Power Sharing in Multiethnic States: Federalism and Consociationalism

Multiethnic states face a persistent dilemma. How do you keep a country united when its people are divided by language, religion, region, or ethnicity? If a single dominant group controls the government, smaller communities feel excluded and may turn to protest, separatism, or even violence. Power sharing is the political science answer to this problem. It is a set of institutional arrangements that distribute political authority across groups so that no single community can monopolise the state. Two of the most influential models are federalism and consociationalism. Both try to hold unity and diversity in balance, but they do so in very different ways.

Why power sharing matters in divided societies 🔗

For much of the twentieth century, scholars believed that stable democracy required a fairly homogeneous society. The assumption was that deep ethnic or religious divisions would tear a democracy apart. The Dutch political scientist Arend Lijphart challenged this view. He argued that democracy can survive and even thrive in deeply divided societies, provided the right institutions are in place. His work showed that what mattered was not the absence of cleavages but how political power was organised around them.

Power sharing rests on a simple insight. When groups fear permanent exclusion from power, the stakes of politics become dangerously high. A community that can never win an election has little reason to accept the rules of the game. By guaranteeing every significant group a meaningful share of authority, power sharing lowers these stakes and makes cooperation more attractive than conflict. Power sharing has become one of the most discussed approaches to institutional design for divided societies, especially in the decades after the Cold War, when many post-conflict states needed new constitutions.

Federalism: dividing power across territory 🔗

Federalism shares power vertically, between a central government and regional governments. Each level has its own sphere of authority guaranteed by the constitution. The central government typically handles matters like defence, foreign affairs, and currency, while regional units control areas closer to everyday life, such as education, land, and local administration.

Federalism becomes a tool for managing ethnic diversity when group boundaries roughly match territorial boundaries. If a linguistic or ethnic community is concentrated in one region, giving that region autonomy effectively gives the community self-governance. The group can run its own affairs, protect its language and culture, and make decisions on local issues without seeking permission from a distant capital. This is why federalism is often described as a method of ethnic conflict regulation.

The Indian example: linguistic reorganisation 🔗

India offers one of the clearest examples of federalism used to accommodate diversity. After independence, the country faced strong demands for states organised along linguistic lines. The central government was initially reluctant, but sustained popular movements forced a change of approach. The States Reorganisation Act of 1956 redrew internal boundaries to create states based largely on language, producing states such as Andhra Pradesh, Karnataka, Kerala, and later Maharashtra and Gujarat.

The logic was straightforward. By becoming the demographic majority within their own state, linguistic communities could capture the state legislature and govern themselves. They gained control over state finances and the power to legislate on subjects assigned to the states under the Constitution. Crucially, this autonomy came without secession. Groups could feel like legitimate members of the nation while still exercising real authority over matters they cared about most.

Comparative scholars describe India as a ‘holding together’ federation rather than a ‘coming together’ one. Unlike the United States, where independent states joined to form a union, India began as a single union and then devolved power to accommodate its diversity. Despite, or perhaps because of, its extreme diversity, India has achieved remarkable success in maintaining relative ethnic peace and political order.

Beyond language: special arrangements 🔗

Federalism in a diverse country is rarely uniform. For regions where ethnic identity does not map neatly onto language, India developed special constitutional provisions. The creation of states like Jharkhand, Chhattisgarh, and Uttarakhand recognised regional and tribal aspirations. In the Northeast, arrangements such as autonomous councils and the Bodoland Territorial Council attempted to give territorially concentrated minorities a degree of self-rule within larger states.

These cases also reveal the limits of territorial power sharing. Researchers studying the formation of Telangana and the Bodoland Territorial Council argue that drawing boundaries on linguistic principles, while necessary, is not always sufficient to accommodate minorities. Problems persist when a dominant group uses language to entrench its own majority, when power-sharing arrangements within a state are weak, or when communities carry long memories of historical injustice. A state created for one group can become a site of exclusion for the smaller groups living inside it.

Consociationalism: sharing power among elites 🔗

Consociationalism takes a different route. Instead of dividing power by territory, it shares power at the centre among the leaders of different communities. This model is most useful when groups are not neatly separated into regions but are intermixed across the country, making territorial division impractical.

Lijphart developed the theory by observing how the Netherlands remained stable despite being divided into separate religious and ideological communities. He found that cooperation among the leaders of these segments held the system together. From this he built a generalisable approach to managing ethnic conflict, defining a consociational state as one with deep internal divisions that stays stable through consultation among group elites.

The four features of consociational democracy 🔗

Lijphart identified four core characteristics that define a consociational democracy. Together they form the institutional toolkit of this model.

Grand coalition. The political leaders of all significant groups govern together rather than competing in a winner-takes-all contest. Executive power is shared, so no major community is left in permanent opposition.

Mutual veto. Each group can block decisions that threaten its vital interests. This reassures minorities that the majority cannot override their fundamental concerns through sheer numbers.

Proportionality. Political representation, civil service positions, and public funds are distributed in proportion to each group’s size. Elections typically use proportional representation rather than first-past-the-post systems that tend to favour large groups.

Segmental autonomy. Each group runs its own internal affairs, particularly in sensitive areas like education, culture, and religion. Where groups are territorially concentrated, this autonomy can take the form of federalism, which shows how the two models can overlap.

Consociationalism in practice 🔗

Several countries are commonly cited as consociational systems. Belgium shares power between its Flemish and French-speaking communities. Lebanon distributes top offices among religious communities by formula, reserving the presidency, the premiership, and the speakership for different groups. Northern Ireland’s Good Friday Agreement built power-sharing institutions to bring unionists and nationalists into a common government. Interestingly, scholars have also identified India as a case of informal consociationalism, where power-sharing operated through political conventions and broad coalitions rather than through formal constitutional rules.

Comparing the two models 🔗

The clearest difference lies in how each model treats territory and identity. Federalism works best when ethnic groups are geographically concentrated, because it hands authority to regions. Consociationalism works even when groups are mixed together, because it shares power among leaders at the national level regardless of where their followers live.

There is also a difference in what each model does to ethnic identity itself. Consociationalism accepts and builds on existing divisions, organising the state around recognised groups. Critics worry this can freeze divisions in place. An alternative approach called centripetalism, developed by Donald Horowitz, argues for the opposite strategy. Instead of replicating ethnic divisions in the legislature, it designs electoral systems that reward politicians for seeking votes across communities, aiming to depoliticise ethnicity and encourage multiethnic parties. Where consociationalism asks elites to cooperate after elections, centripetalism tries to build cooperation into the act of winning them.

The criticisms and trade-offs 🔗

Neither model is a perfect solution, and both have drawn serious criticism. Federalism can entrench regional dominance, allowing a state’s majority community to marginalise its own internal minorities. It can also feed regionalism and complicate national unity over time, as identity becomes tied to territory.

Consociationalism faces the charge that it can produce paralysis. When every group holds a veto, governing can become extremely difficult. Lebanon illustrates the danger vividly. After granting representation and vetoes to its many communities, the country experienced repeated deadlock, including a stretch between 2014 and 2016 when the groups could not agree to elect a president. By recognising and rewarding group identity, consociationalism may also make those identities more rigid and permanent, the opposite of long-term integration.

In reality, many divided societies do not choose one pure model. They combine elements of both, mixing federal structures, proportional representation, and incentives for cross-ethnic cooperation. The best institutional design usually depends on local conditions rather than a one-size-fits-all formula, and arrangements that grow organically out of a society’s own circumstances tend to last longer than those imposed from outside.

Balancing unity and diversity 🔗

Both federalism and consociationalism pursue the same underlying goal. They try to give every significant group a genuine stake in the political system so that diversity becomes a feature to be managed rather than a threat to be suppressed. Federalism does this through territory, granting regions autonomy. Consociationalism does it through elite cooperation, guaranteeing each group a seat at the table. The common thread is the prevention of domination by any single community, replacing the fear of permanent exclusion with the security of guaranteed inclusion. For multiethnic states, getting this balance right is often the difference between durable peace and recurring conflict.

What do you think? Is it better for a divided society to recognise and build institutions around ethnic identities, or to design a system that encourages people to look beyond them? And in a country as diverse as India, do you think federalism alone is enough to accommodate every community, or are additional power-sharing mechanisms necessary?

References

1. https://en.wikipedia.org/wiki/Arend_Lijphart
2. https://en.wikipedia.org/wiki/Federalism
3. https://www.mha.gov.in/
4. https://utsynergyjournal.org/2021/04/06/linguistic-reorganization-and-indian-federalism-the-cases-of-andhra-pradesh-and-punjab/
5. https://nottingham-repository.worktribe.com/OutputFile/943700
6. https://doi.org/10.1080/14736489.2017.1279934
7. https://pesd.princeton.edu/node/246
8. https://adambrown.info/p/notes/lijphart_multiethnic_democracy
9. https://en.wikipedia.org/wiki/Consociationalism
10. https://onlinelibrary.wiley.com/doi/full/10.1111/spsr.12384
11. https://en.wikipedia.org/wiki/Centripetalism
12. https://www.geopoliticalmonitor.com/why-consociationalism-is-not-helping-in-post-conflict-settings/
13. https://onlinelibrary.wiley.com/doi/full/10.1111/spsr.12371