Document Type

Working Paper

Publication Date

Summer 6-1-2011

JEL Codes

F22, I20, O15, I28, J61

Working Paper Number



In deciding upon whether to pursue an undergraduate education in the United States, a foreign student considers the expected probability of securing US employment after graduation. The H-1B visa provides a primary means of legal employment for college-educated foreign-nationals. In October 2003, the government drastically reduced the number of available H-1B visas, hence lowering a college-educated foreign-born worker’s probability of finding US employment, and possibly discouraging highly qualified international students from attending US colleges and universities. However, citizens from five countries are de facto exempt from the 2003 H-1B visa restrictions. Using students from these five exempt nations as the control group and other international students as the treatment, we study the effects of the 2003 H-1B policy change on the pool of international applicants to US schools. We use two datasets: (i) College Board SAT score data on prospective international applicants; and (ii) SAT and high-school GPA data on international applicants to a single highly-selective university. Our difference-in-difference estimates show that restrictive immigration policy has had an adverse impact on the quality of prospective international applicants, reducing their SAT scores by about 1.5%. This effect is driven mostly by a decline in the number of SAT score reports sent by international students at the top-quintile of the SAT score distribution, suggesting that the restrictive immigration policy disproportionately discourages high-ability international students from pursuing US education. Our results are robust to alternative specifications, including the use of high-school GPA as a measure of applicant ability.


Jessica Mawhirt, Kelly Motta, and Hejia Wang supplied valuable research assistance. We thank David Card, Eric Larsen, Todd Sorensen, Melanie Guldi, Barbara Roback, and seminar participants at the University of Connecticut, Mount Holyoke University, the 2010 EALE/SOLE conference, and the 2010 IZA Annual Migration Meeting for helpful comments and suggestions. Sparber thanks the University of Puget Sound for providing summer research support.

Included in

Economics Commons