The ‘Ethnic Fractionalization’ Variable in Development Economics — A Systematic Review

This article consists of a systematic review of 73 empirical articles in economics journals using the ‘ethnic fractionalization’ variable in cross‐country analyses with developing countries. This variable was introduced in development economics in 1997 by two World Bank economists, Easterly and Levine. Their publication was followed by many others using the same variable. What is striking is that the variable does not measure fractionalization of ethnic groups but simply ethnic diversity. The negative connotation of such groups falling apart resulted in many hypotheses about a negative relationship of ‘ethnic divisions’ with a variety of economic development outcomes, including economic growth, investment and innovation. This systematic review reveals that less than half of the 73 publications confirm hypotheses about negative effects of ‘ethnic fractionalization’. Most of the selected studies do not discuss relevant theory and they do not include confounding variables about how ethnic groups relate to each other. Although more research is needed regarding Scholars' motivations for the use of the variable, the way in which researchers have used the variable points to normative choices made in terms of hypotheses, data, variable label and models. This calls for more reflection about research ethics in applied economics.