VoxEU Column Frontiers of economic research Global governance Politics and economics

Ethnic favouritism: Not just an African phenomenon

Ethnic favouritism is widely regarded as an African phenomenon, or at most a problem of poor and weakly institutionalised countries. This column uses data on night-time light intensity to challenge these preconceptions. Ethnic favouritism is found to be as prevalent outside of Africa as it is within, and not restricted to poor or autocratic nations either. Rather, re-election concerns appear to be an important driver of the practice.

It is generally accepted that sub-Saharan African politics operates on the basis of ethnic favouritism, whereby political leaders target preferential policies, and substantial resources, towards their ethnic homelands. Kenya is a prominent example. Both the Kalenjin-dominated government of Daniel arap Moi and the Kikuyu-dominated government of Mwai Kibaki engaged in patronage, corruption, and ethnic favouritism during their 35 years in office ending in 2013 (Wrong 2009).

However, while an abundance of studies has documented the link between ethnic favouritism and African politics, few have focused on more than one country and policy area (notable exceptions are Franck and Rainer 2012, and Kramon and Posner 2013), and none has investigated the prevalence and determinants of ethnic favouritism globally.

We employ a novel approach, drawing on two large and diverse samples of multi-ethnic countries from around the world, and a broad output measure (night-time light intensity) that captures the distributional effect of a wide range of policies, to examine the extent and drivers of ethnic favouritism at the global level (De Luca et al. 2016).

Measuring ethnic favouritism globally

To explore whether incumbent political leaders worldwide favour their ethnic homelands, we utilise data from many ethnographic regions (areas that share the same ethnic composition) in around 140 multi-ethnic countries during the period 1992 to 2013.

For our ethnographic data, we construct two samples based on the two most prominent ethnographic maps available. The first is from the World Language Mapping System and maps the languages described in the Ethnologue (Gordon 2005), a comprehensive list of the world's known living languages. The second is from the Geo-Referencing of Ethnic Groups (GREG) project (Weidmann et al 2010), which has digitalised the classical Soviet Atlas Narodov Mira. Our Ethnologue-based sample includes 7,653 ethnographic regions from 141 multi-ethnic countries (on average 54 regions per country) and our GREG-based sample comprises 2,032 ethnographic regions from 137 multi-ethnic countries (on average 15 regions per country).

Since GDP data are unavailable for ethnographic regions, we follow Henderson et al. (2012) and use data on night-time light intensity based on recordings by US Air Force weather satellites. Due to data availability at the local level and its positive association with GDP, night-time light intensity has become a widely used measure of economic activity and development in studies looking at subnational administrative and ethnographic regions (e.g. Michalopoulos and Papaioannou 2013, 2014, Hodler and Raschky 2014a, 2014b, Alesina et al 2016).

We investigate whether an ethnographic region has more intense night-time light if it is the current national political leader’s ethnic homeland than it would have if the political leader belonged to some other ethnic group. We interpret a positive association between being the political leader's ethnic homeland and night-time light intensity as evidence of ethnic favouritism. In addition, we study the phenomenon’s scope and determinants, and the possible motivations of the political leaders who engage in it. 

Ethnic favouritism is a global phenomenon

We find robust evidence for ethnic favouritism – ethnographic regions that are the current political leader’s ethnic homeland enjoy 7%-10% more intense night-time light, corresponding to 2%-3% higher regional GDP. Furthermore, we show that ethnic favouritism extends to ethnic groups that are linguistically close to the political leader.

Most significantly, these effects are as strong outside of Africa as they are within, challenging the preconception that ethnic favouritism is mainly or even entirely a sub-Saharan African phenomenon. For example, Bolivian presidents tended to favour areas populated by European descendants and Criollos, largely at the expense of the indigenous population. After the election of Evo Morales, a member of the indigenous Ayamara ethnic group, luminosity in indigenous areas grew substantially. Notably, critics suggest Morales gave special attention to the interests and values of the Ayamara at the expense of other indigenous peoples (e.g. Albro 2010, Postero 2010).

Democratisation is not a panacea

Our results further suggest that, while democratic institutions have a weak tendency to reduce ethnic favouritism, their effect is limited. In particular, a change from autocratic regimes to weak democracies does not seem to reduce ethnic favouritism (and may even increase it).

This result could in part be explained by political leaders’ motivations for engaging in ethnic favouritism. We find that the practice intensifies around election years in which the political leader's office is contested, suggesting that leaders may target policies towards their ethnic homelands to improve their re-election prospects, and not solely out of co-ethnic altruism. To the extent that political leaders engage in ethnic favouritism for electoral purposes, democratisation is not likely to be effective in curbing the practice.

Political leaders’ electoral concerns may also explain our finding that the benefits of ethnic favouritism are only temporary, and therefore do not contribute to sustainable development. We show that night-time light intensity in the ethnic homelands of former political leaders who have been replaced by political leaders from other ethnic groups returns to normal levels within two years of ethnic transition. We suggest this could be due to political leaders directing public funds to their ethnic homelands for consumption purposes, rather than investments in infrastructure, pursuant to a patronage logic whereby co-ethnics are considered more likely to maintain their political support when the benefits of that support depend on the leader’s continued power (Padró i Miquel 2007).


The findings of our study – the first of its kind – challenge the conventional view that ethnic favouritism is a primarily African phenomenon that can be curbed through economic development and democratisation. We show that ethnic favouritism is a global axiom of politics that exists in both rich and poor countries, and that the constraining effects of sound political institutions are limited.

These results highlight the need to explore the ways in which political mechanisms can reduce ethnic favouritism or, even better, induce cooperation amongst ethnic groups. After all, Switzerland is a highly ethnically-segregated country, but we find no evidence of ethnic favouritism there thanks to its inclusive form of government and a presidency that rotates between different political parties and ethnic groups.


Albro, R (2010) “Confounding cultural citizenship and constitutional reform in Bolivia," Latin American Perspectives, 37: 71-90.

Alesina, A, S Michalopoulos and E Papaioannou (2016) “Ethnic inequality," Journal of Political Economy, 124: 428-488.

Franck, R and I Rainer (2012) “Does the leader's ethnicity matter? Ethnic favouritism, education and health in Sub-Saharan Africa," American Political Science Review, 106: 294-325.

De Luca, G, R Hodler, P A Raschky and M Valsecchi (2016) “Ethnic favouritism: An axiom of politics?”, CEPR Discussion Paper 11351.

Gordon, Jr, R G, (2005) Ethnologue: Languages of the World, Dallas, TX: SIL International.

Henderson, V J, A Storeygard and D N Weil (2012) “Measuring economic growth from outer space", American Economic Review, 102: 994-1028.

Hodler, R and P A Raschky (2014a) “Regional favouritism”, Quarterly Journal of Economics, 129: 995-1033.

Hodler, R and P A Raschky (2014b) “Economic shocks and civil conflict at the regional level", Economics Letters, 124: 530-533.

Kramon, E and D N Posner (2013) “Who benefits from distributive politics? How the outcome one studies affects the answer one gets", Perspectives on Politics, 11 : 461-474.

Michalopoulos, S and E Papaioannou (2013) “Pre-Colonial ethnic institutions and contemporary African development", Econometrica, 81: 113-152.

Michalopoulos, S and E Papaioannou (2014) “National institutions and subnational development in Africa", Quarterly Journal of Economics, 129: 151-213.

Padró i Miquel, G (2007) “The control of politicians in divided societies: The politics of fear", Review of Economic Studies, 74: 1259-1274.

Postero, N (2010) “Morales's MAS government: Building indigenous popular hegemony in Bolivia", Latin American Perspectives, 37: 18-34.

Weidmann, N B, J Ketil Rod and L-E Cedermanv (2010) “Representing ethnic groups in space: A new dataset", Journal of Peace Research, 47: 1-9.

Wrong, M (2009) It's our turn to eat: The story of a Kenyan whistleblower, London: Fourth Estate.

3,150 Reads