would-be employer has to prove that they tried to find a US citizen to fill the job, but failed.
One thing that isn't supposed to be a factor in the application is the immigrant's country of origin. But even when the researchers controlled for as many variables as they could — from the temporary visa that an immigrant held in the US when he filed the green-card application, to the skill level of the job — they found that approval rates varied widely from one nationality to the next. 90.5 percent of Asian immigrants were approved for labor certification. But only 66.8 percent of Latin American immigrants were.
The regional disparity even showed up in immigrants applying for the same type of job. "Immigrants from Asia seeking employment as restaurant cooks are 41.6 percent more likely to be approved than immigrants from Latin America, all else being equal," the researchers write. It's a problem for high-skilled workers, too: Asian immigrants weren't any more likely than Canadians (for example) to get approved to work as computer software engineers, but Latin American immigrants were over 25 percent less likely.
Is this just because Latin American applicants are less educated?
The biggest problem with the study: the government agents looking over immigrants' applications could see each immigrant's educational background, but the researchers couldn't. That could be a huge factor explaining the variation: maybe Latin American immigrants are simply less educationally qualified for the positions they're applying for.
But the study indicates that can't be the whole story. For example, the researchers looked at immigrants who were already on H1-B high-skilled visas (99 percent of whom have a bachelor's degree or higher) but were applying to upgrade to green cards. Among that group, Asian immigrants were still 11 percent more likely than Canadians to get approved for green cards — and H1B-holding Latin American immigrants were 20 percent less likely.
There's also previous research showing that government officials profile immigrants based on their countries of origin. In one study, in which officials were asked to look over fictional visa applications, the author said that region of origin was being used strongly as a "criterion of a visa applicant's desirability."
Could more complete application evaluation solve the problem?
Most of the time, the DOL makes decisions based on basic information about the immigrant, the job, and the employer, as well as evidence of the employer's failed attempt to recruit US citizens. 90 percent of those cases get approved, and they're typically approved or denied in under 4 months. But in a few cases — thanks to a process that is partly random, and partly not — applications are "audited," and agents take a more in-depth look at an immigrant's background, and the requirements for the position. In those cases, only 57 percent of applications are approved — and it takes about 2 years to come to a decision.
According to the new study, audited applications had one big advantage over non-audited ones: the disparities in approval rates between immigrants from different regions disappeared. That might indicate that whatever is happening to favor Asian immigrants and disfavor Latin American ones in the quicker process is unintentional, since government officials don't appear to think there's good reason to be more suspicious of Latin American applicants.
The study's authors suggest an easy fix: making it impossible for an official to see an applicant's country of origin, just like officials aren't allowed to see applicants' ages or sexes. If this really is a significant problem, however, it's not something that changing the process in the future will be able to fix.
If the government is really making it harder for Latin American immigrants currently in the US legally to get green cards through their employers, they're unfairly forcing Latin American immigrants to make the difficult choice between leaving the country they've been living in for years, and staying in the US after their visas expire. In other words, the government's approval-rate problem might be driving more legal Latin American immigrants to become illegal.