Horizontal inequalities

(This 15-minute presentation accompanying the reading pack was recorded at the GSDRC Seminar on Conflict and Development, London, 23 March 2015.)

Civil wars are one of the main sources of state fragility, low incomes and poor human development. Hence, policies to prevent conflict must be a high priority for those concerned with promoting development in poor countries. There is increasing evidence that the presence of horizontal inequalities (HIs), or inequalities among groups, raises the risk of conflict, and this is the central issue covered in these readings.

There has long been controversy as to the role of inequality in causing conflict with some econometric studies finding little connection (Fearon & Laitin 2003; Collier & Hoeffler 2004). Yet such studies only look at vertical inequality or inequality among individuals or households in a society. In contrast, horizontal inequalities have been shown to be associated with conflict (Cederman, Weidmann, & Gleditsch 2011). Horizontal inequalities occur along a number of dimensions, including:

  • economic dimensions, where it is not just income, but land ownership and employment, among other aspects, that are relevant to people’s wellbeing and grievances;
  • social dimensions, such as access to health and education;
  • political dimensions, encompassing participation and control in central and local government, the bureaucracy and the army, as well as other sources of power; and
  • cultural dimensions, including societal respect for a group’s religious practices, language, or dress.

In civil wars, people mobilise in groups. While people can be grouped in a variety of ways, for understanding conflict, it is necessary to identity differences which are salient to people and thus may form the basis of mobilisation. The relevant categorisation differs across societies and time. Frequently, however, it is ethnic or religious differences that constitute the unifying banner under which people mobilise. Where there are large inequalities in access to socio-economic resources between major groups, people in low-income groups may be ready to mobilise to improve their position, while those in the richer groups may mobilise to protect their privileges. Such differences can lead to violent conflict, if there are no peaceful ways of securing change. In the presence of socio-economic HIs, violent conflict is especially likely where groups are also excluded politically (i.e. also face political HIs) and consequently leaders of a group are unable to participate in government.

This group of readings explores the issue of the relationship between HIs and conflict. Reading 1 provides an overview of the argument and some evidence. More systematic evidence is contained in Reading 2. Reading 3 provides a West African case study to illustrate how HIs can generate conflict in practice and the type of policies that may prevent them leading to conflict. Reading 4 discusses the range of policies that would reduce HIs, arguing that these should be applied in all cases where there are significant HIs and not only where there has been conflict. Reading 5 considers some implications for aid. And reading 6 shows that the relationship between HIs and conflict applies at global as well as national level, and may partly explain current global conflicts concerning Muslims and non-Muslims.


Reading 1: Stewart, F. (2000). Crisis prevention: Tackling horizontal inequalities. Oxford Development Studies, 28(3), 245-262.
This paper analyses the economic and social causes of conflict, drawing conclusions for conflict prevention. There has long been discussion as to whether inequality causes violent conflict. Evidence suggests that vertical inequality (i.e. inequality among individuals in a society) is not associated with conflict. Civil wars normally occur when groups mobilize against each other, on the basis of some cultural characteristic like ethnicity or religion. It is suggested here, with supporting case study evidence, that horizontal inequalities provide the basis for inter-group animosity. Horizontal inequalities are inequalities between groups with a common identity. These inequalities may manifest in political, economic or social dimensions. The paper concludes that policies to limit high horizontal inequalities are needed in all vulnerable countries.

Reading 2: Cederman, L.-E., Weidmann, N. B., & Gleditsch, K. S. (2011). Horizontal inequalities and ethno-nationalist civil war: A global comparison. American Political Science Review, 105(3), 478-495.
Contemporary research on civil war has largely dismissed the role of political and economic grievances, focusing instead on opportunities for conflict. However, these strong claims rest on questionable theoretical and empirical grounds. Whereas scholars have examined primarily the relationship between individual inequality and conflict, here it is argued that horizontal inequalities between politically relevant ethnic groups and states at large can promote ethno-nationalist conflict. Extending the empirical scope to the entire world, this article introduces a new spatial method that combines a newly geocoded data on ethnic groups’ settlement areas with spatial wealth estimates. Based on these methodological advances, the paper finds that, in highly unequal societies, both rich and poor groups fight more often than those groups whose wealth lies closer to the country average. The results remain robust to a number of alternative sample definitions and specifications.

Reading 3: Langer, A. (2008). When do horizontal inequalities lead to conflict? Lessons from a comparative study of Ghana and Côte d’Ivoire. In F. Stewart (Ed.), Horizontal inequalities and conflict: Understanding group violence in multiethnic societies (chapter 8). London: Palgrave.
Available as Working Paper: CRISE WP 82, 2015
This paper reviews the experience of two West African countries in terms of horizontal inequalities and conflict. It shows that in many respects Ghana and Cote d’Ivoire are similar – in economic structure, ethnic composition and horizontal inequalities. Both countries have severe socio-economic horizontal inequalities between the North and the South. Yet Côte d’Ivoire experienced a major North-South civil war from 2002-2007, while Ghana avoided any major national conflict. This paper reviews the reasons for this difference. It provides evidence showing that while the two countries had similar socio-economic inequalities, Ghana was consistently politically inclusive, and consciously respected different cultures and religions. Côte d’Ivoire also followed an inclusive policy until the death of Houphouët-Boigny but subsequently Northerners were excluded politically, and culturally. These political and cultural HIs are argued to be the main factors behind the outbreak of violent conflict.

Reading 4: Stewart, F., Brown, G., & Langer, A. (2008). Policies towards horizontal inequalities. In F. Stewart (Ed.), Horizontal inequalities and conflict: Understanding group violence in multiethnic societies. London: Palgrave.
Available as Working Paper: CRISE WP 42, 2007.
Severe horizontal inequalities are undesirable in themselves and can lead to violent conflict. It is therefore important to reduce them wherever possible. This paper reviews a range of policies that could contribute to reducing HIs in the political, socio-economic and cultural status dimensions. Relevant policies depend on the context and hence a first requirement is a careful assessment of the nature and causes of HIs in the particular society. Among many considerations to be taken into account, two are especially important. First, if possible, policies should be adopted that reduce rather than increase the salience of identities. Hence, the paper considers a range of indirect policies that are likely to reduce HIs because they are designed to help groups in which deprived groups are numerous, rather than direct policies targeted at the groups themselves. Secondly, policies that correct HIs can be provocative, leading to mobilisation (sometimes violent) by previously privileged groups, so caution is needed in design and implementation. Examples of success in introducing such policies and sustaining peace – such as in Malaysia and Northern Ireland – show that policies can be effective without provoking a violent reaction. The successful socio-economic cases are shown to have tackled both social and economic inequalities, while success in reducing political HIs requires political inclusivity at many levels of the political system. Despite the importance of addressing HIs where they are large, the paper concludes that considerations of HI are frequently ignored in policy-making, including in most of the policies advocated by the IMF and World Bank, and in poverty reduction strategies.

Reading 5: Brown, G., & Stewart, F. (2006). The implications of horizontal inequality for aid (CRISE working paper no. 36). Oxford: Centre for Research on Inequality, Human Security and Ethnicity.
This paper argues that the reduction of horizontal inequalities should inform aid policy in heterogeneous countries with severe HIs. It shows how this would change aid allocation across countries, leading to more aid to heterogeneous countries relative to homogeneous ones, the opposite of the existing bias in aid distribution. It explores how adopting an HI approach would affect the use of particular aid instruments, arguing that different instruments are appropriate according to the attitude and capacity of the government in relation to correcting HIs. Drawing on case studies of Ghana and Nepal, it argues that at present there is neglect of HI considerations in aid policy, which can be particularly damaging where aid forms a large part of government resources.

Reading 6: Stewart, F. (2009). A global view of horizontal inequalities: Inequalities experienced by Muslims worldwide (MICROCON Research Working Paper 13). Brighton: MICROCON.
Both within and across countries, most attention has been devoted to measuring inequality among individuals (and globally, among countries). Within countries, increasing evidence shows that inequalities among groups (HIs) are important for wellbeing. However, the global component of HIs is generally neglected. The paper argues that HIs at a global level may also be important for world stability and wellbeing, in much the same way HIs are relevant at the national level. With this perspective, the paper reviews Muslim/non-Muslim HIs within developed and developing countries, and between Muslim and non-Muslim countries, finding that Muslims are systematically disadvantaged across many dimensions. It argues that, despite much heterogeneity among the Muslim population, there is evidence of multiple global connections and of shared perceptions, such that inequalities faced by Muslims in one part of the world may become a source of grievance and potential mobilisation in other parts of the world. Consequently, socioeconomic and political inequalities need to be addressed globally, within countries and between them, and politically as well as with respect to socioeconomic and cultural status.

Questions to guide readings

  • Why might one expect Horizontal Inequalities to be associated with conflict?
  • What bearing, if any, does the greed/grievance debate have on the question?
  • How robust is the evidence supporting the hypothesis that larger HIs are associated with a greater risk of violent conflict?
  • Does the HI approach point to policies that differ from the growth and poverty reduction policies that are generally advocated?
  • Does the approach have any relevance to aid?
  • How would one go about identifying the relevant HIs in a particular country? Consider current conflicts, for example in Ukraine, the Middle East or Nigeria.
  • Does it make sense to apply the HI approach globally?

Send feedback on this reading pack