2014•African Development ReviewRequires access

Oil Wealth, Ethno‐Religious‐Linguistic Fractionalization and Civil Wars in Africa: Cross‐Country Evidence

John C. Anyanwu

Open publisher page 31 citations

Abstract

Abstract We empirically examine the effect of oil wealth and ethno‐religious‐linguistic fractionalization on civil war prevalence in Africa, by using three different estimation strategies and alternative measures of societal diversity. We show that oil wealth and the three distributional measures of ethnic fractionalization, religious fractionalization and linguistic fractionalization are significant correlates of civil war in Africa. These effects persist as we use an alternative measure of the prevalence of civil war. Thus, while oil wealthy, ethnically and linguistically fractionalized countries are more likely to experience civil violence, religiously fractionalized ones are less likely to experience significant civil violence in the Continent. We also find that countries with large population size, rough (mountainous) terrain and coup‐prone are at greater risk for civil war than those that are more democratic with high per capita income and economic growth.

About this research paper

What this paper is about

Abstract We empirically examine the effect of oil wealth and ethno‐religious‐linguistic fractionalization on civil war prevalence in Africa, by using three different estimation strategies and alternative measures of societal diversity. We show that oil wealth and the three distributional measures of ethnic fractionalization, religious fractionalization and linguistic fractionalization are significant correlates of civil war in Africa. These effects persist as we use an alternative measure of the prevalence of civil war. Thus, while oil wealthy, ethnically and linguistically fractionalized countries are more likely to experience civil violence, religiously fractionalized ones are less likely to experience significant civil violence in the Continent. We also find that countries with large population size, rough (mountainous) terrain and coup‐prone are at greater risk for civil war than those that are more democratic with high per capita income and economic growth.

Why it matters

OpenAlex reports 31 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Abstract We empirically examine the effect of oil wealth and ethno‐religious‐linguistic fractionalization on civil war prevalence in Africa, by using three different estimation strategies and alternative measures of societal diversity. We show that oil wealth and the three distributional measures of ethnic fractionalization, religious fractionalization and linguistic fractionalization are significant correlates of civil war in Africa. These effects persist as we use an alternative measure of the prevalence of civil war. Thus, while oil wealthy, ethnically and linguistically fractionalized countries are more likely to experience civil violence, religiously fractionalized ones are less likely to experience significant civil violence in the Continent. We also find that countries with large population size, rough (mountainous) terrain and coup‐prone are at greater risk for civil war than those that are more democratic with high per capita income and economic growth.

Key concepts: Fractionalization, Ethnic group, Spanish Civil War, Per capita, Per capita income, Development economics, Democracy, Economics

Related papers

Back to paper searchBrowse research topicsOriginal source
Oil Wealth, Ethno‐Religious‐Linguistic Fractionalization and Civil Wars in Africa: Cross‐Country Evidence — Research Paper | ScholarLens