Ethiopia’s Tigray War not only caused immense human suffering and economic disruption but also eroded trust within local communities. New evidence from the Young Lives study shows that individuals exposed to violence report lower trust in their neighbours, highlighting how civil war can damage the social ties that underpin cooperation and recovery after the fighting ends.
Violent armed conflict has long been a persistent feature of humanity. Conflicts have such devastating effects on economic development that some have described them as “development in reverse” (Collier et al. 2003), and post-conflict reconstruction attracts billions of dollars in international financing annually. The typical focus is on physical and economic recovery, such as rebuilding infrastructure, stabilising economies, and restoring public services. But what happens to trust, the social glue that is critical for communities to cooperate, markets to function, and peace to hold (Rohner et al. 2013)?
Evidence on whether conflict impacts trust is contested. Several studies have produced both positive and negative findings, and one prominent review concluded that the average effect of war violence on trust is close to zero (Bauer et al. 2016). A key question is whether this genuinely reflects modest effects, or whether important variation (such as which type of trust is being measured and how much violence individuals experienced) is being averaged away.
Our new research, using novel data from Ethiopia’s Tigray War collected through the Young Lives study, sheds light on both questions (Freund and Favara 2026).
Ethiopia’s Tigray war
After two decades of rapid economic growth, Ethiopia descended into civil war in November 2020 when fighting broke out between the federal government and the Tigray People’s Liberation Front. The conflict spread into the neighbouring Amhara and Afar regions before a peace agreement was signed in November 2022. By 2022, it had become the deadliest conflict in the world – with more battle-related deaths than the wars in Ukraine, Yemen, Myanmar, Nigeria, Somalia, Mali, and Burkina Faso combined (Obermeier et al. 2023). Millions were displaced, food systems were destroyed, and independent investigations documented ethnic cleansing targeting Tigrayans (Human Rights Watch 2022, United Nations General Assembly 2022).
Getting accurate data on conflict exposure
Obtaining accurate individual-level data on conflict exposure is difficult. Standard surveys may undercount conflict experiences, as respondents may be reluctant to discuss traumatic events with an unfamiliar enumerator, particularly where they fear identification or retribution (Brück et al. 2016).
We addressed this by using Audio Computer-Assisted Self-Interviewing (ACASI). With ACASI, participants listened to pre-recorded questions asking about exposure to conflict experiences through headphones and recorded responses on a tablet, without any enumerator present. In a randomised pilot, conducted before the main survey, testing the feasibility and accuracy of ACASI versus traditional face-to-face interview, individuals assigned to ACASI were 18% more likely to report at least one conflict experience than those in a standard face-to-face interview (von Russdorf et al. 2024).
We used this approach in the seventh round of the Young Lives study, which has tracked the same individuals in Ethiopia since 2002, collecting data from over 1,500 young adults in 2023–2024. Overall, 57% reported at least one conflict experience. Respondents in Tigray had the highest exposure, reporting on average five of the nine conflict events they were asked about. Across the full sample, 38% had lost a family member or friend to the conflict, 30% had a family member or friend disappear, one in five had personally experienced physical violence, and one in ten had personally experienced sexual violence.
How did conflict affect trust in Ethiopia?
Conflict damages trust between neighbours, but not trust in strangers. We find that greater conflict exposure substantially reduces trust towards people in one’s own neighbourhood. We find no evidence, however, that conflict reduces generalised trust or trust in people from other neighbourhoods; the negative effects on trust appear to be concentrated within the local community.
The decline in neighbourhood trust is concentrated at high exposure levels. The relationship between violence and local trust is not linear. Individuals who experienced only one or two conflict events since the outbreak of the Tigray war show no decline in neighbourhood trust compared to those with no exposure at all. If anything, there is suggestive evidence that limited shared adversity may strengthen local bonds. The damage to trust is concentrated among those most heavily exposed (those who experienced five or more violent events), who report neighbourhood trust roughly 11 percentage points below those who experienced no conflict.
This non-linear pattern may help explain why existing research has produced contradictory findings. Studies drawing on populations with limited exposure may find null or even positive effects on trust, while those focusing on heavily affected communities may find negative effects.
The effects are pervasive across groups and types of exposure. The decline in neighbourhood trust is not confined to those most directly targeted. We find declines in neighbourhood trust among both Tigrayans, who faced the most severe violence, and individuals from other ethnic groups. Effects are present for both men and women, though women additionally show a reduction in generalised trust that men do not.
Furthermore, indirect exposure – through the experiences of family members and friends rather than personal victimisation – also reduces neighbourhood trust. Communities can therefore experience erosion of social cohesion even where individuals have not been personally harmed.
Why does conflict erode local trust?
We examine several potential pathways, though the evidence here is descriptive rather than causal.
We find that conflict exposure is strongly associated with higher symptoms of post-traumatic stress disorder (PTSD), anxiety, depression, and stress. PTSD, in particular, involves social withdrawal and heightened perceptions of threat, and research has shown that PTSD can lead people to judge those around them as less trustworthy (Saraiya et al. 2019). Greater conflict exposure is also associated with lower perceived safety in one’s neighbourhood, which may discourage social interaction and participation in local collective activities, thereby weakening social ties.
Conflict exposure is additionally associated with higher rates of physical displacement and greater food insecurity. Displacement severs existing social networks and places individuals in communities where they lack established relationships, while food insecurity can have a negative impact on collective social functioning. Together, these patterns suggest several interrelated psychological and material pathways through which violence may undermine the everyday interactions on which neighbourhood trust depends.
Key takeaways for post-conflict policy
From a policy perspective, our results have implications for how post-conflict programmes are designed. Our findings highlight the importance of considering social capital in post-conflict recovery and suggest that post-conflict interventions may need to account not only for the restoration of physical and human capital, but also for the social relationships that underpin everyday cooperation. Community-level programmes that restore mental health, perceived safety, and create opportunities for local interaction should be an intentional part of reconstruction efforts, not an afterthought to infrastructure.
In developing country contexts such as Ethiopia, where access to mental health services remains extremely limited, our findings underscore the importance of integrating psychosocial support into broader post-conflict reconstruction strategies.
References
Bauer, M, C Blattman, J Chytilová, J Henrich, E Miguel, and T Mitts (2016), "Can war foster cooperation?" Journal of Economic Perspectives, 30(3): 249–274.
Brück, T, P Justino, P Verwimp, A Avdeenko, and A Tedesco (2016), "Measuring violent conflict in micro-level surveys: Current practices and methodological challenges," World Bank Research Observer, 31(1): 29–58.
Collier, P, V L Elliott, H Hegre, A Hoeffler, M Reynal-Querol, and N Sambanis (2003), Breaking the Conflict Trap: Civil War and Development Policy, World Bank, Washington, DC.
Freund, R, and M Favara (2026), "Love thy neighbour? Violent armed conflict and trust: Evidence from Ethiopia," Journal of Development Economics, 182: 103767.
Human Rights Watch (2022), "We will erase you from this land: Crimes against humanity and ethnic cleansing in Ethiopia's Western Tigray Zone."
Obermeier, A M, H Strand, and G Berry (2023), Trends in State-Based Armed Conflict, 1946–2022, Peace Research Institute Oslo, Oslo.
Rohner, D, M Thoenig, and F Zilibotti (2013), "War signals: A theory of trade, trust, and conflict," Review of Economic Studies, 80(3): 1114–1147.
Saraiya, T, D Fareri, T López-Castro, D Hien, E Fertuck, and R Melara (2019), "The social cognitive appraisal of trustworthiness in individuals with dimensional levels of post-traumatic stress symptoms," European Journal of Psychotraumatology, 10: 1697582.
United Nations General Assembly (2022), "First report of the International Commission of Human Rights Experts on Ethiopia," A/HRC/51/46.
Von Russdorf, S, L Ahlborn, A Hidalgo, G McQuade, and M Favara (2024), "A sound methodology: Measuring experiences of violent conflict through audio self-interviews," Economics Letters, 242: 111879.