Power-sharing and its side effects

One morning in May 2009, Andreas Juon is taking the bus to the Children's Model High School in the centre of Kathmandu, where he has been teaching English for nearly three months, when suddenly he hears loud thumps. Stones bounce off the windshield of the bus, leaving cracks in the shape of spiderwebs.

"It was a moment I'll never forget," says the ETH postdoc almost 12 years later, looking relaxed in his Zurich flat as we talk via Zoom. "Naturally, on the day itself, I didn't reflect on the causes of civil wars," he replies to the question as to whether there is a connection between the attack on the bus by Maoist protesters and his current research. "But my experiences in Nepal undoubtedly heightened my awareness of ethnic conflicts."

Formative experiences in Nepal

In 2009, three years had passed since the government and the Maoist rebels in Nepal had reached a peace agreement, putting an end to a bloody civil war that had raged for ten years. Under the terms of the peace accord, the Maoists and formerly excluded ethnic minorities would form part of the government from then on.

"Without this inclusion of the Maoists, the 2006 peace agreement wouldn't have come about," Juon says with conviction today. However, the medium-term consequences of power-sharing in the country have not all been positive. Due to the newly established veto rights of various players, the government was often gridlocked and unable to act between 2006 and 2009. In May 2009, this turbulent phase culminated in the removal of the Maoist prime minister. Protests and strikes ensued, during which the school bus of the then 19-year-old Juon was pelted with stones.

For the postdoc, who is now a researcher in the International Conflict Research group led by ETH Professor Lars-Erik Cederman, this period of his life was formative - and not only on a personal level. His experiences in Nepal also inform a major aspect of his current research: although power-sharing is an important tool for bringing violence to an end in the aftermath of civil wars, in the medium to long term it leads to a series of unintended and sometimes damaging side effects that should also be considered.

Ethnic civil wars and the integration of minorities

After his five-month stint of civilian service in Nepal, Juon returns to his native Switzerland to study human geography at the University of Zurich. For his Master's degree, he switches to the Center for Comparative and International Studies, which fuses the political science professorships of ETH Zurich and the University of Zurich.

His academic interest in ethnic conflicts is awakened primarily by the lectures and seminars of Cederman. For years, Cederman and his former doctoral students and postdocs, who have gone on to take up research positions at various top universities, have been influencing the international discussion around the causes of ethnic conflicts to an extent pretty much unparalleled by any other group of researchers.

In numerous publications, they demonstrate that civil wars are more likely when ethnic minorities are excluded from majority rule, are well organised, or have already been involved in conflicts in the past. Conversely, the inclusion of large ethnic minorities reduces their dissatisfaction and thus the risk of violence breaking out.

Ethnic civil wars such as those in the former Yugoslavia, in Rwanda, in Lebanon and indeed in Nepal powerfully illustrate that how a country politically integrates minorities is one of the chief questions a political system has to answer, perhaps even the most important. After all, it is a measure not only of how developed a democracy is, but also how stable. This dynamic is borne out by the history of Switzerland, where more and more groups have been integrated in incremental fashion since the foundation of the federal state in 1848.

From Zurich to London and back again

Although Juon is fascinated by the research into ethnic conflicts, he ends up writing his Master's thesis on the side effects of power-sharing on democracy. His supervisor is Daniel Bochsler, who has since become a professor in Vienna and Belgrade, but continues to lecture at the University of Zurich. The supervisor-student relationship quickly morphs into a productive collaboration. Between 2016 and 2020, Bochsler and Juon publish three joint articles dealing, among other subjects, with power-sharing and the inclusion of minorities.

After completing his Master's degree in the autumn of 2015, the now 26-year-old Juon goes to University College London to pursue doctoral studies. He wants to understand how the manner in which ethnic minorities are incorporated into governments and parliaments affects inter-ethnic relations and the attitudes of minority and majority groups. Juon spends four years in London exploring these questions. He is guided in his investigations by the well-established distinction in comparative politics between corporate and liberal power-sharing.

Corporate and liberal power-sharing

Corporate forms of inclusion define explicitly which groups should be represented in the government or the parliament and how strongly. This is usually done by way of quotas or veto rights. Belgium is a good example: the constitution guarantees the Walloon minority half of all cabinet seats, and important laws can be blocked by a majority of each of the three linguistic groups in the country. Another charter that is considered particularly corporatist is the post-war constitution in Bosnia from 1995, which reserves access to important government offices for the Bosniak, Serb and Croat ethnic groups.

"This explicit enumeration of rights offers ethnic minorities the strongest guarantee of being politically recognised and getting their voices heard," Juon says. Given that deep mistrust tends to reign between ethnic groups after civil wars, such guarantees are often the only way to secure peace.

But corporate power-sharing also comes at a price: the exclusion of small minorities. In Bosnia, for example, Jewish citizens or members of the Roma community are explicitly excluded from political life. To this day, Juon's research investigates these fault lines between the incorporation of a few large groups and the exclusion of smaller groups.

In contrast to corporate power-sharing mechanisms, liberal ones are more inclusive. They work via strongly proportional election systems with very low barriers to entry and with big majorities required for important decisions. According to Juon, the advantage of this approach is that the guarantees for minorities are distributed more evenly while also being less strong.

"In liberal power-sharing, it is not the constitution but the electoral system that decides which minorities are represented in the government and the parliament, as well as how strong their representation is. This gives smaller groups the opportunity to further their interests as well." The interim post-apartheid constitution in South Africa is often cited as an example of a liberal variety of inclusion: thanks to a thoroughly proportional election system and low hurdles for participation in government - each party with a 5 percent share of seats in parliament has a right to cabinet posts - the white minority was represented in the parliament and the government even after the end of apartheid.

A new dataset for global power-sharing

So how does one investigate how corporate and liberal forms of power-sharing affect the attitudes both of minority and majority groups? Which of the two forms tends to lead to stability and why? Answering these questions calls for data on the inclusion of minorities from as many countries as possible. But from the start of his doctoral studies in London, Juon realises that this data is woefully inadequate if it exists at all.

So he decides to compile his own dataset. "That was definitely the most laborious part of my dissertation. For over a year, I went through all the constitutions and constitutional amendments of 180 countries from 1945 to the present day and coded them according to whether they leaned more to the liberal or to the corporate power-sharing model," Juon says with a note of pride.

By the end, he has reviewed over 700 constitutional texts. The result is the Constitutional Power-Sharing Dataset. This comprehensive dataset allows the ETH postdoc to statistically analyse for the first time the effect of different forms of inclusion on the attitudes of majority and minority groups. So as to be able to measure the latter, Juon combines a series of comparative surveys. Among them are the World Values Survey, the Eurobarometer and the Afrobarometer, each of which polls how satisfied people are with their government.

Investigating inclusion envy

On the basis of this new dataset, Juon shows that corporate power-sharing leads to higher satisfaction with the government amongst the groups that were explicitly included. This is not especially surprising. People who are part of an ethnic group feel more fairly treated on average when they have a share in political power, as this gives them more influence and a larger say in how they are governed. Their incentive to revolt against the government falls, creating stability in the wake of ethnic conflicts. Juon is one of the first to be able to demonstrate this statistically on such a wide scale.

However, his analysis goes beyond these groups. He also investigates how members of minorities that are excluded from power-sharing respond. Once again, it is not surprising here that these people are particularly dissatisfied with the government. What is unexpected, though, is the discovery that this dissatisfaction increases the more that other groups are integrated.

Juon attributes this to something akin to "inclusion envy". When evaluating their satisfaction with the government, people seem to consider above all how strongly they are included in power-sharing compared to others. "If other minorities are more strongly integrated, this makes their own exclusion seem worse than if all minorities were excluded equally," Juon explains.

Or, to put it another way, a political system dominated by a majority group is perceived as fairer by excluded small minorities than a corporate system in which only the largest minorities are included. "The price of stability is sometimes that you simply cannot integrate all groups equally," Juon says.

Power-sharing as a cause of coups?

Juon's models additionally show that liberal power-sharing institutions also reduce the dissatisfaction of minorities, albeit that the effect is much weaker than in the case of corporate institutions. However, the dissatisfaction is somewhat more evenly distributed amongst minorities, and smaller minorities have better opportunities to take part in the political process.

But liberal solutions are much more difficult to bring about after violent conflicts, as they offer minorities weaker protection. "The more that minorities fear they could be outmanoeuvred by the majority, the more they demand corporate institutions. In a peaceful context, by contrast, liberal regulations are advantageous, as they are more inclusive and minimise unintended side effects," Juon says.

The politically charged question as to the effect of corporate power-sharing on majority groups is somewhat less clear. For African countries south of the Sahara in particular, Juon's analyses indicate that the formal inclusion of large minorities increases the likelihood of a putsch by majority groups.

The mechanism works as follows: The more the majority group is obliged to share power with minorities, the greater its dissatisfaction. Political elites can then exploit this dissatisfaction to undermine the entire corporate system and reinstall a dominant majority government.

"These results show that inclusive systems don't promote stability in and of themselves. Too much power-sharing can also backfire," Juon cautions. It is his analysis of these trade-offs between inclusion and exclusion that make Juon's research so useful for political decision-makers. To reduce the risk of a backlash by the majority, for example, a country can recognise the majority population as the constitutive group of the state.

New research programme looks at autonomous

regions

For the past six months, Juon has been working at the place where he first came into contact with research on ethnic civil wars as a Master's student: in Cederman's International Conflict Research group. "To be able to continue my research in an environment where they've been doing pioneering work in the field of conflict research for years is a unique opportunity for me," Juon says.

Currently his focus is on two research projects: one involves delving deeper into the effect of power-sharing on majority groups. After all, if Juon's thesis is correct, then the dissatisfaction of the majority as a consequence of the inclusion of minorities should become apparent not only in the form of coups, but also at elections. Accordingly, he is investigating how the inclusion of minorities affects the electoral success of right-wing nationalist parties. The election of Donald Trump, for example, can be interpreted as the reaction of a white majority to the presidency of Barack Obama.

Juon's second research project takes a fresh look at autonomous regions in federal systems. Here, again, the ETH postdoc is interested in how the autonomous status affects the relations between majority and minority. Can regional majority groups use their power to strengthen the autonomy of the regions they dominate and so expand their influence? And is there more violence towards internal minorities in these regions? Unlike when he was a doctoral student in London, this time there is good data for him to work with. He should know: he compiled it himself.