Power Sharing

Timothy D. Sisk

September 2003

Mark Gerzon reflects on the importance of symbols of power in his facilitation of a U.S. Congressional retreat.

If parties in intractable conflicts -- particularly in societies divided by deep ethnic, racial or religious differences -- find that they are unable to escalate their way out of conflict, but seek a compromise that assures them a permanent place at the bargaining table, they may turn to power sharing as a potential solution. Power sharing is a term used to describe a system of governance in which all major segments of society are provided a permanent share of power; this system is often contrasted with government vs. opposition systems in which ruling coalitions rotate among various social groups over time.

These are the basic principles of power sharing as traditionally conceived:

  1. grand coalition governments in which nearly all political parties have appointments;
  2. protection of minority rights for groups;
  3. decentralization of power;
  4. decision making by consensus.

Today, there is a more expanded definition of power sharing, such that a wide range of options exist for engendering consensus and compromise in deeply divided societies. One of the most important tasks for practitioners in intractable conflict is pairing thoughtful assessments about the causes and dynamics of a conflict with the wide range of power-sharing options that could potentially ameliorate tensions through consensus-oriented governance.

Ostensibly, power-sharing solutions are designed to marry principles of democracy with the need for conflict management in deeply divided societies. Power sharing involves a wide array of political arrangements -- usually embodied in constitutional terms -- in which the principal elements of society are guaranteed a place, and influence, in governance. From South Africa to Sri Lanka, from Bosnia to Burundi, from Cambodia to Congo, it is difficult to envisage a post-war political settlement that does not, or would not need to, include guarantees to all the major antagonists that they will be assured some permanent political representation, decision-making power, and often-autonomous territory in the post-war peace.

In many situations, the international community works proactively to encourage parties to adopt power sharing instead of waging war. In Afghanistan, for example, following the fall of the Taliban, international mediators worked hard at the Bonn negotiations in December 2001 to ensure that the transitional government under interim (now permanent) leader Hamid Karzai was broadly representative of the major ethnic groups in this highly diverse and long-conflicted country. In Ivory Coast, French mediators have brokered a pact in early 2003 to end that country's civil war; rebel commanders eventually took up appointments in a revamped cabinet.

However, power sharing is no panacea. Indeed, some types of power-sharing systems may contain the seeds of their own self-destruction as the search for consensus turns into deadlock by political leaders aware that they hold the power of veto over government action. Moreover, some elements of status-quo power will violently reject sharing power, as did elements of the Rwandan paramilitary groups in 1994 in opposition to the Arusha Accord of the previous year, leading to the worst genocide since World War II.

This essay assesses recent experience with power sharing as a means of living together in intractable conflict settings. It offers a classification of power-sharing models, and it includes examples of various approaches in practice. The conclusion for practitioners is to recognize that power sharing may be desirable, and necessary, as an immediate exit to deadly conflicts, especially those fought in the name of ethnic identity. In the long run, however, rigid power sharing is not a durable solution to intractable conflicts. Ideally, power sharing should wither away over time, as trust builds and the uncertainty of more "normal" majority rule democracy becomes acceptable. At the same time, practitioners should think innovatively about options that can allow such an evolution from formal sharing of power -- often by exclusive ethnic groups -- to a more socially inclusive and integrated form of representation.

Types and Elements of Power Sharing

Helen Chauncy explains how fear can interfere with coexistence efforts.

A long-standing misconception about power-sharing options for intractable conflicts is that there is a single formula for sharing power, which for many years has been called "consociationalism."[1] The elements of this approach to power sharing are well known: grand coalitions, proportional representation, cultural autonomy or federalism, and the mutual veto. Yet this prototype of power sharing is but one of what is in fact a very broad range of political options for settling ethnic conflicts, the gist of which can be exceptionally different in terms of aims, structures, and effects on promoting inter-group moderation and compromise. What are the principal options for sharing power?[2]

  • Autonomy. For many conflicts today, such as Azerbaijan (Karabagh), Sudan, or Sri Lanka, autonomy is often seen as a reasonable way to balance the claims of states for territorial integrity and the claims of rebel forces for secession. [3] Autonomy, as eminent scholar Yash Ghai suggests, is not a term on which there is a consensus definition. Nonetheless, his best effort at one is useful: "Autonomy is a device to allow an ethnic group or other groups claiming a distinct identity to exercise direct control over important affairs of concern to them while allowing the larger entity to exercise those powers which are the common interests of both sections."[4] Among the forms of autonomy is symmetrical federalism, in which all units enjoy similar powers, and asymmetrical federalism that might provide enhanced powers to a particular region.
  • Power Sharing: Group Building-Block Approach. Another possible option is a looser form of autonomy, not always explicitly territorial, termed consociationalism. The option is in essence a group building-block approach that relies on accommodation by ethnic-group leaders at the political center and guarantees for group autonomy and minority rights; in essence, this approach is "consociational" in that it encourages collaborative decision-making by parties in conflict. The key institutions are federalism and the devolution of power to ethnic groups in territory that they control; minority vetoes on issues of particular importance to them; grand coalition cabinets in a parliamentary framework, and proportionality in all spheres of public life (e.g., budgeting and civil service appointments).[5] Bosnia's 1995 Dayton Accord is a good example of this approach in practice.

Consociational Power Sharing

Table 1: Consociational Power Sharing
Principles Practices Problems
Broad-based coalitions among ethnic political parties Grand coalition governments Elites may initiate conflict to bolster their power at the center
Minority or mutual veto on matters of importance to the group Group rights defined in constitutional terms for named ethnic, racial, religious, or cultural groups Can reinforce the ethnic divisions in society rather than promoting cross-cultural understanding
Proportionality Proportional representation electoral system and the proportional allocation of jobs, spending, representation, and participation by ethnic group leaders May reflect well the divisions in society but does not provide incentives for building bridges across community lines
Group autonomy Federalism, territorial, or "corporate" May contain disincentives for contending groups to live peacefully together
  • Power Sharing: Integrative Approach. By contrast, the integrative approach eschews ethnic groups as the building blocks of a common society. As a distinct set of options for power sharing, this approach rejects cohesive ethnic or other groups (such as "confessional" or religious factions in Lebanon) as the building blocks of society. This approach features options that purposefully seek to integrate society along the lines of division. This approach can be called "centripetalism," because it tries to engineer a center-oriented spin to political dynamics.

The integrative approach seeks to build multiethnic political coalitions (again, usually political parties), to create incentives for political leaders to be moderate on divisive ethnic themes, and to enhance minority influence in majority decision-making.[6] The elements of an integrative approach include electoral systems that encourage pre-election pacts across ethnic lines, non-ethnic federalism that diffuses points of power, and public policies that promote political allegiances that transcend groups. Some suggest that integrative power sharing is superior in theory, in that it seeks to foster ethnic accommodation by promoting crosscutting interests. Others, however, argue that the use of incentives to promote conciliation will run aground when faced with deep-seated enmities that underlie ethnic disputes and that are hardened during the course of a brutal civil war. Table 2 summarizes this option and its related practices and problems.

Integrative Power-Sharing

Table 2: Integrative Power Sharing
Principles Practices Problems
Incentives for elite and mass moderation on divisive ethnic or racial themes A president who stands for all groups and who emphasizes moderation and reconciliation (like a Mandela) Leaders who can rise above the fray of intergroup enmity are hard to find; they can't be simply invented.
Intra-group contestation and inter-group moderation in electoral contests The use of vote-pooling electoral systems, such as the Single Transferable Vote or the Alternative Vote People may be unwilling to vote for candidates who are not from their community
Minority influence, not just representation Federalism is a way to give all minority groups access to power in various regions; the regions serve as a training ground for national-level moderates Political leaders and key public figures may not be willing to respond to the incentives for moderation, preferring that minority representation remain token or symbolic

Although this typology presents two conceptually distinct approaches, it is clear power-sharing options can be pieced together in a number of ways. In deciding which power-sharing institutions and practices might work, there is no substitute for intimate knowledge of any given country. In multiethnic Fiji, for example, a four-year expert review of the country's political system produced a set of recommendations for a recently adopted constitution that combines measures to guarantee a minimum level of traditional Fijian (as opposed to Indo-Fijian) representation in parliament (a group building-block option) with measures to promote the formation of political alliances across group lines (an integrative option). The Fiji experience points to how a well-conceived process, featuring a balanced panel of experts with firm political support, can arrive at creative solutions specifically tailored to a unique set of problems.[7] The Fiji case is instructive precisely because the efforts of spoilers to disrupt integration along ethnic lines were only temporarily successful; as Fiji recovers from the attempted coup d'etat of 2000, it has returned to an integrationist formula for resolving its ethnic tensions.

Matching Problems to Solutions

A key feature of consociational power-sharing is the mutual veto, whereby decisions are only made with the widest possible consent and only with a near consensus. However, this often leads to the use of "political blackmail." Unable to get consensus, governance stagnates and policy-making drifts; the result is a "cold peace," in which the parties refrain from violence but have not embarked on a serious process of reconciliation, either. In many ways, this is the sad story of post-war Bosnia , which has muddled through elections and a period of peace without much progress in effective consensus-oriented government. The inability to make or implement policy due to protracted disagreement can lead to frustration and eventually defection from a peace accord. War can erupt anew. Historically, the outbreak of civil wars in Angola, Cyprus, Lebanon, Sierra Leone, and Sudan have all been the result of broken power-sharing agreements that led to renewed violence.

Power-sharing solutions make for good transitional devices, but in the long run the best outcome may well be a much more fluid form of democracy that allows for the creation of flexible coalitions that bridge the ethnic divide. A central question that has yet to be fully explored is the terms under which power-sharing, consensus-oriented forms of democracy can evolve into more flexible institutions that can foster reconciliation and a broader national identity. If sustainable peace comes through "conflict transformation," as argued by John Paul Lederach (1997), power sharing is often too rigid a system to allow for the social and political changes necessary for addressing the underlying causes of conflict that give rise to war.[8]

How can the rigid structures of political power-sharing wither over time to the point where the guarantees for group security they contain are no longer necessary? This is not a purely academic question. In Bosnia, for example, the ability of NATO's international peacekeepers to end their occupation is premised on the ability of the power-sharing institutions forged in the 1995 Dayton Agreement -- now dominated by nationalists -- to melt into more moderate and ethnically mixed political institutions.[9]

If power sharing is at best a transitional device, this conclusion begs the question of what types of political institutions can be expected to allow democratic decision-making to prosper in post-war environments in which politics remain deeply divided. There is no way to say prima facie which type of power-sharing system -- consociational or integrative -- is inherently best.

In matching options to solutions, much depends on the level of enmity between the contending groups, the trajectory of the war (e.g., the extent of ethnic separation that occurred) and whether or not in their negotiations they can accept any degree of uncertainty or vulnerability to political loss. Critical to analysis of the problems is a coherent assessment of the role that ethnicity plays in the turn to violence and the prominence of identity as a cause of conflict. At some point, it becomes impossible to live together in broad, tolerant, multiethnic coalitions; in such cases, perhaps consociational democracy is the best alternative to violence. When consociationalism can not work, autonomy might be a solution. When even autonomy is not possible, the time may be ripe to consider complete separation (or, in the jargon of the field, partition).


Intractable conflicts are characterized in part by the inability of the parties in them to completely prevail by escalation. When they reach a stalemate and are highly motivated to de-escalate, conflict-resolution practitioners may be in a position to help the parties arrive at a workable solution for power sharing. While there may be understandable pressures for power sharing, there are immediate risks to such an arrangement from spoilers (as in Rwanda) and longer-run risks from the design of political institutions (as in Bosnia).

To reconcile immediate imperatives with the sustainability of peace over time, power sharing will work best when it can, over time, wither away. Whether in South Africa, Northern Ireland, Bosnia, or Lebanon, in the immediate term, formal power sharing has been an effective confidence-building device to ensure that all groups with the capacity to spoil a peace settlement should be included in the institutions and given influence in decision-making. Over time, however, postwar societies need to move beyond the mutual hostage taking that a guaranteed place at the decision-making table implies, the deadlock it inevitably creates, and the construction of postwar societies around the fixed and unyielding social boundaries of ethnicity.

Integrative power-sharing solutions have an inherent advantage, if they can be achieved. When successful, they engineer a moderation-seeking, centripetal spin to the political system, one that allows for ethnicity but promotes fluid coalitions that transcend the cleavages of conflict in war-torn societies. A practical way to begin is to purposefully manipulate the electoral system to provide new incentives to moderate and coalesce across group lines, as suggested above. Electoral systems should be designed to give politicians real incentives to motivate, moving beyond a perhaps natural instinct to play the communal card to attain power. There is emerging evidence that such clever design can promote moderation in intractable conflicts, as examples from Northern Ireland, Fiji, and Papua New Guinea attest.[10]

The clever design of power-sharing institutions, no matter how careful, can not resolve some of the inherent problems that lead to discord and the inability to reach consensus in today's deeply divided societies. If thoughtful analysis and clever design were sufficient, it is likely that the Cyprus dispute -- which resists settlement even though United Nations mediators have labored for years extensively over every detail of a mutually acceptable power-sharing solution -- would have been resolved decades ago. Regrettably, recent efforts by the international community to negotiate acceptance of the United Nations power-sharing plan apparently failed again in early 2003.[11] Nevertheless, if and when the Cypriots and others in similar situations are ready to settle, they will find themselves facing basic choices about sharing power and how best to do so. As a means for exiting intractable social conflicts, there seems to be no alternative.

[1] For the seminal articulation of this approach, see Lijphart, Arend. Democracy in Plural Societies: A Comparative Exploration (New Haven: Yale University Press: 1977).

[2] For a more thorough overview of power-sharing options, see Timothy Sisk, Power Sharing and International Mediation in Ethnic Conflicts (1995), and Peter Harris and Ben Reilly, Eds. Democracy and Deep-Rooted Conflict: Options for Negotiators (1998).

[3] See also Ruth Lapidoth, Autonomy: Flexible Solutions to Ethnic Conflicts (1997) and Hurst Hannum, Autonomy, Sovereignty, and Self-Determination: The Accommodation of Conflicting Rights (1990).

[4] Autonomy and Ethnicity: Negotiating Competing Claims in Multi-Ethhic States. Cambridge University Press (2000). 

[5] For a recent assessment of consociational power-sharing in Europe, see Ulrich Schneckener, "Making Power Sharing Work: Lessons from Successes and Failures in Ethnic Conflict Regulation," Institut fur Interkultrelle und Internationale Studien, University of Bremen (Working Paper Nr. 19/2000).

[6] For the classic articulation of this approach, see Donald Horowitz, Ethnic Groups in Conflict (Los Angeles and Berkeley: University of California Press, 1985).

[7] See the report of the Constitutional Review Commission, The Fiji Islands: Toward a United Future, Suva, 1996.

[8] For an excellent summary of underlying causes of ethnic conflicts, see Brown (1996).

[9] For example of the practical policy challenges, see "Turning Strife to Advantage: A Blueprint to Integrate the Croats in Bosnia and Herzegovina" an International Crisis Group Report (March 15, 2001), available at www.intl-crisis-group.org.

[10] Ben Reilly, Democracy in Divided Societies: Electoral Engineering for Conflict Management (Cambridge: Cambridge University Press, 2001).

[11] See "Cyprus peace talks end in failure," on www.CNN.com, http://edition.cnn.com/2003/WORLD/europe/03/10/cyprus.meet.reut/

Use the following to cite this article:
Sisk, Timothy D.. "Power Sharing." Beyond Intractability. Eds. Guy Burgess and Heidi Burgess. Conflict Information Consortium, University of Colorado, Boulder. Posted: September 2003 <http://www.beyondintractability.org/essay/power-sharing>.

Additional Resources