Children in rural schools are often outperformed by their urban peers, especially in developing countries. As Scientific Director of TSE Human Capital Center, Matteo Bobba leads a team of economists focused on how to close these stark educational gaps. In a NBER working paper with coauthors including former TSE PhD researcher Tim Ederer, he reveals how smart compensation policies — designed using data on teacher preferences and effectiveness — could dramatically reduce inequality in Peru.
What motivated you to investigate Peru’s educational divide?
We wanted to understand how the way teachers are paid and assigned can either reinforce or reduce inequality. Teachers are central to children’s learning and, because they can move, they respond to where conditions are better. Yet most public systems pay them the same regardless of where they work.
In Peru, which has many small, isolated communities facing persistent structural inequalities, we saw an opportunity to study the impact of reforms that use teachers’ wages to enhance the fairness and effectiveness of the education system.
How do teachers decide where to work in Peru?
We know surprisingly little about why teachers choose certain schools. In systems with fixed salaries, teachers likely prioritize non-monetary factors like working conditions, cultural factors, prestige, connectivity, or safety. In our survey, nearly two-thirds of applicants ranked proximity to home as a top-three priority.
W e found that Peru’s best teachers tend to choose urban schools, which are much more likely to have libraries, sports facilities and internet access. In rural schools, which often lack even basic facilities such as electricity and drinking water, we found that teachers are less qualified and students perform far worse on mathematics and language tests.
How have government reforms changed the game?
In 2015, Peru introduced a more transparent, centralized recruitment system for public teachers, requiring all applicants to take the same competency test. In order of their score, applicants then choose their preferred schools from a national list of vacancies. This reform helped us to build an exceptionally rich dataset for observing teacher quality and job choices, as well as their impact on student achievement.
By offering wage bonuses to teachers in rural areas, the Ministry of Education also created the kind of natural experiment that researchers dream of. Bonuses were tied to precise population and geographic thresholds, favoring smaller, more remote localities. This meant we could compare otherwise similar schools just above and below the eligibility cutoffs to identify the causal effects of higher pay.
What happened when salaries rose in the more remote schools?
The local effects were impressive. Rural schools that qualified for the wage bonus were able to attract teachers with much higher scores on the national exam. These newly recruited teachers performed substantially better, with their students’ test scores rising by about half a standard deviation in math and one-third in language.
Even so, the bonuses were too small to fully compensate for rural disadvantages. Teacher effectiveness in rural schools remains substantially lower and explains around a quarter of the urban-rural test-score gap.
Did higher wages also change how existing teachers behaved?
We found that higher pay drew stronger new applicants into rural schools, but did not noticeably change effort among existing staff. This suggests that salary incentives can be powerful tools for attracting talent but may not strongly motivate those already in post.
Your study is a powerful demonstration of TSE’s expertise with structural models. What’s the value of this approach?
Going beyond evaluating what actually happened in Peru, our model lets us ask what could happen under different policies. In other words, we can test not just what worked, but what might work even better.
To simulate alternative compensation systems, our teacher sorting model combined two crucial insights. First, teachers have dramatically different preferences over job attributes. Second, their effectiveness can vary considerably for different types of students.
What do those simulations suggest? What are their policy implications?
When we simulate pay policies we found that large equity and efficiency gains are attainable at no additional cost for the government. Importantly, these gains mainly come from attracting applicants with higher average effectiveness into public teaching rather than reallocating existing teachers to exploit their comparative advantage in teaching different types of students. This suggests that, despite the large potential gains from teacher reallocation, leveraging the extensive margin of recruitment is more cost-effective Overall, our findings suggest that Peru could substantially reduce the teacher-quality gap at no additional cost by tailoring incentives more intelligently.
What is the main lesson for governments trying to reduce educational inequality?
That compensation policy is not just an accounting issue — it’s a policy lever. By aligning teachers’ preferences and abilities with the needs of schools and students, we can make public education more equitable and more efficient. The challenge is to design systems that recognize heterogeneity among teachers and to use this to society’s advantage. Pay reform can then become a driver of opportunity.
KEY TAKEAWAYS
• Pay rigidity reinforces inequality – High-quality teachers choose urban schools with better amenities, leaving rural students further behind.
• Money talks — Peru’s rural wage bonuses attracted better teachers and boosted learning outcomes, but not enough to catch up with urban schools.
• Spend it wisely – Targeted bonuses that account for teacher skills and preferences could close the urban-rural gap at very little cost.
