Having more educational opportunities should be a good thing for everyone, right? It is a boon for the students, who benefit from the increased access. And more broadly, it is a win for society, which benefits from a healthier and more skilled workforce.
But experts differ on just how—and how quickly—access should be increased. New research suggests that, at least in the short term, sharp enrollment hikes may hurt student learning.
The evidence comes from historical data from Italy. In 1961, the Italian government enacted a sweeping education reform that expanded access to science, technology, engineering, and math (STEM) degrees at state-run institutions. In just a few years, the new policy led to an abrupt increase in enrollment and in the diversity of the student population in these majors.
By analyzing the records of students enrolled in STEM majors before and after the reform, the Kellogg School’s Nicola Bianchi uncovered some unintentional consequences. Namely, the enrollment boom caused congestion in the affected majors. The student–faculty ratio skyrocketed, leaving students with less access to professors and teaching assistants. Moreover, with students now coming from a broader range of backgrounds, there was more variation in students’ levels of preparedness.
These factors made learning the relevant course material more difficult. And the drop in learning affected the financial premium associated with a STEM degree—even decades later. Income for students who should have benefited from the increased access to education stayed largely flat, while income for students who had access to STEM majors pre reform actually decreased.
“Policies may have indirect effects, which can decrease the benefits for targeted students or generate negative effects for other students."
“Obviously, new policies should bring benefits to the students,” says Bianchi, an assistant professor of strategy. “But what my research shows is that an education system is a complex setting. Policies may have indirect effects, which can decrease the benefits for targeted students or generate negative effects for other students who shouldn’t have been affected.”
Autopsy of an Education Reform
To understand the data, you need a short primer in Italian education.
There are three main types of high schools in Italy: academic schools that prepare students for a university education, professional institutes that prepare students for direct entry into specific careers and vocations, and technical schools. Among technical schools, there are several tracks, including the industrial track, which trains students for fields like mechanics, electronics, biotechnology, and construction; the commercial track, which prepares students for fields like accounting; the language track; and the education track.
Though some university courses were open to a wide range of students, STEM courses had traditionally been restricted to students who graduate from academic schools. But Italy’s education reform of 1961 opened STEM majors to students in industrial-track technical high schools (though not to students in other technical schools). For three years, this access was still somewhat restricted by enrollment caps; by 1965 these caps were lifted.
The reform had an exaggerated effect on student enrollment in a very short amount of time. Bianchi wanted to know: Did the government’s rapid expansion of access to STEM majors affect how much students learned in the classroom?
He needed to be able to track students from high school through university. So Bianchi visited Milan, going from high school to high school and knocking on principals’ doors to glean access to thousands of student archives from 1958–1968. At the University of Milan, he created digitized records of university transcripts for the same set of students. This allowed him to match up a given student’s high school and university records.
Fewer Resources, Less Learning
In order to determine whether the influx of new students had affected learning, Bianchi compared the grades of academic-track students in the years before and after the reform was implemented. In other words, he examined the very students whom the reform should not have affected. (Because assignments were not graded on a curve, the letter grades students earned are a suitable proxy for learning.)
Bianchi found that the grades of the academic-track students got worse after the reform. Why did student learning take such a hit? Bianchi’s evidence suggests an overtaxing of university resources that hurt the classroom experience.
The decrease in learning was particularly steep after 1964, when all enrollment caps were finally removed—meaning that students would have had an even harder time seeking out teachers and teaching assistants because so many other students were doing the same thing. And the drop in grades was most pronounced in courses where the material would have been unfamiliar to industrial-track students. In these courses, the student body would have been particularly diverse in terms of levels of preparedness, likely making classroom instruction or discussion less effective.
“The new students had a different set of skills relative to incumbent students,” Bianchi says.
In his study, Bianchi also looked at the long-term implications of learning less in the classroom. He linked the students’ records to their 2005 tax returns. How much money, he wondered, were these former students making now that they were in their fifties and sixties?
Students who graduate with university STEM degrees can generally expect to make more income than students who earn other university degrees. But notably, Bianchi found that for academic-track students who attended the university when the majors were at their most congested, the income premium decreased.
Yes, supply and demand is responsible for some of this decrease—after all, with more STEM graduates, there was simply more competition for jobs.
But Bianchi’s model estimates that three-quarters of the post-reform decline in income can be explained by the students’ poorer university experience, including the overcrowded classrooms and generally less prepared classmates.
Bianchi also noticed something else surprising: some academic-track students with an aptitude for STEM actually migrated away from these majors after the reform—in favor of other restricted majors such as medicine and law.
“This tells us that the reform affected major vocation choices of students the policymakers did not want to alter at all,” Bianchi says. “One of the bigger contributions of this paper is to identify that you could have negative effects from an education reform while the students are still in school, even before they enter the labor market.”
Exploring the Industrial Track
Overall, Bianchi’s results suggest that the reform had negative consequences for academic-track students. But what about the intended beneficiaries, the industrial-track students? How did they fare?
Bianchi compared the post-reform outcomes of industrial-track technical high school graduates, many of whom took advantage of the increased university access, and commercial-track technical high school graduates, who were not included in the reform.
Relative to the commercial-track students, those on the industrial track initially appeared to benefit from the reform, with long-term income increasing by nearly 15%. But the gains did not last. After the enrollment cap was entirely lifted and the STEM majors experienced overcrowding, the benefit to industrial-track students dropped to 6% (an amount so small it could have occurred by chance).
That is, although these industrial-track students received more education than the professional-track students, they did not actually benefit financially from the extra education.
Bianchi’s work shows that dramatic educational policy reforms can have consequences that detract from their intended goals.
When governments want to increase access to education, Bianchi explains, often their solution is to encourage students into certain fields in state-controlled universities. The Italian experiment shows that this way of implementing increased access can hurt student learning—and long-term earnings.
Bianchi instead recommends a more focused approach to increasing access, such as identifying students who have a high aptitude for STEM fields but who face barriers to furthering their education.
“You might want to instead target people that show promise, and give them vouchers or give them scholarships so they can choose their own majors,” Bianchi says. Scholarships would drive students to specific universities; vouchers would allow students to attend the university of their choice. Either option would strike a better balance between increasing access and not overtaxing the resources of a few select colleges or majors.
The research also highlights the importance of some elements of the classroom environment—namely access to teaching fellows and professors, and the quality of a student’s peers. Such factors can tangibly affect how well students learn. Policymakers who anticipate that a new reform will boost enrollment should strongly consider hiring more instructors to minimize overcrowding and help students who are less prepared.
“If you increase the scope of the education market, a lot of important challenges must be thought through [in advance] because they can have long-lasting effects on the students involved,” Bianchi says.