Ethnic Voting in Latin America

Prof. Dr. Detlef Nolte
2015 - 2019
German Research Foundation (DFG)

Dr. Fernando Mayorga, Universidad Mayor de San Simón, Bolivia

Dr. Eduardo Rodriguez Veltze, Universidad Católica, Bolivia

Dr. Simón Pachano, FLACSO, Ecuador



Research Questions

What are the levels of ethnic voting among different ethnic groups in different Latin American countries across time? How can variance in ethnic voting across groups, countries, and time be explained?

Contribution to International Research

The project contributes to research on interethnic relations, electoral behaviour, and party-voter linkages. Research on ethnic voting in Latin America lacks comparative historical analyses and focuses primarily on indigenous groups, ignoring Afro-Americans. This research project analyses variance in ethnic voting across time, Latin-American countries and ethnic groups. Its aim is to generate insights on the dynamics of ethnic voting and in so doing to contribute to a deeper understanding of interethnic relations, electoral competition, and democratic representation n Latin America and beyond.

Research Design and Methods

The aim of the project is to further develop theories on ethnic voting and test them empirically on the basis of a longitudinal comparative study over an extended period of time (25-30 years) of free and (relatively) fair elections. The three central theories along which ethnic voting is tested are identity based, patronage and spatial models of electoral competition. The data employed to measure ethnic voting consists of census data on ethnic group size according to self-categorisation, official election results and survey data. It makes use of a large variety of methods including quantitative, comparative, and qualitative methods.

Preliminary Results

The comparative analyses point to important variance in ethnic voting across time and countries. It shows the highest levels of ethnic voting for Bolivia, Ecuador, Guyana, and Suriname. Medium levels of ethnic voting are observed in Guatemala and Panama. Very low levels of ethnic voting can especially be found for Costa Rica and Brazil. On average one can observe an increase in ethnic voting over time. This increase is most evident for Bolivia, Guyana, and Suriname. Also in Chile an increase in ethnic voting seems to be a systematic phenomenon.

While comparative analysis shows that Bolivia is currently one of the countries with the highest level of ethnic voting, this was not always the case. In earlier periods ethnic voting in Bolivia has been weaker. Hence, since we observe important dynamics in Bolivia we have investigated this case in more detail. More specifically, we have particularly focused on differences in the electoral preferences of indigenous highland and lowland peoples, of which the latter have until now been treated as a homogenous unit. We have found that the rise in ethnic voting in Bolivia started with the first party with a pro-indigenous agenda, Conciencia de Patria (CONDEPA), and increased with the rise of the Movement for Socialism (MAS) led by Evo Morales. A paper entitled “We were forgotten. Explaining ethnic voting in Bolivia’s highlands and lowlands” has recently been published in the Journal of Ethnic and Migration Studies.

In contrast to Bolivia we find very low levels of ethnic voting in Brazil. This is surprising regarding the fact that Afro-Brazilians are heavily underrepresented among the members of parliament. In our analysis of candidates’ electoral success we find that this underrepresentation is not due to electoral discrimination of the Afro-Brazilian candidates but their lower socio-economic status reflecting ethnic inequalities in society as a whole. Currently we are investigating how ethnic inequalities at the societal level translate into socio-economic inequalities between European-descendent and Afro-Brazilian candidates.

Finally, it has been criticized that the study of ethnic voting underestimates the fluidity of ethnic identities. Hence, in a final step we are investigating the stability of ethnic identities in the region. More specifically, we analyse changes in ethnic voting in Ecuador between 2001 and 2010 with a panel of around 350'000 Ecuadorians. The analyses show that ethnic identities remain mostly stable and that changes in ethnic identities at the individual level are better explained with acculturation and internal migration than with social mobility.