Douglas A. Irwin: This paper focuses on an important and interesting issue: the flow of students from developing countries to the developed world. As the author notes, many highly educated workers in developing countries received their training in OECD countries. In addition, student migration is one of the largest and most unregulated categories of immigrant flows into the United States and other Western countries. Thus, there are a host of questions that can be raised about this migration and its impact.
The paper sets up a horse race between two competing models of student migration: the school-constrained model and the migration model. The school-constrained model suggests that foreign students come from countries with high returns to education but with few domestic opportunities to invest in human capital. In this case, students seek training in the United States and elsewhere with the ultimate goal of returning to their home country and reaping the rewards of the high return to education. The migration model suggests that students will acquire schooling abroad as means of entering and staying in the foreign country when the returns to education are low in their home country. In this case, students are simply escaping the low wages at home in search of higher incomes.
The main difference between the two models is the relationship between student migration and the domestic returns to education. If the returns to education increase in the student's home country, more students will seek education abroad in the school-constrained model, but fewer students will seek education abroad in the migration model.
The two models also have different implications for educational policy in the home country. In the school-constrained model, an increase in the quantity and quality of home education institutions will reduce the number of students who seek education abroad. In the migration model, an increase in schooling [End Page 87] opportunities will complement outward student flows, and may increase student outflows.
Looking at U.S. data, Rosenzweig finds strong evidence that students come to the United States and stay when the return to education is low in their home country. In other words, according to the author, there is overwhelming evidence for the migration model. Part of the reason, the author suggests, is that student migrants are motivated not by the return to education in their home country, but the huge gaps in wage levels (regardless of skill level) between OECD and developing countries. In contrast, even if the local returns to education are significant, the gross wage differentials between developed and developing countries are simply enormous.
The author should be commended for setting up a simple framework in which to think about and evaluate the question of student migration. But, at least to this outsider to the field of labor economics, there are some reasons for skepticism about drawing strong conclusions from the results.
First, the paper notes that the apparent stay rate of students (that is, the proportion of students who stay in the United States and do not return to their home country) is low, at about 10 percent, although the author suggests that this is an underestimate because of imperfect measures of visa adjustments, among other reasons. That is close to what was found by Bratsberg who focused exclusively on the stay rate of students and found it to be about 12 percent in the early 1970s.1 He argued that this number was also improperly measured and was much too low, perhaps being just a third of the true figure. Granting the many difficulties of estimating the true stay rate, and even granting that the stay rate may be as high as 30 percent (something one would certainly like to know with greater certainty), the question remains: If students are largely motivated by the huge wage gaps between countries, as the paper contends, why do only a third stay in the United States? Furthermore, why would foreign students undertake the effort to get...