Renzo Severino

Immigrants, New Students, and Academic Performance in Peru

Renzo Severino

Immigration from Venezuela to other countries in South America rose sharply starting in 2015 in response to rapidly deteriorating economic conditions, food and medicine shortages, and increased political repression.

Venezuelan presence in Peru--the main destination in South America after Colombia--reached 800,000 immigrants by the end of 2019, increasing from around 7,000 people just four years earlier. Between 10% and 15% of the newly arrived immigrants were children of schooling age.

This paper studies the impact of the introduction of immigrant students on the performance of natives. It is the first to highlight the similarities between recently arrived immigrant students and natives new to their high school. It connects two strands of literature: research on the impact of immigrants on native students and research on domestic student mobility, both of which contain mixed results.

Several features of the Peruvian school system aid to the identification of a causal effect:

Start of year: Within a school-grade, children are assigned a classroom at the start of the academic year (March)

No Tracking: There is no established practice of sorting students based on past academic performance. Nevertheless, all regressions account for this possibility.

No Movement: All students of a particular classroom take all their classes in the same room, with teachers moving between classrooms.

No Influence: Students or parents typically have no influence on classroom assignment

This allows me to take the entrance of new groups of students as quasi-random events conditional con school-by-year fixed-effects. In particular, I compare the standardized test scores of Peruvian students in classroom with varying numbers of Venezuelan students or different numbers of new Peruvian students.

Introducing immigrant students or new native students impact incumbent native students in a similar way: there are no statistically significant differences in effect sizes or in their pattern of heterogeneity. This suggests that a shared characteristic between immigrant students and newly enrolled natives drives the results.

The evidence I find is consistent with a story of natives reacting to the influx of new students to their school--both native and immigrant--by reducing their effort in learning. This is consistent with natives spending more time either befriending or antagonizing recently arrived students instead of studying.

I offer here a novel and subtle interpretation of what is behind the effect of immigrant shocks on academic performance. Policymakers should think carefully before attributing observed effects to characteristics inherent to immigrants instead of characteristics shared with a subset of native students. For example, public resources may be better spent on teacher training programs to help incorporate new students into high schools instead of enforcing policies specifically directed to immigrants. Find the draft here