��ɫ��Ӱ

Many states are confronting a fresh crisis brought on by the COVID-19 pandemic: Rising teacher shortages as educators exit the profession rather than return to the classroom. To fill staffing gaps, some states are lowering standards for who can teach.

The policy response could have real effects on the academic qualifications of teachers, says Bradley Larsen, a ��ɫ��Ӱ assistant professor of economics. In a  released this week by the National Bureau of Economic Research, Larsen shows how licensing rules affect who becomes a teacher — by deterring people who are the least academically qualified from entering the profession. If states make it easier to become an educator, the academic preparation of teachers is likely to suffer.

“With COVID-19, policymakers may need to lower standards to get teachers into classrooms,” he says. “For the first time, we now have evidence of what that might mean: more teachers with lower academic ability entering the profession.”

The insight may seem intuitive. But many economists have long argued that occupational licenses, for teachers and other professions, do more harm than good. In a  earlier this year, Larsen and a team of researchers find that, for an array of home improvement services, reputation is more valuable than licensing in consumers’ eyes.

Previous research on teacher licensing has looked at average quality and found that stricter standards don’t improve overall quality. 

But Larsen, a faculty fellow at the ��ɫ��Ӱ Institute for Economic Policy Research (SIEPR) and the study’s lead author, delves deep into the data. He looks specifically at what effect undergraduate course requirements for prospective teachers have on who becomes a teacher and finds that additional requirements significantly reduces the number of would-be educators coming from low-ranked colleges.

“These results indicate that stringent teacher certification may be effective at weeding out less-academically prepared candidates from the profession,” Larsen says. He finds this outcome across public school districts generally, including those in low-income, high-minority areas.

The study is significant, too, because it is the most expansive and in-depth to date on the impact of certification laws on teachers specifically. “No one has looked at the question of teacher quality the way we have,” Larsen says.

A dataset unrivaled in scope

Policymakers have long grappled with ways to attract college graduates with strong qualifications to become public school teachers. In many countries, according to Larsen, it’s not unusual for the best college students to become K-12 educators. But in the United States, data show that public school teachers graduated high school with GPAs in the bottom half of their class.

“Everybody is concerned about having high-quality public education, but people disagree on what you should do to achieve it,” Larsen says.

To many, licensing might seem like an obvious quality guarantee, and these proponents argue for stricter licensing. But critics argue that licensing drives up costs and drives away top talent. Larsen calls this the Wall Street dilemma: If it’s too hard or costly for highly capable people to become teachers, they may pursue more lucrative careers in finance or law. 

One way state policymakers have tried to address the problem is through teacher certification. Standards  vary widely and have changed significantly over time. Some states, like Kentucky and North Dakota, make it very difficult to become a teacher. Rules in Florida and Georgia, on the other hand, are considerably more relaxed. Still others, like Michigan and Arizona, have gone both ways at various times.

The variation in state rules across time gave Larsen and his collaborators — ��ɫ��Ӱ economics PhD students Ziao Ju and Chuan Yu and Princeton assistant professor of economics Adam Kapor — a unique opportunity to study teacher licensing rules as a natural experiment. They collected information on 37 distinct dimensions of teacher certification requirements across 50 states and Washington, D.C. from 1991 to 2007. 

The researchers find that rules requiring future teachers to take additional courses in, for example, math, English, and the natural sciences, account for the lion’s share of differences in state licensing rules and the magnitude of their effect on teacher quality. They focus on teachers’ academic ability (one measure of quality) based on the average SAT scores of the teachers’ undergraduate institutions.

“We are concerned about all dimensions of teacher quality, but a large chunk of the teacher certification debate revolves around this particular dimension – the academic qualifications of teachers, how good their resumes look relative to those who chose other careers,” Larsen says. “For example, are the best students becoming teachers, or going to some other job? And are licensing laws changing that pattern?” 

Insights for policymakers

Larsen’s analysis shows that states requiring more academic coursework for future teachers have fewer secondary school educators coming from less-selective colleges. In other words, aspiring teachers who come from less-selective colleges are far more likely to step off the education track when asked to take additional classes as part of their certification.

“These are very robust results: In every test we run — controlling for many different kinds of changes in policy, labor markets, and such across states and years — we consistently see that low-quality teachers are being driven out by increased stringency,” Larsen says. 

While the researchers find that raising the licensing bar also lowers the number of high-quality teachers, they find that that effect is not statistically significant. They also find that stricter requirements don’t appear to disproportionately harm high-poverty or high-minority districts or reduce racial diversity among teachers.

Larsen is quick to caution that the study doesn’t say anything about other measures of teacher quality, such as classroom effectiveness in raising student test scores, long-term outcomes for students, or parental satisfaction. And because of collective bargaining and other distinct characteristics of education, the findings don’t necessarily generalize to other occupations.

But, he says, there are other important takeaways for policymakers and public education reformers. Most researchers who analyze licensing outcomes focus on averages without looking at segments of the teacher population that are most affected. Larsen’s approach to analyzing the whole distribution of teacher characteristics, rather than just the average, builds on his  methodology studying effects of trade with China on U.S. wage inequality, described in a SIEPR policy brief.

“Policymakers have to look beyond averages to really understand whether teacher certification laws — as well as occupational licensing laws more broadly — are effective,” Larsen says. “Our analysis allowed us to dive into the data and see what happens specifically to people who appear to be less qualified and those who have greater abilities.”

Larsen also says that this study — and the statistical model his team designed — offer a new approach to understanding aspects of teacher licensing that work or not. For example, their analysis uncovers evidence to suggest that other teacher certification requirements, such as exams, background checks, and citizenship requirements, don’t improve quality. 

“If compliance with these other certification requirements burns real resources (for teachers, governments agencies, and teacher prep programs) without affecting quality,” the paper states, “the effectiveness of these other requirements is questionable.”

He says that, for states looking to relax licensing rules to address COVID-19-induced teacher shortages without compromising quality, waiving non-course requirements may be the better policy.

Krysten Crawford is a freelance writer