Ethnically heterogeneous societies are widely held to be more difficult to govern than homogeneous ones. Theoretical arguments often attribute the problems to the fact that ethnic divisions result in divergent group preferences. However, the core measure in empirical research on the problems with heterogeneous societies is ethno-linguistic fractionalization (ELF), which contains no information about the likelihood of particular groups holding divergent preferences. Since the ELF measure assumes that substantive differences between all groups are the same, its use requires the implicit assumption that the potential for group-based conflict within a society is simply a function of how likely an individual is to interact with someone from a different group. Yet we know that individuals from different ethnic groups often have similar preferences.
For a large number of issues that arise in politics, group economic differences will be at the core of inter-group disputes. Income differences across groups can lead to different needs with respect to public goods, feelings of alienation or discrimination, different attitudes toward redistribution, and different "class" identities. The central objective of this research project is therefore to draw on household income and budget surveys to create a new cross-national dataset that measures economic differences across groups.
The data will be useful for exploring how ethnic, linguistic, and regional economic differences are related to outcomes that have been central to the literature on ethnic heterogeneity. These include economic growth, public goods provision, redistribution and conflict. By incorporating information about economic differences into measures of ethnic cleavages, it will be possible to explore more carefully the mechanisms that have been proposed to link ethnicity with poor outcomes. The data will also open a range of new research opportunities.
The primary benefit of the study will be to enhance the infrastructure for research and education by creating and disseminating a cross-national data set that measures inter-group economic differences. The data set, which will be made up of 80-100 countries, will include a wide range of measures of inter-group economic differences. It will also include standard ethnic fractionalization and polarization scores.
