For years, increasing the number of students pursuing science, technology, engineering, and mathematics (STEM) degrees has been a priority for President Barack Obama, other policymakers, and business leaders. More STEM graduates, according to the White House, “can both create jobs and enhance our national competitiveness.” But as research by Nicola Bianchi, who is finishing his Ph.D. in economics at Stanford University in Palo Alto, California, illustrates, a rapid hike in STEM students may produce results different from what those policymakers expect. For example, when the ranks of STEM majors were expanded, Bianchi found, many of the best students left science for better-paying majors.
Bianchi examined the outcome of a dramatic reform of the Italian education system in the 1960s that more than doubled the number of university STEM majors in only a few years. Changed regulations allowed graduates of technical high schools to enroll in majors previously restricted to graduates of more elite academic high schools. “Peculiar features” of the Italian system permitted Bianchi to discern educational effects, and tax records allowed him to follow earnings. “[T]he direct effects of the reform were as intended: many more students enrolled and many more obtained degrees,” Bianchi writes in a paper based on his dissertation.
For the US, the evidence suggests that students are already highly elastic [that is, willing to change fields] with respect to economic returns in STEM fields.
But the reforms also had other effects, less anticipated and less desirable. First, “many more people enrolling in these STEM majors all at once made learning more difficult [resulting in] the students acquiring … a lower amount of skills,” Bianchi tells Science Careers in an interview. Why? Resources didn’t expand as much as enrollments did. Also, before the reform, “the professors were used to teaching a very homogeneous type of student;” they now had to teach a more heterogeneous population. Curricula were not adjusted to account for the somewhat different preparation that the technical students brought to their university studies.
Many of the technical school graduates nonetheless managed to complete their degrees—often, though, at lower levels of achievement than the academic school graduates. This appears to have diluted the value of the degree, as shown by the fact that “the new university-educated [technical school] individuals did not earn higher incomes than people from earlier cohorts who were denied access to STEM majors,” Bianchi writes in the paper. What happened? “Economic theory tells us that [when] we increase the supply of STEM-educated workers by a lot, by whole lot, … wages should go down unless the demand for these workers was so high that they were all absorbed, ... which I think is unlikely,” he explains in the interview.
Another important consequence: Many of the best academic high school graduates were driven away from the majors affected by the reform and into “other majors, high paying ones, [that] were kind of sheltered by being not accessible to the technical students,” Bianchi says. The academic students “moved to other majors like medicine, which were high paying but without the technical students and therefore [without] the negative impact on learning” created by the expansion.
“The most able students are those that are more flexible in major choice. … They can succeed in different occupations,” Bianchi emphasizes in the interview. “[F]or the US, the evidence suggests that students are already highly elastic [that is, willing to change fields] with respect to economic returns in STEM fields,” he writes in the paper. Research by B. Lindsay Lowell of Georgetown University and co-authors has also found a tendency among top U.S. STEM students to seek other career options in response to changes in the attractiveness of STEM careers.
The Italian reform, Bianchi stresses, had particular features that make it unique; he declines to draw conclusions from it about current policy proposals. Nonetheless, his research strongly suggests that the results of manipulating student populations for policy ends can be more complicated than predicted.