Workshop on Exposure to Political Violence and Economic Behaviour

Monday 5th April 2021 13:50-16:30

Tuesday 6th April 2021 13:50-16:30

We are happy to announce that the online workshop on Exposure to Political Violence and Economic Behavior will take place on April 5-6, 2021.

The workshop is going to be the first of a series of similar dissemination events on the microlevel consequences of political violence exposure. Our goal is to bring together a core group of scholars from political science, economics and psychology to share and discuss their latest works on political violence and its consequences. In the first event of this series we will be focusing on economic behavior. Please join us for a very interesting program of presentations by esteemed scholars of political violence. Participation in the workshop is by registration only and is free of charge.

Please note this event will be held over Teams.

Joining instructions will be sent to you the day before. We look forward to seeing you soon!

Register

Workshop Program

April 5 - Monday

13:50 Opening Remarks

14:00 Kristian Gleditsch, Essex University

Title: Can Information Change Views under Polarization? A Survey Experiment on Support for the 2016 Colombian Peace Agreement
Authors: Mónica Pachón, Manuela Muñoz, Cesar Mantilla, Kristian Skrede Gleditsch
Abstract: The proposed peace agreement between the Colombian government and the Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia, FARC) was narrowly defeated in a referendum on 2 October 2016, with 50.2% of votes cast against the agreement. In this study we examine how individuals evaluate specific information about the peace process agreement, based on their prior opinions or beliefs, in an environment of political polarization. We conduct two survey experiments to examine how support for the peace agreement varies by framing the information about the economic costs or benefits of peace, and giving factual information about the trade-offs in terms of justice and the number of victims of the conflict.

Discussant: Jordi Munoz

14:45 Maarten Voors, Wageningen University

Title: Effect of an asset transfer program on mental health of displaced persons and host populations in Democratic Republic of Congo: a randomised controlled trial

Authors: Maarten Voors
Abstract: Humanitarian crises affect over 200 million people globally, exacting a large toll on population mental health. We assess the impact of an asset transfer program on the mental health of internally displaced persons and host populations in eastern Democratic Republic of Congo (DRC).
We conducted a randomised trial among vulnerable households residing in 25 villages in North Kivu Province, DRC, where a large United Nations program responds to population displacement by providing asset transfers in the form of vouchers for essential household items (EHI). Households in need of assistance but outside the program’s standard eligibility criteria were randomly assigned (1:1) to “voucher” or to “no intervention”. Households in the voucher group received US$50-92 worth of vouchers to use at a fair where EHI, such as blankets, clothes, buckets, and pans, were sold. The head woman of each household was interviewed just before the fair, six weeks after, and one year after. The primary outcomes were standardized indices of adult’s mental health, children’s physical health, social cohesion, and resilience. Effects were assessed in least-squares regression models adjusting for baseline levels.
Between August 2017 and March 2018, we enrolled 976 households in the study. 488 were randomly assigned to the EHI voucher and 488 to no intervention. 88% of respondents were female. At baseline, 33% of respondents had an anxiety/depression score suggesting clinical significance. At six weeks, the voucher group had a 0.32 standard deviation units (SDU) improvement on the mental health index (95% CI 0.18 to 0.46), and, after one year, the voucher group had a 0.19 SDU improvement (95% CI 0.02 to 0.34). There were no effects on the child health, social cohesion, or resilience indices. 
Asset transfers can improve the mental health of vulnerable populations in humanitarian crises.

Discussant: Anke Hoeffler

15:30 Break (15 min)

15:45 Arzu Kibris, Warwick University

Title: The Impact of Exposure to Political Violence on Risk and Ambiguity Attitudes

Authors: Arzu Kibris and Neslihan Uler

Abstract : We conduct an incentive-compatible field experiment with a large representative sample to study how exposure to political violence in a civil conflict context affects risk and ambiguity preferences of individuals. We identify random exposure to violence by relying on a natural experiment in Turkey created by the military institutions and the long running civil conflict in the country. We show that the effect of violence exposure on risk and ambiguity preferences depend on the type of exposure. We find that while being exposed to the conflict environment induces individuals to become more risk-tolerant, having traumatic direct experiences in that environment creates the opposite effect and renders individuals extremely risk averse. Such individuals also become more averse to ambiguity. We also show that time of exposure should be considered in determining the overall effects. Overall our findings indicate that preferences on risk and ambiguity are history-dependent.

Discussant: Maarten Voors

16:30 Albert Falco Gimeno, University of Barcelona

Title: Retrospective Bullets: The Conditional Effect of Terrorism on Incumbent Support

Authors: Albert Falco Gimeno and Jordi Munoz

Abstract: Empirical evidence on the effects of terrorism on support for the incumbent is so far inconclusive. Models of retrospective voting predict a negative effect, as terrorism is detrimental to voters’ welfare. However, the literature has also described the “rally around the flag” effect, that runs in the opposite direction. Following a terrorist attack, voters may cling to the incumbent government to receive protection, and often opposition parties express explicit support to the government. We claim that both mechanisms can be teased out if we take into account the timing of the attack relative to election. The rally around the incumbent should fade out at a faster pace than the accountability mechanism, so we expect a positive effect of terrorism on incumbent support when the elections occur shortly after the attack, and a negative effect when there is more temporal distance. We evaluate the argument against a data-set of all deadly attacks by ETA in Spain during the period 1977-2004 matched with municipal-level national election results. Using municipality- and year- fixed effects models to control for various sources of heterogeneity, we find an average negative effect of exposure to terrorism on incumbent support. However, attacks occurred during the last third of the term appear to trigger rallies, and incumbents reap an electoral benefit. The temporal heterogeneity is especially pronounced for the most deadly and indiscriminate attacks.

Discussant: Jessica di Salvatore

April 6 - Tuesday

14:00 Anke Hoeffler, University of Konstanz

Title: Mental Health and Everyday Violence in the Eastern DRC
Authors: Samuel Carleial, Anke Hoeffler, Anke Koebach, Sabine Schmitt and Katy Robjant

Abstract: People in the DRC have experienced waves of violence and the Eastern part of the country has been the worst affected. The consequences are severe, there is widespread extreme poverty as well as physical and mental health problems. While there is currently no large scale armed conflict, rural communities experience considerable everyday violence while still recovering from the trauma of war. Using a rich dataset from seven communities in rural North and South Kivu we analyse the prevalence and correlates of violence. While we can confirm that adolescents and young adults are at high risk of victimization as well as being more likely to perpetrate violent acts, we find a number of puzzles. Although the prevalence of Post-traumatic Stress Disorder is very high, it appears to have little impact on the perpetration of violence, but makes continued victimization more likely. These mechanisms need to be better understood while the DRC is making an attempt to provide mental health care for its traumatized population.

Discussant: Arzu Kibris

14:45 Philip Verwimp, Universite Libre De Bruxelles

Title: Health in Early-Childhood, Civil War and Teenage Cognitive Development

Authors: Philip Verwimp

Abstract: This paper uses tests scores for different academic disciplines in Burundi to infer the effect of undernutrition in early childhood on cognitive development 10 years later. We use the universe of test score data from the Concours National, a nationwide competitive test taken at the end of primary school which determines access to secondary school for the period 2010-2012. We combine these data with anthropometric measurement obtained from a household survey in early childhood. Employing the historic record of the location and timing of the civil war we define exposure to civil war in childhood and use it as an IV for child health status. We find that a decrease of one-standard deviation in the health status - caused by exposure to civil war - leads to a 3 to 5% lower score on the test ten years later. This effect is large enough to loose one’s seat in Burundi’s competition for access to secondary school. We conclude that pupils who are affected by civil war are disadvantaged in Burundi’s education system.

Discussant: Andres Moya

15:30 Break (15 min)

15:45 Marc Rockmore, US Government

Title: The Implications of Aggregate Measures of Exposure to Violence for the Estimated Impacts on Individual Risk Preferences

Authors: Marc Rockmore and Christopher B. Barrett

Abstract: A rapidly expanding literature causally links exposure to violence to changes in a variety of behavioral parameters. The estimated coefficients, however, vary greatly across studies in both magnitude and sign. Using original panel data and disaggregated measures of exposure to plausibly exogenous violence in northern Uganda, we investigate the effect of aggregating exposure to violence at the individual and geographical levels. We demonstrate that exposure to violence affects individual risk preferences in strikingly heterogeneous ways depending on the nature of the individual’s exposure. Estimates that rely on individual measures that aggregate across types of violence exposure or on geographic measures that aggregate across individuals therefore reflect mixture distributions that depend on the underlying distributions of exposure to violence. Simple sampling differences can thereby generate the sort of variability of estimated effects that has been reported in the literature to date.

Discussant: Neslihan Uler

16:30 Andres Moya, Universidad de Los Andes

Title:Psychological Trauma, Socioemotional Skills, and Labor Market Outcomes:Evidence from Internally Displaced Persons in Colombia

Authors: Andres Moya, Suzanne Duryea, Carolina Gonzalez-Velosa

Abstract: In this paper we explore the association between psychological trauma and socio-emotional skills, performance in a job-training program, and labor market outcomes for IDPs in Colombia. We focus on young IDPs enrolled in Transfórmate (Transform Yourself), a job-training program that is part of a series of governmental interventions that stem from the Victims Law (Ley de Victimas, 1448 of 2011) and that seek to provide reparations for IDPs and victims of the armed conflict in Colombia. Our empirical strategy exploits the severity in symptoms of posttraumatic stress disorder to understand how it correlates with (1) socio-emotional skills; (2) academic performance; and (3) labor market outcomes. Our results reveal strong and statistically significant negative patterns of association between psychological trauma and socio-emotional skills, performance in the program, and labor market outcomes.

Discussant: Marc Rockmore