Early in 2011, on a plane from New York to New Delhi, I met a disappointed young chemist on his way home after spending several years at a U.S. university. As I recounted in a blog entry, this returnee, whom I called Ashok, had been a postdoc at a mid-ranked institution but had lost his visa when his principal investigator's grant was not renewed. Just like that, Ashok's position ended and he could not find another one.

Talking across an aisle in the jam-packed steerage section of a long-haul flight did not conduce to deep exploration of this Ph.D. scientist's experiences or motivations, as a researcher or job hunter. But before we turned our attention (by unspoken mutual agreement) away from the details of events that evidently embarrassed Ashok and toward the airline's movie offerings, he indicated that he would have preferred to stay in the United States had his job search worked out differently.

For years now, homeward-bound foreign-born scientists have been on the minds of many people, most prominently President Barack Obama and the authors of the "Gang of Eight" immigration reform bill that passed the Senate earlier this year. (Ashok doesn't fit the profile of the scientists to whom the bill would grant blanket permission to stay in this country, because he did not earn his Ph.D. at a U.S. university. Nonetheless, he does resemble them in intelligence, ambition, and advanced technical education).

In a speech several months after my airborne chat with Ashok, President Obama bemoaned, to an applauding audience, the inconsistency of U.S. policies that "provide students from around the world with visas to get engineering and computer science degrees at our top universities. But then our laws discourage them from using those skills to start a business or a new industry here in the United States."

Multiple incentives

Whether scientists stay or go turns out, however, to be more complicated than the president suggests, according to a recent study tracking the decisions of foreign-born, U.S.-educated Ph.D. researchers. Attracting Talent: Location Choices of Foreign-Born PhDs in the US, by economists Jeffrey Grogger of the University of Chicago and Gordon H. Hanson of the University of California, San Diego, looks back over 5 decades of data from the National Science Foundation's Survey of Earned Doctorates (SED) to identify the factors that influence how newly graduated scientists decide where to live and work. As the authors note, SED tabulates every Ph.D. awarded in the United States. Visa status, they find, plays a significant role in outcomes—but for many holders of new doctorates, circumstances other than U.S. immigration policy are extremely influential. 

Historically, new Ph.D. researchers' desire to stay in the United States has been high, Grogger and Hanson find. On average, 87% of those from low-income countries, 70% of those from high-income countries, and 67% of those from middle-income countries told SED that they intended to live in the United States "after graduation," the authors write. Over time, China and India have been the largest low-income contributors to the U.S. supply of newly minted Ph.D. researchers; now, however, both belong to the ranks of countries the World Bank considers middle income.

To dispose of the obvious first, Ph.D. researchers who are neither citizens nor permanent residents generally can stay only if they can find a job that will provide either a visa, such as a temporary H-1B, or the coveted green card. Some others extend their student visa for a year of work "directly related to [their] major area of study," under the Optional Practical Training program, according to the U.S. Citizenship and Immigration Services Web site. 

Many new Ph.D. researchers who want to stay in the United States succeed in finding ways to do so. By the time they received their degrees, more than 80% of SED respondents who had wanted to stay had either found a position or "were in the advanced stages of landing" one, Grogger and Hanson write. More than half had already committed to a job or a postdoc appointment, which the authors count as employment. Less than a fifth were still on the job (or postdoc) market with "no specific prospects."

The likelihood of finding visa-providing work may well correlate with a scientist's overall intellectual chops, the authors note. New Ph.D. researchers with "stronger academic ability," as indicated by having received university support in the form of a fellowship or assistantship or by having at least one parent with a college degree (which, especially in poor or middle-income countries, places the family among an educated class that is much smaller than in a rich country) are likelier to stay than those who lack these presumed correlates of academic strength. 

But an even more powerful predictor of staying in the United States, the authors write, is whether "in recent years the US economy has had strong GDP [gross domestic product] growth or the birth country of the foreign student has had weak GDP growth. Foreign students are less likely to remain in the US if they are from countries with higher average income levels." A recent move of the native country toward democracy also draws scientists home, but even so, Grogger and Hanson conclude, "Economic conditions are perhaps the most important factor shaping the location choices of recent PhD recipients in [science and engineering] fields." Furthermore, "business-cycle considerations aside, we do see that as countries develop they become more attractive locations for PhDs in science and engineering."

Unlinking visas and jobs

Should the Senate-passed Gang of Eight immigration reform bill, with its provision to "staple a green card" to every science, technology, engineering, and mathematics (STEM) graduate degree granted to a foreign national, become law, new Ph.D. researchers will no longer have to find work in order to stay. But, as we have also previously noted, "programs that do not link admissions to the promise of a job end up with too many unemployed or under-placed skilled immigrants," according to University of Texas, Austin, political scientist Gary Freeman. "[E]xperiments with broad efforts to recruit skills without regard for domestic demand, as with Canada, have on the whole not been satisfactory."

Canada originated the system of admitting immigrants based on skill levels rather than on employment promises, but it will drop this approach in 2014 because of what Maclean's magazine calls the "ugly reality facing so many" of the skilled immigrants in the country: great difficulty finding work and low earnings when they do. In place of Canada's skills-based "points" system (which a number of U.S. politicians admire), Canada will use something closer—though not identical—to the current American system of employment-based visas. Under the new Expression of Interest program, modeled on plans used in New Zealand and Australia, skilled would-be immigrants will apply for inclusion in a pool of approved employment candidates. Employers will then select from among those on the list, whereupon permission to work in the country will be granted.

Creating opportunities

The Great Recession year when Ashok—and a number of other young foreign-born Ph.D. researchers—left the United States was brutal for chemists, as we noted at the time. The American Chemical Society reported record joblessness among its members.

Would Ashok have been better off staying, without any apparent prospects of finding work that used his highly developed abilities? Would he have been likely, as President Obama's oration implied, to join an iconic group of immigrant entrepreneurs and job creators? Or would he, without the chance to work in his scientific field, have resembled a different iconic figure, one less touted by politicians but familiar in many cities: the ultra-educated, foreign-born cabbie? There's no way of knowing, but the Canadian experience seems to argue for the latter. And considering that, the year Ashok returned home, India's GDP grew more than three times as fast as the U.S. GDP did, it is not unlikely that the opportunities available to him were better there than here. 

The United States did, of course, lose Ashok's abilities, but this country didn't seem able to make good use of them when it had the chance. If policymakers and politicians want more scientists to stay in this country and contribute to our economy, they ought to think about how to provide work opportunities in their fields that would make that possible. Come to think of it, it wouldn't hurt if policymakers found ways to do the same for this country's own un- and underemployed home-grown science talent.

Beryl Lieff Benderly writes from Washington, D.C.

10.1126/science.caredit.a1300190