LEEM Seminar - Elisa Cavatorta (King’s College London)

"Does exposure to violence affect reciprocity? Experimental evidence from the West Bank"
Quand ? Le 24-10-2018,
de 12:00 à 14:00
Où ? Salle SCI - ISEM
S'adresser à
Ajouter un événement au calendrier vCal

Séminaire LEEM

Elisa Cavatorta (King’s College London)

"Does exposure to violence affect reciprocity? Experimental evidence from the West Bank"

Authors: Elisa Cavatorta (King’s College London)
Yousef Daoud (Doha Institute of Graduate Studies), Daniel John Zizzo (Newcastle University and BENC)

Abstract: Violent conflict is often a dynamic situation where the onset of conflict depends crucially on the response to the first attack. If the tendency of a victim of violence is to respond to aggression with aggression, it is possible that conflict begets more conflict. On the other hand, exposure to violence may detract other people from engaging into violence. The extent to which people affected by violence reciprocate other people’s actions is a matter of great interest for peace and conflict resolution, but underappreciated.

This paper is concerned with how reciprocity is affected by exposure to violence in early age. Using an experimental methodology and lab-in-the-field experiments, we study how conditional cooperation (positive reciprocity) and conditional aggression (negative reciprocity) in adolescents vary as a result of exposure to violence. We focus on young Palestinians in the West Bank region of the Palestinian Territories.

To elicit conditional cooperation, we implement a “voluntary contribution mechanism” game eliciting unconditional and conditional contributions (Fischbacher et al, 2001). To elicit negative reciprocity, we implement a decision task in which retaliatory behavior can, but need not, occur. This is a simplified form of the vendetta game developed by Bolle et al. (2014).

One challenge in the analysis of the effect of violence on these attitudes is that violence may be more likely to occur in places where people are less cooperative already. To solve this issue, we employ a research design based on the location of Israeli settlements pre-dating the first episode of widespread violence against Palestinians (the First Intifada). Within the design, some respondents have the obligation to regularly cross military checkpoints (and are more exposed to violence) and some do not have to (and are less exposed to violence). This strategy allows us to create exogenous variation in the exposure to violence.

Our sample includes 1,172 students in 23 pairs of schools, one in each pair located in an area of high likelihood of violence and one in an area with low likelihood of violence. Our analysis shows that exposure to violence (proxied by the obligation to cross a checkpoint) does not affect unconditional behavior but does influence conditional behaviour. In particular, we observe that subjects obliged to cross checkpoints are more “reciprocators” in both domains: they tend to cooperate more when others do the same, and retaliate more when other engage in “aggressive” behavior.

One of the mechanisms through which violence may impact the degree of reciprocity is through an effect on beliefs about what other people would do in general. To test the importance of this mechanism, we introduce an additional treatment: we elicited the subject’s expectations about other participants’ behavior before their decisions in the games. As expected, we observe that behavior is positively associated with the expectations on the same behavior by others. The initial results support the existence of this channel of impact.