Ethnic Based Party Systems and Democratic Stability: Implications for Ethiopia

(Abdissa Zerai, PhD)

According to Edmund Burke (as cited in Sartori, 2005, p. 8), a political party is understood as “a body of men (and women) united, for promoting by their joint endeavors the national interest, upon some particular principle in which they are all agreed.” As noted by Lupu (2015), although political parties are rarely mentioned in constitutions, there is not a single modern democracy that does not have them. Indeed, it has long been noted that they are indispensable to mass, representative democracy. In fact they are said to be a prerequisite for modern elections.

In a modern society where the population of one electoral constituency runs into hundreds of thousands, political parties perform the function of mobilizing the electorate (Ujo, 2001). What is more, political parties resolve collective action and social choice problems in legislative decision-making (Aldrich, 1995). They serve as heuristic/experiential devices and bundle/push policy dimensions for voters who lack the time or inclination to learn about every issue and every candidate (Popkin, 1991). They structure electoral competition with a stable menu of options that makes uncertain elections more predictable and extends the time-horizon of politicians (Mainwaring and Scully, 1995). And they discipline politicians and hold them to account even when voters cannot (Alesina and Spear, 1988). Political parties are also responsible for creating alternative or competitive ideologies that cater for divergent electorate and thereby provide a framework for representation.

As Horowitz (1985, p. 296) notes, the main characteristic of a political party is its quality of mediating, of comprising a combination of interests. Referring to Huntington, Horowitz (1985, p. 297) argues, “parties become the buckle which binds one social force to another and which creates a basis for loyalty and identity transcending more parochial groupings.” Along a similar line, Sigmund Neumann (as cited in Horowitz, 1985, p. 297) describes a party as “the great intermediary which links social forces and ideologies to official governmental institutions and relates them to political action within the larger political community.” To be a party, Sartori (as cited in Horowitz, 1985, p. 297) contends that a political organization must be “capable of governing for the whole”, and this means it “must take a non-partial approach to the whole.” By this he means that the party must attempt to serve the “public interest” or the “common good” and “organize the chaotic public will.”

However, the conceptualizations and features of parties provided above tend to be more in sync with the characteristic features of conventional civic-oriented political parties. But political parties can also be ethnic rather than civic in their orientation. If that is so, how does ethnic-oriented party differ from civic-oriented party?

Ethnic-oriented parties and party systems

According to Dowd and Driessen (2008, p. 1), an ethnically based party is defined as a party which portrays itself as the champion of a particular ethnic group or category to the exclusion of others and makes such a strategy central to its strategy to mobilize voters. On the other hand, they conceptualize ethnically dominated party systems as party systems in which all or most of the major political parties are ethnically based. From the perspective of Dowd and Driessen (2008, p. 1), in ethnically dominated party systems, parties are not distinguished from each other based on what they represent but rather based on who they represent. There are few if any policy differences between parties or, if there are policy differences, few people could tell what these differences are.

Problems associated with ethnic based parties

The conventional association of a political party with the quest for public rather than particularistic interests is inimical to the very basis of an ethnic party. This is because an ethnic party identifies narrow group interests with the totality of the common interest, owing to a real or perceived incompatibility of those interests (Horowitz, 1985, p. 297). Ethnic or particularistic parties behave much like interest groups or pressure groups rather than political parties. As such pressure groups monopolize the loyalty of their members, intergroup cleavages are widened and deepened. And the growth of groups dedicated to the promotion of narrow group claims places greater strain on the social mechanisms for the settlement of group conflict. Thus, ethnic parties make the mediation of group interests difficult, and this helps explain why ethnic party systems are so often conflict prone (Horowitz, 1985, p. 298).

What is more, Scholars (Rabushka and Sheplse, 1972; Horowitz, 1985, as cited in Dowd and Driessen, 2008, P. 3) argue that

…when most major ethnic groups have their own political parties, ethnic minorities fear permanent domination, and ethnic majorities are thus continually focused on devising ways to co-opt or exclude rival ethnic groups. When and where ethnicity and race are politicized and that politicization is institutionalized in the party system, the political landscape becomes frozen along the ethnic dimension. Ethnic minorities fear permanent exclusion and, within parties, elites compete for leadership by outdoing one another in proving their loyalty to their respective ethnic groups and stoking fears of government dominated by rival ethnic groups. Consequently, the likelihood of violent conflict increases and the prospects for good governance are lowered where the party system is ethnically dominated.

Along similar lines, Sisk (1996, as cited in Dowd and Driessen, 2008, p. 3) contends that ethnically dominated party systems lower the quality of democracy because they limit citizens’ electoral choices to members of their own ethnic groups. Those who suggest that it may be wise to vote for a candidate supported by a party other than the one claiming to represent their own ethnic group are accused of being traitors or harassed. This might be especially true where region and ethnicity overlap, as they do in many deeply divided societies.

There are also those who suggest that the quality of democracy is lower where there are ethnically dominated party systems because politicians are focused more on the interests of their respective ethnic groups than on the needs of the country as a whole. For instance, it is noted in Dowd and Driessen (2008, pp. 3-4) that

…when ethnic identities are indeed politicized, elites are expected to win the maximum rights and privileges for their ethnic groups. Where there are ethnically dominated party systems, one’s ethnic group is either in or out of power. If there is more than one established ethnic party, other ethnic groups are themselves compelled to mobilize along ethnic lines in the hopes of gaining some influence in national politics, thereby fueling an ethnic crowding out effect …. In developing country settings, gaining access to the state becomes more pressing for survival and ethnicity becomes an ever more indispensable political instrument. Politically institutionalized competition for the rights to these resources increases the likelihood of political instability, violent conflict, and the demise of democracy.

Dowd and Driessen (2008, p. 4) went on to argue that once ethnicity becomes politicized, the electoral choices with which citizens are presented are essentially limited to the ethnic plane. What appears to be multiparty democracy at the national level, could really be thought of as a collection of illiberal one-party ethnic states at the sub-national or regional level. According to Dowd and Driessen (2008, p. 4), “Even though there may be several national political parties, where there is an ethnically dominated party system and ethnic groups are geographically concentrated, parties essentially do not compete with each other for support at the local level. Each party has its ethno-regional bailiwick [jurisdiction.]”

One may logically argue that since the presence of ethnic parties does not prevent the formation of national parties that can cater for a broader constituency, ethnic parties can effectively be challenged by organizing such national parties. But as Dowd and Driessen, 2008, p. 4) notes,

There may be no legal constraints that prevent the formation of new parties that are less ethnically based, but politicians, once in power, control access to scarce and valuable resources and use these resources to prevent other politicians from appealing to anything but ethnicity. Further, politicians who have built successful political careers by appealing to ethnicity may collude to prevent other politicians from appealing to anything but ethnicity. Individual political identity is thereby reduced to ethnic identity as citizens essentially delegate their vote to entrenched ethnic elites. If this argument is correct, citizens where there are many political parties that are ethnically based are subject to the narrow choices and abuses of power that characterized previous single-party authoritarian regimes.

In summary, it can be argued that an ethnically dominated party system is ill-equipped to promote the public interest or the common good; it is likely to limit electoral choices by producing parties that appeal to segmented constituencies; it could foster ethnic outbidding and thereby generate and/or exacerbate group conflict; it would not only reduce the quality of democracy but also would be less likely to effectively deliver public goods. Furthermore, although such a system might produce stable parties, it might also generate unstable politics.

Implications for Ethiopia

As we all know, since the demise of the military regime and the coming into power of the EPRDF 1991, the Ethiopian political landscape has been dominated by ethnic based party system. In a deeply divided society like ours, the emergence and prominence of such a party system can be attributed to what various scholars call the dynamics of supply and demand. Whereas the supply side has to do with the decisions made by elite politicians to play the ethnic card in reaction to institutional incentives that make appeals to ethnicity more effective when trying to win political power, the demand side has to do with the importance of ethnic identity in people’s lives. After more than two decades of experimentation, we have got to the point where we are now.

As we frantically look for the way out of the current crisis and ponder the way forward, some are calling for a clean break from ethnic-oriented politics and embracing civic-oriented politics, whereas some others are cautioning not to throw the ‘baby with the bathwater.’ In fact with all the pitfalls of ethnic based party systems discussed above, the first impulse is to give credence to the first proposition. However, a serious question one has to first answer is that with the current heightened sense of ethnic consciousness and deep-rooted social divisions, can we do away with ethnic politics at once in favor of civic-oriented politics? What is the best way to do that? Should we do this by undertaking a constitutional amendment that bars establishing ethnic based political parties? Would that be feasible, or even, desirable? As appealing as the proposition might sound, the ‘how’ of it does not appear to be a straightforward one.

Since politics is the art of the possible rather than the art of the ideal, the alternative question might be the following: would it be wise to look for a better way of structuring multi-ethnic parties that would help us contain the deleterious effects of ethnic parties and serve as a stepping-stone for a gradual transition to a civic-oriented party system? This is the question I would like to touch upon in the sub-section below:

According to Elischer (2008, p. 8), ethnic based political parties are of three types: the mono-ethnic, the multi-ethnic alliance and the multi-ethnic integrative parties.

The mono-ethnic party promotes the interests of one ethnic group and openly appeals to this group to unite politically under its party banner. Accordingly, in terms of its leadership composition, it is homogenous. All its leading figures come either from one ethnic group or from ethnic groups with cultural similarities. Consequently, its party factions do not represent ethnic factionalism but rather generational differences… or power struggles between individuals within the party. The party’s ability to field candidates is restricted to the area(s) where the group it represents is located.

As noted by Elischer (2008, p. 9), there are two types of multi-ethnic parties. These are the multi-ethnic alliance and the multi-ethnic integrative party, both of which represent various ethnic communities.

The two basic differences between the two multi-ethnic types are their respective motivation and degree of internal stability: The alliance type corresponds to the logic of Donald Horowitz’s coalitions of convenience and coalitions of commitment. A coalition of convenience is formed with the sole motivation of gaining a parliamentary majority. Coalitions of commitment indicate more amicable relations, yet anticipated gains on election day again provide the key impetus for communities to unite. Driven by such strategic considerations, the alliance type makes extensive use of “ethnic arithmetic”: It tries to include as many groups as necessary in order to secure electoral victory. Consequently, these types of ethnic coalitions prove to be internally fragile and short-lived. By contrast, the integrative type’s purposes transcend election day. This corresponds to Horowitz’s logic of permanent coalitions of ethnic parties.

Arguing further, Elischer (2008, p. 9) notes that the integrative party aims to form a long-lasting political force in which two conditions are fulfilled: Firstly, it bridges its country’s dominant ethnic cleavages (past or present) by incorporating influential community leaders from both sides of the cleavage into its leadership structure. Secondly, the integrative party is formed long before election day and survives electoral defeats as well as leadership contests without any major changes (splits and mergers) in the groups that make up the party. By staying together as a united political force, it demonstrates that it has overcome the divisive logic of ethnic arithmetic. What is more, within the integrative party, argues Elischer (2008, p. 9), ethnicity no longer leads to exclusion; thus, the party no longer represents an “ethnic party” as conventionally understood in the literature and as represented by the mono-ethnic or multi-ethnic alliance party. Both the multi-ethnic alliance and multi-ethnic integrative parties have the capacity to field candidates nationwide. However, the alliance type might focus on its “home regions.”

After discussing the three types of ethnic parties, Elischer (2008, p. 8) is of the opinion that “Given the diverse social fabric of African societies, Africa’s lack of an industrial revolution, and the continent’s more or less limited ability to design social welfare policies, the integrative party is instead the closest an African party gets to the model of the catchall party found in Europe and the United States.”

When we look at the ethnic based political parties that currently exist in Ethiopia, we can see that they are either mono-ethnic parties or multi-ethnic alliance parties, and there are no parties that can qualify as a multi-ethnic integrative party. As it pertains to Ethiopia, I found Elischer’s argument persuasive and relevant given the fact that more than eighty (80%) percent of the Ethiopian society lives in the countryside in small and tight communities where communitarian ethos outweigh liberal values and ethnic attachment is much stronger. And any political party that aspires to political power at the national level cannot achieve its aspiration without a significant support from such a constituency. Thus, at this stage, it seems prudent that we encourage the formation of multi-ethnic integrative parties in order to blunt the destructive edge of ethnic nationalism and lay a ground work for a gradual exit to a more civic-oriented politics. Obviously, this does not preclude the formation of broad-based national or catchall parties concurrently, as such parties are important; but as their constituency might be largely limited to urban areas (which constitute between 18-19% of the electorate), their chances of becoming a governing party would be limited when compared to multi-ethnic integrative parties. Otherwise, there is no doubt that their role would be crucial, especially in the urban centers, and, hence, their formation ought to be encouraged. Overall, however, at this critical juncture, it appears to me that focusing on the formation of multi-ethnic integrative parties is the less ideal but the more practical and possible course of action we might consider taking as we plan our next step.


Aldrich, John H. (1995). Why Parties? The Origin and Transformation of Political Parties in America. Chicago: The University of Chicago Press.

Alesina, Alberto, and Stephen E. Spear. (1988). An Overlapping Generations Model of Electoral Competition. Journal of Public Economics 37 (3): 359-79.

Dowd, Robert A. and Michael Driessen. (2008). Ethnically Dominated Party Systems and the Quality of Democracy: Evidence from Sub-Saharan Africa. Afrobarometer Working Paper no. 92.

Elischer, Sebastian. (2008). Ethnic Coalitions of Convenience and Commitment: Political Parties and Party Systems in Kenya. German Institute of Global and Area Studies Working Papers.

Horowitz, Donald L. (1985) Ethnic Groups in Conflict, Berkeley: University of California Press.

Lupu, Noam. (2015). “Political Parties and Party Systems,” in Routledge Handbook of Comparative Political Institutions, eds. Jennifer Gandhi and Rubén Ruiz-Rufino. London: Routledge, pp. 128-144 (2015).

Mainwaring, Scott, and Timothy R. Scully. (1995). “Introduction: Party Systems in Latin America.” In Building Democratic Institutions: Party Systems in Latin America, eds. Scott Mainwaring and Timothy R. Scully. Stanford, CA: Stanford University Press. 1-34.

Popkin, Samuel L. (1991). The Reasoning Voter: Communication and Persuasion in Presidential Campaigns. Chicago: The University of Chicago Press.

Sartori, Giovanni, (2005) Parties and Party Systems: A Framework for Analysis, Colchester: ECPR.

Ujo, A.A. (2001) Understanding Party Politics in Benue State, Nigeria, Kaduna: Anyaotu Enterprises and Publishers Nigeria Ltd.

* The writer, Abdissa Zerai (PhD), is a visiting faculty at the University of New Mexico, USA; and former head of the School of Journalism & Communication at Addis Ababa University. He can be reached at

Previous articles from this author:

Striking a Grand Elite Bargain on the Terms of Our Coexistence: The Fierce Urgency of Now

From Making Ethiopia Safe for Ethnic Diversity to Making Ethnic Diversity Safe for Ethiopia: The Ethiopian Political Elites’ New Challenge

View Comments (1)

Leave a Comment

This website uses cookies.