George J. Borjas Harvard University October 2005

1. IMMIGRATION IN HIGH-SKILL LABOR MARKETS: THE IMPACT OF FOREIGN STUDENTS ON THE EARNINGS OF DOCTORATES George J. Borjas Harvard University October 2005

2. 1. Introduction • Rapid growth in the number of foreign students enrolled in American universities has transformed the higher education system, particularly at the graduate level. • Fraction of doctoral degrees awarded to foreign students rose from 11.3 to 24.4 percent between 1976 and 2000. • Nonresident aliens receive a very high share of the doctoral degrees awarded in the physical sciences (36.5 percent in 2000), engineering (50.7 percent), and the life sciences (25.7 percent). • Over half of the foreign-born doctorates remain in the United States, suggesting they may have a sizable impact on the labor market for high-skill workers. • Have foreign students harmed the economic opportunities of competing native workers?

3. 2. Preview • I show that the foreign student influx has differentially affected different fields at different times. I exploit this variation in the supply shock to identify the impact of immigration on high-skill labor markets • Empirical analysis shows that a foreign student influx into a particular field at a particular time has a significant and adverse effect on the earnings of competing doctorates in that field who graduated at roughly the same time. • A 10 percent immigration-induced increase in the supply of doctorates lowers the wage of competing workers by about 3 to 4 percent. About half of this adverse wage effect can be attributed to the increased prevalence of low-pay postdoctoral appointments in fields where immigration has softened labor market conditions.

4. 3. Data • The National Science Foundation’s Survey of Earned Doctorates (SED) and Survey of Doctoral Recipients (SDR). • The SED provides a population census of all doctorates granted by a U.S. institution, with a response rate of around 92 percent. I use the SED to calculate the magnitude of the immigrant supply shock by field and year of degree. • The SDR is a biennial longitudinal file with a 7 percent sample of doctorates in science or engineering granted by U.S. institutions, and contains detailed information on a worker’s earnings. I pool data from all five waves, 1993-2001. • The SDR data contains information on labor market outcomes of doctorates in 22 science and engineering fields. Because the SDR data contains information on labor market characteristics of doctorates only in science and engineering, I restrict the analysis of the SED data to those persons who received doctoral degrees in those fields. • An “immigrant” is a person who is either a naturalized citizen or a non-citizen; all other persons are classified as “natives.”

5. 4. Doctorates awarded to foreign-born persons, 1968-2000

6. 5. The supply shock • Consider the population of persons who are granted a doctorate in field f in calendar year c. Let Mfc gives the number of immigrants in cell (f, c) and Nfc gives the corresponding number of natives. The foreign-born share in this particular field-cohort cell is given by:

7. 6. Doctorates awarded annually

8. 7. Immigrant share in doctorates awarded annually

9. 8. Immigrant supply shock, selected fields

10. 9. Doctorates awarded to foreign-born persons, 1968-2000

11. 10. Descriptive regression analysis • Let wifc(t) denote the annual earnings of worker i, who has a doctorate in field f, received his doctoral degree in year c, and is observed at time t. I estimate the regression model: log wifc(t) = pfc + xifc(t) + df + yc + t+ (df × t) + ifc(t), • where xifc(t) is a vector indicating the number of years that the worker has been in the labor market; df is a vector of fixed effects indicating the worker’s field of doctoral study; ycis a vector of fixed effects indicating the worker’s year-of-graduation cohort; tgives a vector of period fixed effects indicating the calendar year in which the worker’s earnings are observed. The worker’s experience is defined as the number of years elapsed between the time the worker is observed in a particular SDR wave and the time the worker received the doctoral degree. • A cohort is defined either by classifying workers into 3-year groups or as a 5-year moving average.

12. 11. Alternative measures of earnings • Adjusted annual salary as constructed by the NSF from information on a worker’s income per pay period. • Total annual (earned) income in calendar year prior to the survey. Although total annual income is a preferable variable, it is not available for the 1993 survey.

13. 12. Basic estimates of wage impact of immigration(Coefficient of immigrant share)

14. 13. Sensitivity analysis (coefficient of immigrant share)

15. 14. Estimation of labor demand function, part 1 log wifc(t) =  log Lfc + xifc(t) + df + yc + t+ (df × t) + eifc(t), • Lfc gives the total number of doctorates in field f and cohort c. The parameter  gives the factor price elasticity. • Potential problems: endogeneity of size-of-workforce variable and incorrect standard errors due to clustering. • Solution: two-stage estimation procedure. • First stage, get individual fixed effects and aggregate fixed effects to the field-cohort level. • Second stage, use IV to estimate vfc(t) =  log Lfc + df + yc + efc(t), • Instrument: log of the number of immigrants in the (f, c) cell.

16. 15. Factor price elasticities (IV estimates)

17. 16. Summary statistics on postdocs, native-born doctorates, 1993-2001

18. 17. The impact of immigration on the probability of being employed as a postdoctoral fellow (IV estimates)

19. 18. Factor price elasticities for workers not in postdoctoral appointments (IV estimates)

20. 19. Predicted wage impact of the 1990-2000 immigrant influx, by field

21. 20. Interpretation • High-skill labor markets likely adjusted to the supply shocks and these adjustments cloud interpretation of the results. • Suppose native students would have taken the place of the foreign students admitted to the various graduate programs if there had been a prohibition on the entry of foreign students. Total supply of doctorates in particular field-cohort cells would then have been the same and the wage structure in the doctoral labor market today would be exactly what we now observe. • Suppose native students responded by moving to other fields, or by going to law or business school. These “internal migration flows” would lower wages throughout the entire high-skill sector, not just in the fields penetrated by immigrants. The measured labor market impact of immigration would then underestimate the actual impact, since the supply response of native students arbitrages wage differences.

22. 21. Many remaining questions • Paper addressed an important component in a cost-benefit analysis of the foreign student program. But what are the benefits? Does the sizable increase in the skill endowment of the workforce accelerate the rate of scientific discovery? These benefits could be very large and accrue to particular parts of the population. • What are the consequences of the distributional impact on human capital investment decisions of native workforce? Do these consequences matter?