Document Library

Key Text Do Horizontal Inequalities Matter for Civil Conflict?

Author: G Ostby
Date: 2004
Size: 41 pages (279 KB)

Access full text: available online

Summary

Do nationwide inequalities between social groups increase the likelihood of civil conflict? This paper from the International Peace Research Institute, Oslo, compares levels of economic, social and health-related inequality between ethnic groups with the risk of civil conflict in 33 developing countries. It argues that, while there is no statistically important link between economic or health-related indicators and civil war, there is a strong connection between social inequality and conflict.

Horizontal inequalities (HIs) are defined as systematic disparities between culturally-formed groups – in this case, ethnic groups. Civil conflicts refer to armed disputes between groups with a common identity, usually shared ethnicity. Previous studies, which cast doubt on links between civil conflict and inequality, have focused on vertical (class-based) inequalities and economic differences, rather than incorporating ethnically-based, social and health-related disparities.

Using Demographic and Health Surveys data, economic HI is here measured by disparities in household asset ownership, social HI by variations in educational and occupational opportunities and health-related HI by differences in infant mortality rates and access to children’s health cards. The impact of other factors on conflict was also measured, including regime type, GDP per capita, vertical income inequality, fractionalisation (ethnic diversity) and polarisation (the size and distinctness of a society’s various ethnic groups).

The statistical models produce various findings on the factors contributing to civil conflict:

While economic and health-related HIs are not statistically significant factors, social HIs are strongly linked to the risk of conflict. The result from the social HI model confirms previous case study findings.
This suggests that HIs in educational and occupational opportunity are more significant in promoting ethnic grievances and a marginalised group identity than disparities in household assets. This may be because social HIs are a more direct expression of discrimination than other HIs. However, the economic HI measure – household assets – may not adequately capture ethnic economic differences or the economic priorities of certain ethnic groups.
Ethnic fractionalisation and polarisation and vertical income inequality are not found to have a statistical connection to internal conflict. Regime type and GDP growth seem also to be statistically unrelated to conflict. Surprisingly (since the countries sampled are low- to medium-sized economies), there is a strong negative relationship between the level of development (GDP per capita) and conflict.

Researchers should undertake more investigation into the “inequality-conflict nexus”:

They should study further the impact of both vertical and horizontal inequalities on conflict and incorporate more refined measures of the various inequality indicators.
They should use national surveys as sources. These provide rich and reliable datasets that avoid government bias, record descriptive, rather than evaluative, data and give information on disparities within, as well as between, ethnic groups.
They should investigate the impact of intra-group inequalities on civil conflict. Some scholars claim that intra-group disparities lessen the risk of civil conflict by reducing a group’s cohesiveness and weakening collective identity. Others assert that group leaders who abuse group identity for their own ends are more likely to collaborate opportunistically with elites from other groups.
They should develop a measure of “horizontal polarisation”, an index which would incorporate the distribution of welfare indicators both within and between groups, as well as clusters of strong ethnic identity within a society.

Access full text: available online

Source: Ostby, G., 2004, ‘Do Horizontal Inequalities Matter for Civil Conflict?’, International Peace Research Institute, Oslo
Author: International Peace Research Institute, http://www.prio.no