Fjelde, H. and K. Höglund (2016): "Electoral Institutions and Electoral Violence in Sub-Saharan Africa". British Journal of Political Science 46(02): 297-320. DOI:    


An article reviewed on this blog earlier this year came up with a weighty question: Does the type of electoral system shape electoral quality? The authors concluded – it does. Across 170 countries, Proportional Representation (PR) systems display less controversies and accusations of fraud than plurality systems. But this might only be the tip of the iceberg. In contexts where democratic principles are less established, the effects of majoritarian electoral rules might lift fraud to another level. Where political institutions are too weak to channel competition in a peaceful manner, electoral tactics might entail a physical dimension: the use of violence or coercive intimidation to influence the electoral outcome.


In their latest article, Fjelde and Höglund delve into this dimension of electoral systems’ effects. An analysis of elections in 47 Sub-Saharan African countries from 1990 to 2010 reveals that majoritarian electoral rules are more likely to induce electoral violence than PR rules. How to explain these findings? According to the authors, violent electoral tactics emerge if formal electoral rules reinforce informal institutional dynamics.Formal electoral rules determine the distribution of power and thus define what is at stake in the electoral contest. While PR systems disperse political power across a broader range of groups, the stakes are higher if majoritarian electoral rules are applied: Large parts of the political spectrum are excluded from power; the winner takes it all. In well-established democracies, this works. The loser waits for its time to come, for the next election. Actors comply with democratic rules because solid democratic institutions convey a certainty that losers will still be allowed to advance their interests in the future.


Yet, what happens in contexts of under-institutionalization? What if formal institutions are paralleled by powerful informal institutions? Fjelde and Höglund argue that pervasive patron-client relationships (particularly along ethnic and/ or geographic lines) increase the cost of electoral defeat. A great deal of Africa’s emerging democracies is plagued by these structures, where political power translates into economic resources – and economic resources secure political power.Thus, the stakes, already high under majoritarian rules, are raised even higher. Elections are perceived as zero-sum games. Hence, the fear of electoral defeat increases the incentives for both government and opposition to resort to violent electoral tactics. The prevalence of electoral violence depends on two factors that shape the structure of these informal institutions: First, on the size of the largest excluded group. And second, on what is at stake economically. Findings of this paper prove: The larger the excluded ethnical groups are and the narrower wealth is distributed the higher the risk of electoral violence.


Fjelde and Höglund’s theoretical argument is convincing: Majoritarian electoral systems are more likely to suffer from electoral violence, particularly if the excluded groups are large and the economic stakes high. The data analysed in this respect support their hypotheses – the electoral system matters; economic situation and ethnic structure influence electoral violence in the predicted direction. Yet, a different operationalization of two key variables might advance our knowledge even further: For one, Fjelde and Höglund’s dependent variable, electoral violence, does not capture the time dimension – and, thus, does not allow for a differentiation between pre- and post-election violence. This, however, could be an interesting path for future research as incentives to use violence might change over the electoral cycle. For another, one could reconsider the assumption that the size of the largest excluded ethnical group has a linear effect on electoral violence. While the authors suggest that the risk of violent tactics increases with group size, a more nuanced, U-shaped relationship is conceivable: strong excluded groups might abstain from violence as they have a viable option to win the election in a non-violent manner and weak parties might do so because they are simply not able to exert violence to a meaningful degree – but what about those in-between? Might they be more likely to choose the violent path to power?


Author: Lena Ramstetter, University of Salzburg, in November 2016

Kommentare: 0