The Pitfalls of Power Sharing

People are calling for a so-called unity government to stem the violence in Mozambique. But there is a better way to set the country on the right course.

By Alexander Noyes

January 2025

Mozambique is burning. Things may go from bad to worse when the purported winner of the country’s disputed election — Daniel Chapo of the ruling Frelimo party — is sworn in on January 15. But calls for a so-called unity government should not be heeded. Instead, regional and international actors should push for a rerun of the October 2024 polls, this time with more robust international involvement.

Upstart opposition leader Venâncio Mondlane claims that the October 2024 poll was rigged and that he is the rightful winner. The country’s constitutional court awarded Chapo 65 percent of the vote to Mondlane’s 24 percent, but domestic and international election observers noted grave concerns about the polls, including “widespread irregularities, subsequent violence, and fraudulent practices.” More than 250 people have already died in postelection violence protesting the disputed results during the last three months, most at the hands of security forces.

On January 9, Mondlane returned from self-imposed exile, saying he is ready to negotiate an end to the crisis. The United States, South Africa, the African Union (AU), and the Southern African Development Community (SADC)  have called vaguely for restraint and dialogue to end the violence. Many of the protagonists, including the influential Catholic church and unnamed moderate politicians in the ruling party, appear open to some sort of power-sharing deal. But this would be a mistake.

Compromised Coalitions

Mozambique is far from the only country that has considered the power-sharing model amid conflict or crisis. Indeed, some form of power-sharing pacts are now the default response to address a wide range of conflicts worldwide, from Afghanistan to Zimbabwe. Historically tried in countries experiencing long and bloody civil wars, including Mozambique in the early 1990s, power sharing is now frequently used to end a range of other less intense conflicts, such as electoral violence or coups.

Yet, all too often power sharing merely leads to compromised coalitions. Especially when used in the wake of election violence, unity governments risk rewarding incumbent power grabs and constraining democratic competition.

My forthcoming book on power sharing, Compromised Coalitions: The Paradox of Post-Conflict Power Sharing in Africa, is based on a decade of research and more than a hundred interviews with senior politicians and officials involved in unity governments. It shows that while power-sharing governments in the wake of election violence can reduce tensions and lead to some democratic reforms, the popular but little-understood model of conflict resolution is rarely the right answer and brings with it significant downsides. These include infighting and ineffectiveness, corruption, and the incentives for greater violence.

The experiences of Kenya, Togo, and Zimbabwe, all countries that chose power sharing after violent disputed elections, are instructive. Zimbabwe’s unity government, in office from 2009 to 2013, was largely a failure. The ruling party refused to share any real power and blocked almost all reform efforts, except for a new constitution, and SADC failed to effectively monitor or enforce the agreement it helped birth.

On the other hand, Kenya’s unity government, in office from 2008 to 2013, achieved some real successes. The parties proved willing to compromise and passed a host of significant democratic reforms, including a new constitution, and made considerable progress on reining in the country’s notoriously abusive police. International and regional pressure, especially from the United Nations and the AU, proved critical to helping get these reforms across the finish line. The government, however, failed to halt corruption or hold perpetrators accountable for the postelection violence.

Togo’s experience with power sharing in 2006 also showed mixed results. The unity government made some headway on electoral reform, delivering a relatively free and fair legislative election in 2007. It also took some important initial steps to reform the security forces. Strong outside pressure from the European Union and the Economic Community of West African States was again key. Yet the power-sharing accord failed to pass any constitutional reforms or to put the country on a lasting democratic trajectory. Importantly, the opposition failed to win subsequent elections, a commonality across all three cases.

Mozambique’s Moment

My research offers several takeaways for Mozambique. First, policymakers should strongly consider alternative options that are less undemocratic by nature. Indeed, mediators need to seriously rethink their current default approach to power sharing and use the model much more sparingly. The downsides of these compromised coalitions all too often outweigh the benefits.

Instead of power sharing, the parties in Mozambique should negotiate a rerun of the election. To ensure that the poll truly reflects the will of the people this time, international actors, including the United States, the UN, the AU, and Mozambique’s democratic neighbors, will need to be more deeply involved. These outside governments and organizations should help fund the rerun and require that observers be present for the vote-counting process. Funding should also be contingent on the inclusion of a full parallel vote tabulation, a tool used around the world to help detect fraud.

Convincing the ruling party to do so will take forceful coordinated action from the region and international partners, including the United States. U.S. policymakers should make it clear that if the ruling party refuses to budge, aid will be cut, especially from a recent Millennium Challenge Corporation program that is contingent on good governance.

This option may prove a bridge too far for the ruling party or too risky for international actors who may fear mass violence or losing influence to competitors such as China or Russia. If domestic intransigence or risk-averse outside players veto a rerun, power sharing may well become the least-worst option in Mozambique. If this highly imperfect path is nonetheless chosen, policymakers will need to pay close attention to the makeup of the coalition cabinet and ensure that real power is shared. To work, power sharing needs the conflict parties to agree to genuine political inclusion and practice democratic compromise within the unity government.

Finally, this real sharing of power will then need to be backed by coordinated and intense regional and international pressure to implement electoral and other political reforms ahead of the next election to help even the playing field.

Domestic officials, local civil society organizations, and external partners would therefore be wise to push for concrete power-sharing provisions that take into account the actual power of “hard”- and “soft”-power positions and ministries. Failing to do so would mean that the actual balance of power between the parties would be lopsided, allowing the ruling party to stymie any real reform under power sharing, as seen in Zimbabwe.

They should also stubbornly refuse to countenance any agreement that does not include robust third-party enforcement, perhaps with the UN and the AU designated as guarantors. The experiences of Kenya and Togo show that coordinated external pressure can make all the difference, while SADC’s lackluster monitoring and enforcement in Zimbabwe hindered progress on democratic reform. Getting the ruling party to agree to an equitable share of real power and an enforcement mechanism with teeth will also require strong and integrated pressure from the same cast of regional and international actors, with a combination of carrots and sticks.

Power sharing is not the answer for Mozambique, yet it may still be chosen as the only viable way forward. In that case, prioritizing political inclusion and outside enforcement of the pact will help to make any coming coalition in Mozambique slightly less compromised.

Alexander Noyes is a fellow in the Strobe Talbott Center for Security, Strategy, and Technology at the Brookings Institution. He is the author of Compromised Coalitions: The Paradox of Post-Conflict Power Sharing in Africa, forthcoming from the University of Pennsylvania Press.

Copyright © 2025 National Endowment for Democracy

Image credit: Alfredo Zuniga/AFP via Getty Images