Why do ethnic conflicts in some parts of the world flare up so easily and spread so fast? Is ethnic hate and intolerance contagious? Researchers from the Czech Republic and Slovakia joined forces to try to find the answers to some of those questions and arrived at some surprising conclusions. I spoke to Associate Professor Michal Bauer, an expert on experimental and behavioral economics at CERGE-Economics Institute, who is one of the authors of the study,and began by asking him what motivated the research in this field.
“The original motivation for thinking about this question is the observation that a lot of real-life conflicts often arise out of nothing. Different groups live next to each other for a long time and then something snaps, something occurs and inter-community hostilities spread. So what we tried to do was that in a controlled experimental environment we tried to measure how easy it is to change individuals’ hostile or destructive behaviour towards an ethic minority.”
How did you do that? Who did you work with and what kind of experiment was it?
“The experiment took place in Eastern Slovakia, in around 15 localities, typically in villages or smaller towns in which the Roma minority and Slovak majority live together, and the participants were adolescents from local schools.”
And how did you measure the degree of ethnic hostility?
“We used modern techniques of experimental economics. If one wanted to measure hostile behavior or attitudes by asking questions in a survey one would very likely not get a truthful answer, so we used incentivized tasks in which the participants make choices which impact their own rewards and the rewards of somebody else. We used a game in which a contestant receives rewards or money and he or she is asked to make a decision whether to sacrifice one’s own reward in order to harm the other person even more. When people pay to harm someone else, we can measure individual willingness to harm somebody, and in this case destroying someone else’s resources is really nasty, because I need to pay in order to harm somebody.”
And teenagers influenced each other in making these decisions?
“That’s right. What we tried to look at were essentially two things. We tried to measure the extent of discrimination towards the Roma as compared to an anonymous person from the Slovak group by providing information about who is the experimental counterpart, who is the person that will get harmed by the decision. And by looking at the differences between these different types of partners we can measure discrimination. Now to get to the main question about how one can identify the influence of peers, the participants made choices in groups of three individuals and they made choices sequentially, so some of the subjects were making decisions after observing that somebody from their own group was behaving non-destructively, while others were making decisions after observing that somebody from their own group was being hostile to the other person. And the question was to what extent the behavior of my peers was going to affect my own behavior. That’s the crucial measure we used to identify the effect of spreading of hostile behavior – i.e. to what extent I’m going to copy, or conform to what my classmates are doing.”
Are individuals so vulnerable then, to peer-influence?
“Yes, it turns out that peers matter a lot. For instance, the prevalence of hostile behavior, if I see that my class mate is non-aggressive is around 17 percent, while the prevalence of destruction increases to more than 70 percent, if I am acting after observing a destructive classmate. So peers have a very strong influence, especially if the decision impacts a member of the Roma minority, then the extent of conformism is more than twice as large.”
Is this something deeply inherent? Is this how bullying starts?
“That’s a very good question and it is hard to say at this moment from just this particular study. We would need more data about how generalizable this is. There is some indication from a follow-up study we conducted in the same setting that looks at the social appropriateness of harming somebody, essentially how people perceive social norms and what we see is that the social norm regulating hostile behavior towards my own group is relatively stable. People pretty much always –independently of what their classmates are doing – consider destructive behavior as an act that is not socially desirable, while this perception is more fragile when it comes to hostile behavior towards the Roma.”
So, if I understand you correctly, this is translatable –or applicable –to conflicts in other countries as well?
“I think it is still an open question since the data that we have are from one study, at the same time there are indications from other studies that would point in this direction. Let me give you one example. If we look at the situation in the UK after the Brexit vote, one interesting observation in crime data is that the prevalence of hate crime increased by around 50 percent after this aggressive campaign against immigrants. This again suggests that it is relatively easy to change social norms of behavior against individuals from different groups.”
You worked with teenagers, but what you are saying now would suggest that the same applies to different age groups as well …
“Yes, we studied teenagers. Again, we do not have hard data to test this, but it would be very interesting to see if this applies to adults as well. From other types of studies it seems that there are some social spillovers; that hostility is easy to spread even among adults.”
“Yes, that assumption is very consistent with what we find –if politicians are aware of the social inflammability of being aggressive towards ethnic outgroups then that may explain why certain types of politicians – in literature they are called “entrepreneurs of hatred” - often choose to have aggressive political campaigns towards migrants or other minority groups.”
So would this research also explain the wave of anti-migrant sentiments in some parts of Europe today?
“I think a lot of people in Europe are worried by politicians playing this card –focusing on the fear of migrants – worried that it might be relatively easy to change people’s attitude towards migrants. I earlier mentioned this example –the increased prevalence of hate crime following the aggressive campaigns in favour of Brexit that focused on migrants. That is consistent with the idea that people’s attitude towards migrants may be easily changed. I think one of the exciting questions for future research would be to look at whether the contagious nature of the hostility that we see towards the Roma –whether the same holds for other groups, for instance migrating groups, maybe whether it interacts with previous experience with those groups, meaning whether the tendency to follow others’ aggressive behavior can be mitigated by previous contact and knowledge of those groups.”
Did you also consider how to counterbalance or stifle this aggression? I saw in one of your articles an emphasis on the strict punishment of hate crimes. If individuals see hate crimes swiftly punished would they not be less inclined to violence themselves?
“One implication of what we find is that essentially the social cost of being hostile to hate crimes against people of a different ethnicity is considerable, because they can cause the quick spread of destruction. This implies that it may be optimal for societies to have hate crime laws that punish more severely ethnically or religiously motivated hate crimes. What we have studied is a stylized experiment, but it implies that the social costs of this type of anti-social behavior are particularly large. Another implication is that essentially early diagnosis appears to be very important –it seems important to focus on the first signs of inter-ethnic tensions because they may easily spread among others and it may be too late to start fighting them at a later stage.”
“Our study essentially offers a methodology how to study the spread of hostilities in a controlled environment. There have been a lot of great studies that looked at whether people discriminate in normal circumstances when they make choices on their own and whether discrimination exists in different societies against different groups – what this study is offering is an experimental method focusing on the question, not just whether discrimination exists in normal circumstances, but also how easy it is to trigger discrimination in a social environment depending on what others are doing from my own group.”
Is this not a strange area of study for economists?
“I think economists are increasingly open to thoughts from other fields and they try to offer some of their tools to systematically test the importance of some of the concepts that have been developed by other social sciences and I think it is a fruitful exchange. But you are certainly right that this is not a standard economic topic –it is a topic at the intersection of fields.”
What about other triggers or roots of ethnic hatred – will you be investigating them as well? Do you plan to take this further?
The results of this study were published in the Proceedings of the National Academy of Sciences of the United States.
“We are thinking about other directions. One potentially interesting question could be related to whether people are inclined to punish collectively – if somebody does something bad to somebody else whether people hold responsible only the given individual or also all those who look similarly as the perpetrator –ie. belong in a given group. Another interesting question relating to triggers could be whether economic conditions influence people’s willingness to act anti-socially, stress factors. A lot of conflicts seem to happen after economic crises or negative income shocks and the question is whether these economic shocks have certain psychological effects, either by creating stress or through some other channels that make people more likely to act anti-socially. I think these are all exciting questions and hopefully we’ll get some answers in the years to come.”