From Social Disorder to Civil Wars: The Far Reaching Effects of Segregation

MIPP researcher demonstrates the impact of segregation on all levels of national conflict.

By Henrieta Isufllari

 Segregation in its broadest terms refers to the forceful geographical separation of a particular group of people based, among others, on their ethnic, religious, or language background. The most known types are racial and ethnic segregations, with numerous examples drawn from the infamous segregation of the African American population in the United States during the Jim Crow laws, or the Apartheid regime in South Africa. 

Although a present-day legal platform that legitimizes segregation may no longer exist, different forms of segregation are currently observed in all corners of the world. Its de facto presence can still be seen in residential segregation (housing patterns), income segregation, or education segregation – a type of segregation which is also very predominant in Chile and has been the subject of a few recent studies both in the social and economic contest.

How does segregation affect conflict? The latest study1 by MIPP researcher Alejandro Corvalán and his colleague, Miguel Vargas, provides an empirical analysis of the link between segregation and conflict, by studying the different mechanisms that affect the motivation for conflict when group concentrations are prevalent. The focus of the study is on three dimensions of segregation: ethnic, language, and religious. “The relation between segregation and conflict is far from being clear, and in fact some political scientists have even recommended segregating communities to avoid conflicts between them,” states Corvalán.


Researchers in the past have continuously shown that the fractionalization or polarization among groups with different racial or ethnic compositions has a positive impact on conflict. But as is often the case with conflict, although the reasons that may cause divergence are clear, the factors that fuel its escalation to protests and even civil wars are not easily identified. Furthermore the magnitude of the impact of segregation is hard to pinpoint.

Corvalán and Vargas confirm the effect of segregation on the escalation of conflict by analyzing two channels directly influenced by segregation: trust levels between groups of different backgrounds, and secession threats arising from a need to fight for autonomy that is often dominant in highly segregated communities. The study shows that both channels have an impact on the effect of geographic group influence on conflict. The lack of trust of the outside groups, which is present in segregated groups, causes an increase in animosity, while the clear separation of territory gives the separated groups a sense of belonging, increasing the chances of claims for autonomy. 

The University of Diego Portales professors use a recent and multidimensional dataset on national segregation to study the impact of segregation on all of the intensities of conflict in 165 different countries. They classify the intensity of conflict in low, medium and high (civil war equivalent), based on the amount of deaths per conflict as reported on the PRIO Dataset - up to 25, between 25 to 1000, and more than 1000 deaths per year, respectively. For the period from 1960 to 2000, the dataset identifies 89 countries with low conflict, 55 with intermediate levels and 42 countries with high levels of conflict.

A cross-country measure of segregation is then used, with full segregation occurring when every ethnic group lives in separate regions within a country; while no segregation means that every region has the same ethnic group composition as the country as a whole. Based on this classification, the five least ethnically segregated countries are Germany, Sweden, Netherland, South Korea, and Japan. On the other end of the spectrum stands Zimbabwe as the most segregated country, followed by Guatemala, Afghanistan, Uganda, and Turkey.

To account for the fact that sometimes segregation can be more of the effect of a conflict rather than one of its causes, the researchers identify that this might be true in the event of civil wars, but also in the cases of low intensity conflict. “These results for civil disorder and protest,” adds Corvalán, “suggest that the causality is from segregation to conflict and not the other way round.”

The study concludes that both ethnic and language segregation have a direct effect on the incidence of conflict, and this is true for all conflict levels, while religious segregation is not linked with any type of conflict. Neither type of segregation serves as a catalyst for the outbreak of violence; their impact is only limited to the extent of reinforcing and escalating existing conflicts. The evidence across countries shows that the biggest impact of segregation on the escalation of violence stems from a decrease in trust or an increase in secession claims, both consequences of geographic group concentration.  


1"Segregation and conflict: An empirical analysis" by Alejandro Corvalán and Miguel Vargas, Journal of Development Economics (2015)

Recent Comments

Top Commenters