Working Paper: CEPR ID: DP13509
Authors: Dominic Rohner; Alessandro Saia
Abstract: This paper studies the impact of school construction on the likelihood of conflict, drawing on a policy experiment in Indonesia, and collecting our own novel dataset on political violence for 289 districts in Indonesia over the period 1955-1994. We find that education has a strong, robust and quantitatively sizeable conflict-reducing impact. It is shown that the channels of transmission are both related to economic factors as well as to an increase in inter-religious trust and tolerance. Interestingly, while societal mechanisms are found to have an immediate impact, economic channels only gain importance after some years. We also show that school construction results in a shift away from violent means of expression (armed conflict) towards non-violent ones (peaceful protests).
Keywords: education; conflict; civil war; fighting; schools; returns to education; polarization; protest
JEL Codes: C23; D74; H52; I20; N45
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| School construction (L74) | Conflict likelihood (D74) |
| Increase in schools built (H52) | Decrease in conflict risk (D74) |
| Education (I29) | Conflict likelihood (D74) |
| Education (via schools) (I29) | Economic opportunity costs of fighting (H56) |
| Education (via schools) (I29) | Interreligious trust and tolerance (Z12) |
| School construction (L74) | Nonviolent forms of expression (peaceful protests) (D74) |
| School construction (L74) | Violent means of resistance (D74) |