The choice of an electoral system in a culturally plural society affects the potential for future violent conflict. However, no single electoral system is the answer. This paper, prepared for the National Research Council’s Committee on International Conflict Resolution, challenges the assumption that a system that is appropriate for the West can be transplanted into the developing world: Transitional democracies have a greater need for inclusiveness and a lower threshold for adversarial politics, while it is appropriate for consolidated democracies to strive for a minimal winning coalition.
There are several factors that should be taken into account when considering the appropriateness of an electoral system in a divided society: The degree that ethnicity is politicised in the society; the intensity of the conflict; and the demographic distribution of groups. Four systems are identified as being appropriate: (1) Consociationalism, a power-sharing arrangement within government, brokered by autonomous, clearly defined segments of society with minority veto powers, and using proportional representation (PR); (2) centripetalism, which uses the single transferable vote (STV) or alternative vote (AV) to encourage vote-pooling and preference-swapping so that politicians are dependant on groups other than their own; (3) integrative consensualism, which uses both consociationalism and centripetalism to encourage both cross-cutting cleavages and a fair representation of minorities in decision making; and (4) explicitism, where community groups are formally recognised through a variety of devices (such as communal electoral rolls, reserved seats, ethnically mixed or mandated seats).
Yet, how much can or should political outcomes be engineered through institutional structures? It is rare for electoral systems to be deliberately chosen or designed for the conditions of a country, and the choice made is often bad for its long-term political health. Findings were that:
- Consociationalism assumes that leaders will be moderate and are willing to compromise (this is not always the case), and that there is a clear definition of groups. There are also few incentives for people to go beyond their own ethnic base
- Consociationalism can be effective for democracies in transition, but not as effective for consolidating democracies. If things are good it may actually increase tension. It is best as a ‘stop-gap measure’ for countries with little hope
- Centripetalism works best under extreme ethnic fragmentation or in cases where there are a few large ethnic groups, widely disbursed. It has been adopted in cases of more moderate conflict and fluid identities
- Centripetalism encourages moderate, centrist policies and can counter extremism
- Integrative consensualism is only an option in societies which show signs that ethnicity will not endure as the sole driving force of democracy. However, it has the potential to be plagued by paralysis
- The electoral system that is best for ending the conflict may not be the best for long-term conflict management.
It is important not to overestimate the power of electoral systems to resolve conflicts. However, an appropriately crafted electoral system can do some good in fostering accommodative tendencies; an inappropriate system can do harm. Policy recommendations are that:
- There is no ‘right’ way of doing things. A one-size-fits-all approach will not work. An electoral strategy needs to be tailored to electoral geography and a realistic ‘diagnosis’ is a necessary precondition
- However, a broad rule for electoral systems is that those systems that represent groups in close approximation to their power and size are better for countries with rigid communal identification
- Systems that are majoritarian should also be avoided
- Sometimes, devices are in conflict with each other and it is necessary to prioritise what is important for a particular conflict
- The four systems are not mutually exclusive. A combination of systems could be used to suit the specific circumstances of a country, although this has yet to be put into practice
- Moderate reforms are often better than jumping to a new system.
