The Impact of Foreign Students on the Earnings of Doctorates
Immigration increased the supply of doctorates in computer science and mechanical engineering by 36 percent, causing wages to fall by an estimated 10 percent.
The textbook model holds that immigration will depress the wages of comparably skilled native-born workers, but empirical studies of wage changes in U.S. metropolitan areas with large numbers of immigrants generally have found that immigration has only a small effect on native-born wages. Those empirical studies have had two main drawbacks, though. First, if immigrants cluster in cities with thriving economies, then the results may reflect a correlation between immigration and local economic growth. Second, if natives respond to immigration by moving to another city, then the effects of immigration would be spread throughout the U.S. economy. This increases the probability that researchers looking at a single city would conclude that immigration has no effect on wages.
In Immigration in High-Skill Labor Markets: The Impact of Foreign Students on the Earnings of Doctorates (NBER Working Paper No. 12085), NBER Research Associate George Borjas analyzes the effect of the large number of foreign students who earned doctorates in science and engineering between 1968 and 2000 on the wages of PhDs in those fields. He uses data from the Survey of Earned Doctorates and the Survey of Doctoral Recipients. In the 1970s, Borjas notes, nearly 20 percent of those earning science and engineering doctorates were foreign-born students who intended to stay in the United States. By the 1990s, that number was over 33 percent.
Foreign doctoral students are not equally distributed across the various science-and-engineering fields, however. In the 1990s, more than half of the doctorates awarded in civil engineering were earned by foreign-born students intending to say in the United States, while in psychology, the fraction was only 5 percent; in health and related sciences it was almost 17 percent; in biological sciences it was over 27 percent; and, in electrical and mechanical engineering it was roughly 49 percent.
As the number of people with doctoral degrees has increased, it has become more common for PhDs in some fields to take a post-doctoral appointment (a "post-doc") as their first job rather than to proceed to a more secure, and higher paying, permanent appointment. The number of young PhDs in post-doctoral positions is nearly 29 percent in the biological sciences, over 17 percent in physics, and almost 10 percent in chemistry. Yet, on average, a post-doc under the age of 40 earns $36,000 a year while the average salary for a regular appointment is approximately $65,900 a year. According to the 2000 Census, U.S. men with undergraduate degrees who were between the ages of 25 and 29 earned an average of $33,000; those who were 30-34 years old earned $42,300. In short, salaries in post-doctoral appointments do not reward the additional years of education required for a Ph.D.
After adjusting for the possibility that the demand for people to fill post-docs is driven by extraneous factors, like increased funding, Borjas concludes that an immigration-induced 10 percent increase in the supply of people with doctorates raises the probability of being employed in a post-doctoral position by about 4 percent. The effect on younger native-born doctoral graduates is even larger. For them, a 10 percent immigration-induced increase in supply increases the probability of employment as a post-doctoral by nearly 22 percent.
Immigration also appears to affect the wages of those who do not take post-docs. Between 1993 and 2001, immigration increased the supply of doctorates in all fields by an average of almost 14 percent. The wage of the average worker with a doctorate in science or engineering fell by close to 4 percent. In some fields, the effect was much larger. Immigration increased the supply of doctorates in computer science and mechanical engineering by 36 percent, causing wages to fall by an estimated 10 percent. In all, the influx of foreign students reduced the wage growth of science and engineering PhDs by about 40 percent.
Borjas notes that low returns to earned doctorates may cause bright undergraduates born in the United States to pursue professional occupations in which wages have not been depressed by immigration. As a result, research labs employing large numbers of post-docs will find that "natives do not want to do the type of work that immigrants do," at least not at the wages they offer. This encourages the labs to press for more recruiting from abroad, with the result that wages for PhDs will continue to be depressed. Borjas cautions that it would be a mistake to assume that lower wages for PhDs is necessarily bad policy. Although immigration may reduce the economic returns to earning a doctorate in the United States, having more skilled labor in the population also may lead to more, and more rapid, scientific discovery. If that is the case, then the overall benefit from high-skill immigration could be very large.
-- Linda Gorman