Location Choices of Foreign-Born Ph.D.s
Almost 80 percent of foreign-born Ph.D.s who earned degrees between 1960 and 2008 reported intending to stay.
The share of U.S. science and engineering Ph.D.s earned by foreign-born students rose from 23 percent in 1970 to 56 percent in 2007. In Attracting Talent: Location Choices of Foreign-Born Ph.D.s in the US (NBER Working Paper No. 18780), Jeffrey Grogger and Gordon Hanson note that science-and-engineering graduates have "relatively high propensities to produce and commercialize patents" and to launch high-technology business ventures. Thus, where they choose to locate after they have completed their education is "likely to affect the global distribution of innovative capacity."
The National Science Foundation Survey of Earned Doctorates contains information on the characteristics of all individuals receiving a Ph.D. from a U.S. university since 1958 and includes a question on whether the graduate intends to stay in the United States. Previous work has shown that the reported intent to stay closely tracks actual rates of stay as calculated from Social Security Administration earnings reports. Almost 80 percent of foreign-born Ph.D.s who earned degrees between 1960 and 2008 reported intending to stay.
The authors find that the most important factor affecting post-graduation intent to stay is economic performance. An increase of 1.2 percent in lagged U.S. GDP raises the intent-to-stay rate by 3.3 percent. An increase of 4.3 percent in a graduate's birth-country GDP is associated with a 2 percent decline in the intent-to-stay-rate: as economic development accelerates in birth-countries, graduates become more likely to return home. Graduates are also more likely to return home if the government in their birth country becomes more democratic.
Graduates with stronger academic ability, measured by whether they received university support for a research or a teaching assistantship, had a 6.8 percent higher stay rate. Assistantships were the primary means of support for 52 percent of the sample; 11 percent of the sample was supported by a university fellowship or scholarship, which was associated with a 2.7 percent higher stay rate. Students supported by foreign funding, 4 percent of the sample, had an intent to stay that was 26.7 points lower.
As one would expect, ties to the United States are also important. Foreign-born U.S. citizens have a 19.5 percent higher stay rate; green card holders have a 15.3 percent higher stay rate; and U.S. college graduates have a 3.6 percent higher stay rate.
To more closely examine how economic growth might affect stay rates, the authors analyze the characteristics of the 9.3 percent of U.S.-born Ph.D. graduates who have reported planning to work for a business since 1984. The shares were highest in plastics engineering (50.5 percent), ceramics sciences engineering (44.6 percent), metallurgical engineering (41.3 percent), polymer chemistry (42.3 percent), forestry science (40.0 percent), and food science (33.8 percent). The effect of lagged GDP growth on the foreign-born stay rates in these fields was one-third larger than in other fields. Of the non-science fields, only economics and management had intent-to-stay rates that were similarly sensitive to the business cycle.
China, India, Korea, and Taiwan accounted for 60.7 percent of all U.S. science and engineering Ph.D.s awarded to non-U.S. students in 2007. Over the period, new U.S. Ph.D. rates per 10,000 birth-country inhabitants averaged 34 for the United States, 30 for Taiwan, 13 for Korea, 3 for China, 2 in non-U.S. high income countries, 1.7 for India, 1.5 for middle-income countries, and 0.6 for inhabitants of other low-income countries. Almost all of the variation in the stay rate over time for India was explained by lagged U.S. and home country GDP growth. Stay rates for graduates from Korea and Taiwan were increased by the 1997 Asian financial crisis. Stay rates for graduates from Korea were reduced by that country's transition from military rule to democracy in 1988.
--Linda GormanThe Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.