How does constitutional power-sharing, adopted in numerous heterogeneous and post-conflict countries around the world, affect democracy? We argue that it has two faces: on the one hand, it fosters pluralism; on the other, it limits free competition and confines political power to a narrow set of often undemocratic elites. Moreover, we argue that the importance of these contrary consequences diverges between two institutional types of power-sharing: First, a more flexible and open liberal type with clear benefits for democracy; second, a more rigid, corporate type with ambiguous consequences for democracy. Using new data on constitutional power-sharing for 138 multi-ethnic countries between 1945 and 2016, we find support for these arguments.
Power-sharing is a popular institutional prescription for divided societies. Multi-ethnic and post-conflict countries around the world have adopted it to establish or maintain political stability and democracy: from Belgium and Switzerland to post-apartheid South Africa, over Bosnia and Kosovo, to Nigeria and post-Hussein Iraq. In this article, we investigate the democratic merits of the two most widely-used institutional types of power-sharing. Thereby we provide critically relevant evidence on whether power-sharing institutions can fulfill these hopes pinned upon them.
We make two key arguments: First, in corporate power-sharing systems, which are based on rigid, ethnically-based institutions, the illiberal ‘face’ of power-sharing will be dominant and eclipse the democratic gains of increased inclusiveness. An example for this is Lebanon’s nominally democratic system, which is plagued by persistent neopatrimonialism and democratic shortcomings. Second, in liberal power-sharing systems, which rely on more flexible, electorally-based institutions and indirect incentives, this trade-off is less severe and gains accruing from higher inclusiveness will prevail. An example for this is Nigeria’s territorial power-sharing system, which has enabled gradual progress towards the regularized, peaceful alternation of power.
We argue that both types of power-sharing positively affect democracy through two pathways. First, they create political stability; second, they render democratic transitions acceptable for all parts of diverse societies.
However, we also identify negative consequences of power-sharing. We argue that these are more likely to accrue in its corporate subtype. First, corporate power-sharing is confined to specific segments; if other groups seek access to political power, they need to aim for radical institutional changes, which risks destabilizing the democratic system. Second, it politicizes individuals predominantly as group members and thereby is inherently in conflict with liberal democratic rules. Third, it relies heavily on compromises between segmental elites. Thereby, it risks deemphasizing mass participation and open competition and might prevent the emergence of a viable opposition. In contrast, liberal power-sharing makes use of institutions that can more flexibly accommodate new societal groups; moreover, it avoids locking in citizens permanently into their group identities.
We use new data to test these expectations. We measure power-sharing as an institutional model, as defined by Lijphart (1977), using a newly updated version of the Constitutional Power-Sharing Dataset (CPSD). This enables us to measure the degree of institutional power-sharing on the country-level and, a key innovation of our article, distinguish between its corporate and liberal subtypes for 138 ethnically diverse countries between 1945 and 2016 (see figures below). In order to avoid overlaps with power-sharing, we rely on a minimal concept of democracy, using both a new measure thereof based on NELDA and a conventional measure based on the Polity index.
The results conform to our theoretical arguments. They indicate that the liberal-pluralist type of power-sharing increases democracy levels, whereas the effects of the rigid, corporate type are mixed (see figure below). Our findings are robust to numerous specifications, including country-fixed effects and first-differenced dependent variables, and various operationalizations. They also hold if we use instrumental variables and matching procedures, which account for the fact that the most unstable countries may adopt power-sharing institutions in the first place.
In sum, our findings can be read as a cautionary tale for ‘institutional engineers’: to engineer a positive peace going beyond the absence of violence, they should exercise caution when employing rigid corporate forms of power-sharing. Whenever possible, a stronger reliance on flexible, liberal power-sharing rules seems to offer better prospects to develop democratic rule over time.
For detailed findings and methodology, please check out the article here.
The two f
Andreas Juon 