Measuring Ethnic Fractionalization in Africa

Abstract

In most studies of the impact of ethnic diversity on economic growth, diversity is hypothesized to affect growth throughI n 1997, William Easterly and Ross Levine published an article titled "Africa's Growth Tragedy: Policies and Ethnic Divisions" (Easterly and Levine 1997), which found a statistically and economically important negative effect of ethnic diversity on economic growth in a cross-section of countries. Specifically, Easterly and Levine (hereafter E&L) found that moving from an ethnically homogeneous country to one with a diversity of ethnic communities corresponded with a decrease in annual economic growth rates of more than 2 percent. They then applied this finding to Africa, reasoning that, because African countries are typically ethnically diverse, the strong link between ethnic heterogeneity and slow growth was quite likely an important part of the explanation for that region's "growth tragedy." E&L's findings have been broadly accepted. Thanks largely to their article, it is now de rigueur for economists Daniel N. Posner is Assistant Professor of Political Science, University of California, Los Angeles, Los Angeles, CA 90095-1472 (dposner@polisci.ucla.edu).I thank Kanchan Chandra, James Fearon, Asim Khwaja, David Laitin, Drew Linzer, Jean-Laurent Rosenthal, Steven Wilkinson, and two anonymous reviewers for their comments, as well as members of the Laboratory in Comparative Ethnic Processes (LiCEP) and participants at seminars at the World Bank, the Claremont Graduate School, and the UCLA Von Gremp Workshop in Economic History. Drew Linzer, Johanna Birnir, Robert Dowd, Bernadeta Killian, Elin Skaar, Lahra Smith, and Susanna Wing provided valuable research assistance. All errors are my own.

Ethnic Politics and Women's Empowerment in Africa: Ministerial Appointments to Executive Cabinets

“…Using two different measures of ethnic political mobilization-a fractionalization index of politically relevant ethnic groups (Posner 2004) and a count of the number of politically relevant ethnic groups (Cederman, Wimmer, and Min 2010)-we show that women are systematically appointed to a smaller share of cabinet positions where incumbents must contend with distributive demands from a larger number of such groups. This finding is robust to the inclusion of a variety of controls as well as to different estimation techniques.…”

“…The first measure is Posner's (2004) fractionalization index of politically relevant ethnic groups (PREG). The PREG measure is calculated as an ethnic fractionalization index, indicating the likelihood that two randomly selected individuals from a country will be from different politically relevant ethnic groups.…”

The Causal History of Africa: A Response to Hopkins

“…More importantly, it predicts the under-provision of public goods and the adoption of bad policies. Posner (2004a) shows that this effect becomes even stronger when the measure is restricted to politically relevant groups. This holds even within countries; schools in the more ethnically diverse parts of western Kenya receive less funding and have worse facilities (Miguel and Gugerty, 2005).…”

“…Hopkins (2009) charges that Easterly and Levine (1997) treat their data on fractionalization as "unproblematic," though he does note how others (e.g. Alesina et al (2003), Fearon (2003), andPosner (2004a)) have improved on their measure. He suggests that an understanding of the origins of fractionalization is lacking.…”

When Do Votes Count?

“…Recent work has shown that common proxies for cleavages suffer from serious problems (Posner, 2004). Unlike these indices, their measures capture both geographic concentration and fragmentation of ethnopolitical groups within a country.…”