Fiseha Haftetsion 12-26-16
This brief paper deals with the impact of the electoral system on political participation in the Ethiopian Federation. The simplest form of the majoritarian electoral system, i.e. First-Past-The-Post (FPTP), is practiced at all levels in Ethiopia. The electoral system in place substantially influences the state of political participation in the Ethiopian Federation. The existing literature (see Yonatan 2009; Beza 2013) focuses on the election-related rights of people living out of their states of origin. Thus, the electoral system is not investigated from the angle of realizing the political pluralism and equitable representation of diverse political interests of the country as envisioned by the 1995 Constitution.
The government has recently announced that it has noticed weaknesses in the system and it will work to fix them. This paper aims to provide some explanations on the very common electoral systems, the drawbacks in the existing electoral system of Ethiopia, and the most feasible way forward. Showing the linkage between the electoral system and political participation in the Ethiopian context is the focal point. Whether the existing electoral system is the best option in helping realize the political objectives of the Constitution, as indicated above, will be analyzed. The most feasible electoral system for Ethiopia, in the eyes of the author, is provided towards the end.
The type of electoral system a country adopts is highly linked with the development of its party system, the nature of its executive, and the nature of the relationship between its executive and legislative branches of government (Lijphart 2008, 161). There are various types of electoral systems. However, the two broad categories are the plurality or majoritarian (also called the FPTP or winner-take-all) electoral system and the Proportional Representation (PR) electoral system. There are many varieties of the two electoral systems. There are also countries that follow the so-called mixed systems by either creating a system that combines some features of the two main systems or implementing the two systems in parallel. More is provided about the mixed system in the last section.
Within the majoritarian or plurality system, we have the FPTP, second ballot or majority run-off, and the Alternative Vote (AV). According to the FPTP, which is the simplest form of the majority system, the country is divided into single-member electoral districts and the candidate who wins the most votes in an electoral district is declared the winner (Reynolds et al. 2005, 35). In this case, the voters elect only a single candidate and the election is conducted only once. Thus, theoretically speaking, a candidate who secures only a couple of votes can be declared a winner if none of the other candidates secures more votes (see Norris 1997, 301). The second ballot or the majority run-off refers to an electoral system whereby votes are cast for the second time if none of the candidates secures a certain, usually an absolute majority or fifty plus percent, of the votes cast in an election (see Norris 1997, 302). In this case, similar to the plurality system, the voters elect for a single candidate and the candidates who participate in the second round can be either the top two in the first round or more than two candidates who secured a certain percentage of the votes cast in the first round. On the other hand, the alternative vote refers to a system whereby, unlike the above two, the voters are allowed to indicate preferences in addition to indicating their most favored candidate in a single-member electoral district (Reynolds et al. 2005, 48). In this case, the candidate who secures an absolute majority of the first preferences is automatically declared a winner. However, if there is none who secures an absolute majority of the votes in the first round, the votes obtained by the candidate who secured the lowest first preferences are distributed to the others depending on the second preference. The process continues until a winner with an absolute majority of the votes is obtained.
The key idea in the PR system is that seats should be allocated in proportion to the votes secured by parties (Reynolds et al. 2005, 57). The two common PR systems are the List PR and the Single Transferable Vote (STV). Under the List PR, parties present the list of candidates and the voters vote for a party. The List can be either open or closed. If the List is open, voters are given the chance to influence the rankings; if the List is closed, party officials do the ranking of candidates and voters do not have the power to influence the rankings (Reynolds et al. 2005, 84). The parties are allocated with seats in proportion to the votes they won by allocating seats to candidates according to the ranking in the List. Under the STV system, voters elect and indicate order of preference in multi-member electoral districts (Reynolds et al. 2005, 76). Because voters can indicate order of preference, STV looks similar to AV. However, the two are different. AV is a majoritarian system in which the electoral districts are single member and the candidate who wins absolute votes is elected whereas STV is a PR system whereby the districts are multimember and a candidate needs to secure only a minimum percentage (threshold) of the votes cast in order to secure a seat. Votes extra to the required quota are also redistributed to other candidates based on order of preference until sufficient candidates are elected.
As discussed in the forthcoming section, the two systems have their own advantages and disadvantages. The Mixed Systems, therefore, are aimed at remedying the defects of one and at capturing the advantages of the other. A mixed system may, for e.g., run both systems in parallel for electing members of a lower house. Alternatively, a Mixed System can employ the PR system to remedy loses of seats because of the FPTP system used. However, it is argued that the advantages and disadvantages of each system are better evaluated not in the abstract but considering a given context (see Yonatan 2009, 334). Here under is a brief discussion on the comparative advantages and disadvantages of the two systems.
Although there are scholars who insist that one is always better than the other (see, for e.g., Lijphart 2008), it is obvious that context matters a lot in designing the most relevant electoral system for any country. To begin with, one of the often-cited advantages of the majoritarian system is that it tends to create few majority parties more often than not; one with rightist and another one with leftist tendencies (Lijphart 2008, 161). This has the advantage of removing small extremist parties from the political scene. Even if extremist parties manage to secure seats in the legislature, they will be less likely to put serious impediments on governance efficiency. Moreover, the majoritarian system better ensures stable governance as it usually results in a single party forming governments without the need to resort to coalitions, which are often fragile (Norris 1997, 304). The majoritarian system allows voters to not only choose parties but also individual candidates. Thus, it is better in establishing direct association between candidates and voters. The majoritarian system is also the simplest electoral system for election authorities to administer it and for voters to understand it.
Despite the foregoing advantages, the majoritarian system has many disadvantages. Primarily, it tends to exclude and disenfranchise smaller parties (Norris 1997, 301). The latter in its turn heavily reduces minority representation in government. Moreover, since the majoritarian system does not guarantee that the seats in the legislature will be proportional to the votes cast, it results in votes’ wastage (Reynolds et al. 2005, 43).
The PR system, on the other hand, has the following advantages. It more or less translates the votes cast to seats in the legislature (see Norris 1997, 310), thus there is less votes’ wastage. More importantly, the PR system is believed to perform better in ensuring minority representation in government (see Moser 2008, 273). This means that there will be less possibility of minority exclusion and disenfranchisement in the PR system. Since, usually, votes are pooled at the national or regional level, PR encourages soliciting votes from everywhere. This encourages parties to address wider audience. Since there is a less possibility of switching from the right to the left or vice versa because of changes in party controlling government power, PR systems tend to ensure better stability and continuity in terms of policymaking and policy implementation (Reynolds et al. 2005, 58). PR further makes power sharing and coordination among parties and interest groups more feasible than in the case of the majoritarian system (see Reynolds et al. 2005, 58).
However, the PR system has also its own drawbacks. This system often produces coalition governments with the accompanying risk of breakup at any time (see Reynolds et al. 2005, 58). Moreover, because of high possibility of parties with ideological differences forming a government, it may be very difficult to implement a certain policy. At least theoretically, it can be argued that this reduces government efficiency. Although we saw earlier that because votes are pooled, parties are encouraged to appeal to a wider audience, there is a high possibility for the fragmentation of the party system. This is because any interest group regardless of the number of people sharing its goals is sure that it can manage to have a seat in the legislature for just winning, for example, only five percent of the vote. Thus, not only does this create fragmentation but also it rewards extremist parties. It also tends to disproportionately empower smaller parties because of the high possibility that coalition governments will often have to include such parties. PR system’s rules are more complex than those of the majoritarian ones are. This makes implementing the system relatively difficult (Reynolds et al. 2005, 59). Furthermore, PR system does not establish strong links between legislators and their constituencies as the voters are not empowered to elect individual candidates. Particularly in closed List PR, the party headquarters determines who may be elected and thus top party officials are very powerful in the case of PR systems.
As mentioned earlier, it is more concrete to assess the advantages and disadvantages of the different electoral systems based on contexts. What is clear, however, is the fact that electoral systems determine the nature of politics in a given country. Thus, each country must carefully design an electoral system that helps it achieve its political goals better. It is argued here that the electoral system a country adopts has a bearing on political participation. Let us analyze which system may better serve the political goals envisioned by the 1995 Ethiopian Constitution and the implications of the existing electoral system on political participation in the Ethiopian context.
4 The implications of the majoritarian electoral system on political participation
According to Article 9(3) of the 1995 Constitution, assuming government power in any manner other than what is provided in the Constitution is prohibited. It is stipulated that a political party or a coalition of political parties that has the greatest number of seats in the House of Peoples' Representatives (HPR) should form the executive and lead it (Art. 56 of the Constitution). The Constitution further stipulates that there should be a neutral electoral board whose members are appointed by the HPR upon the recommendation of the Prime Minister (Art. 102). Article 54(2) of the Constitution and the Ethiopian amended electoral law, on the other hand, unequivocally confirms that the system at work is the simplest form of the majoritarian system i.e. the FPTP. Accordingly, a candidate who receives more votes than any other candidate is declared the winner and all constituencies are single member constituencies (Art. 25 cum. 28(3) of Procl. 532/2007). Furthermore, according to the law, it is not mandatory for an Ethiopian to participate in elections (Art. 76(3) of Procl. 532/2007).
It is indicated, in previous works, that the simple FPTP electoral system adopted by Ethiopia since the inauguration of the 1995 Constitution has fared unsatisfactorily in representing people who live out of their states of origin (see Yonatan 2009, 337-8; Beza 2013, 99). Hence, regardless of their national backgrounds and for election purposes, these people are considered a minority and the existing electoral system is criticized for not ensuring their representation. When seen against the preceding theoretical discussions and considering the Ethiopian situation it can be observed that the PR system could have given them more representation and hence more voice in both the national and state-level legislatures. Thus, the criticisms are valid.
However, it is also critical to examine the relevance of the majoritarian system from a wider perspective and from the political objectives envisioned by the 1995 Constitution. It is apparent that in such diverse countries as Ethiopia there are diverse political interests. To ensure long-term stability, not only the governance system but also the mechanism to access government power should be as inclusive as possible. The system should as far as possible avoid winner-take-all scenarios.
Nevertheless, it is wrong to understand the Ethiopian way of accessing power as a pure adversarial Westminster model. Indeed, it reflects some features of the consociational model. The key features of the consociational model are: “(1) grand coalition governments that include representatives of all major linguistic and religious groups, (2) cultural autonomy for these groups, (3) proportionality in political representation and civil service appointments, and (4) a minority veto with regard to vital minority rights and autonomy” (Lijphart 2008, 42).
The ruling party, the EPRDF, is a coalition of parties that represent four major states of the Federation: Oromia, Amhara, SNNP, and Tigray. Different parties that are affiliated to the EPRDF rule the remaining states. Moreover, de facto, different national and religious groups are represented in the cabinet almost in proportion to population number. Since the consociational model envisions grand coalition government encompassing different parties with different ideologies, the Ethiopian situation does not fulfill the first requirement. However, it should be noted that the party system attempts to represent all the national groups.
The Ethiopian national groups have cultural autonomy. In fact, according to the 1995 Constitution, they have the right to a full measure of self-government, which includes the right to establish institutions of government in the territory that they inhabit, and to equitable representation in the Federal and state governments (Art. 39(3)). Guaranteeing cultural autonomy is considered as one of the most visible successes of the federal system in place (see Alemseged 2004, 604). This fulfills the second requirement.
According to the consociational model, the electoral system should normally be the PR system (Lijphart 2008, 48). This is therefore the weakest link of the Ethiopian system in the eyes of the consociational model. Nevertheless, the Ethiopian national groups are in overwhelming majority of the cases territorially concentrated and it can be argued that the system does not exclude any national group from being represented in different tiers of government (see Yonatan 2009). Although not formally very well regulated, practice indicates that there are attempts to diversify the civil service and include the representatives of different national, religious, and gender groups. This partially fulfills the third requirement.
Finally, every nation, nationality, and people is entitled to full measure of self-administration including the right to secession and hence a national group can veto the federal arrangement. This fulfills the fourth requirement. It can be seen that the governance system as a whole does not squarely fit into the Westminster (two party- majoritarian) model. Indeed, the Ethiopian Federation depicts some consociational features. The forthcoming discussions, therefore, aim at highlighting the problems associated with adopting the majoritarian electoral system.
Despite the fact that accommodation of diversity is the founding principle of the 1995 Constitution, one can observe that the electoral system in place is not the best in terms of fulfilling this objective. Theoretically speaking, according to the existing law, a candidate who obtains a single vote can be elected to the legislature if none of the other competitors manages to obtain one. This is not only very wasteful in terms of valuing votes but also has potential negative political consequences.
The experience of the Republic of South Africa with its first democratic election in 1994 is instructive here. Before 1994, in South Africa, the electoral system at work was the FPTP. The ANC was by far the largest party with a higher possibility of obtaining more seats under the majoritarian system (Reynolds 2005, 63). However, in the interest of long-term stability and inclusiveness, the ANC, in its first electoral participation, agreed to change the electoral system to PR thereby enabling smaller parties secure more seats in the National Assembly than they could have obtained under the majoritarian system (Reynolds 2005, 63).
Let us see the main issue at hand: analyzing the impact of the majoritarian system on political participation in the Ethiopian context. As mentioned earlier electoral systems should not be evaluated ‘in abstract’ (see Yonatan 2009, 334). In the Ethiopian case, therefore, it can be, so far, observed that the majoritarian system has strengthened the concentration of power in the hands of a dominant party. The more power is concentrated in a single or a coalition party, the higher the possibility that the party system bypasses the constitutional power demarcations. Although the electoral system is not the sole factor, it is apparent that it has contributed to the disproportionate concentration of power in a single party. Cases in point are the 2010 and 2015 countrywide elections. In the 2010 election, the EPRDF and its affiliates secured all the seats in the HPR except for two seats that were won by an opposition member and an independent candidate. In the 2015 election, the EPRDF and its affiliates won all the seats in the HPR. Following this, in response to those who doubted the democratization process in Ethiopia and those who challenged whether the election was up to the standards, it was stated officially that the votes cast were not as uniform as the legislature looked like. Indeed, it was the case. This implied that it was the electoral system in place that should be blamed. The most important point is that, in such cases, there is a high possibility that whatever proposed by the dominant party may easily become a law without enough scrutiny since there is very high uniformity of voices in the legislature (see also Assefa 2015, 258-259). This can have a serious impact on long-term political stability.
The arguments of writers such as Yonatan (2009) and Beza (2013) that the existing electoral system is good enough to safeguard the representation of territorially concentrated national groups is at least contestable since it ignores the issue of internal diversity and political pluralism. It is normally expected that national groups have internal political diversity and different political parties who claim to represent such interests and the PR system is by far better in accommodating the representation of such internal diversities.
The better internal diversities are accommodated, the higher the possibility that state-level or local-level legislatures are more diverse. This results in relatively stronger legislatures that can debate laws better; with a consequent higher possibility of bringing political stability. In other words, diverse legislatures, which are highly likely in PR systems than in majoritarian systems, are in a better shape to ensure the representation of diverse political opinions. The latter quality in its turn boosts the potential for institutional vibrancy; creates opportunities for balanced exercise of power by both the federal government and the constituent units. On the other hand, the majoritarian systems tend to create 'an elected dictatorship' (Norris 1997, 311). However, we saw that the PR system has weaknesses, too. Therefore, there is a need to look for a system that combines the strengths of both systems.
5 The way forward
It is essential to reemphasize that the electoral system Ethiopia may adopt should be one that combines the strengths of the majoritarian and the proportional systems and, at the same time, that minimizes the weaknesses of the two. Moreover, given the literacy level of the society and the level of development of the country, the electoral system should be as simple as possible. A complicated voting system, however fair it is, may not be suitable or efficient.
This author is of the opinion that shifting to pure PR system is not the ideal option. This is because of the disadvantages of the PR system discussed earlier. The preferred option is the Mixed System. Within the mixed system, there are two options. One of them is called the Mixed Member Proportional (MMP) system and the second one is called the Parallel System. In this system, the Majoritarian and the PR systems are employed at the same time in a manner that the PR system is used to compensate seats lost due to the majoritarian system (Reynolds et al. 2005, 91). The PR system employed here can be closed or list PR. According to the Parallel System, the two are employed independently and simultaneously to elect representatives (Reynolds et al. 2005, 91). The first option tends to bring proportionality while also keeping some of the advantages of the majoritarian system such as establishing direct link between voters and candidates. The Parallel System tends to be disproportional, as the PR system is not used to compensate seats lost because of the Majoritarian System.
The most feasible option for Ethiopia, the author argues, is the MMP electoral system. The country now is divided into 547 electoral districts. The Constitution stipulates that the seats in the HPR should not exceed 550 and among them at least 20 seats should be reserved to minority nationalities and peoples (Art. 54(3) of the Constitution). If the electoral system is to be changed to the MMP, since the Constitution has limited the number of the seats, there is a need to restructure the number of the electoral districts. The number of seats provided in the constitution should be divided into two categories: one category that is to be filled by the majoritarian system and another one that is going to be allocated proportionally. The amount of seats reserved to the two categories differs from a country to another. In Germany, the two categories are given equal weight (50% of the seats in the Bundestag are reserved for each category). In Italy 25% of the seats are reserved to the PR system and 75% to the Majority System; in Bolivia it is 48% to 52%; in New Zealand 46% to 54% respectively (Reynolds et al. 2005, 91). Countries that allocate fewer seats via the proportional mechanism tend to lack seats that can compensate disproportionality that is created because of the majoritarian system (Reynolds et al. 2005, 91).
The author is of the opinion that both the Majoritarian and the PR Systems should be given equal weight in the Ethiopian MMP system. This makes it necessary to restructure the electoral districts of the country. The districts can be restructured to around 275 while maintaining the Constitution's stipulation that the seats should not exceed 550. Accordingly, 275 of the seats can be filled according to the Majoritarian System and the remaining seats can be allocated based on the PR system. Restructuring the districts to 275 will create a situation where hundreds of thousands of people are represented in a single district (on average close to 400,000.00). However, some minorities may fail to fulfill this threshold and they may not be represented in the HPR. This defect can be remedied by expanding the seats reserved to minorities further (probably, up to fifty depending on the number of numerical minorities). According to this arrangement, a party that won 200 out of the 275 seats according to the plurality system but that have won only 60% of the total votes will get 330 seats in the 550 seat HPR.
This is how it works: if calculated proportionally, a party that won 60% of the votes obtains 330 seats (60/100*550). Since the party has already secured 200 seats through the Majoritarian System, it will be awarded additional 130 seats. The same goes to all the parties that participated in the election. To discourage the development of fragmented party system, minimum threshold can be introduced i.e. only parties that won a certain amount of votes are entitled to compensatory seats. This writer is of the opinion that the parties that should benefit from the compensatory seats should be those parties that won, at least, 3% of the total votes or those parties that won at least three seats in a normal district. Normal district in this sense is a district that is not designed to ensure the participation of numerical minorities. The common experience elsewhere suggests that the parties should win either 5% of the total votes or a seat through the Majoritarian System to be eligible to compensatory seats via the PR System. In the Ethiopian case, given the higher population number, setting the minimum threshold at 5% may result in votes wastage, thus it is preferred to set it at around 3%. The other optional requirement is for a party to win in at least three electoral districts. This is introduced to protect the system from over representation of minority nationalities and peoples. If the requirement is reduced to, for e.g., winning a single district, minority parties that won a seat in a district that was designed to ensure minority representation will get the chance to compensatory seats. This will inflate the overrepresentation of minorities further.
It was mentioned earlier that the electoral system should be as simple and clear as possible. One of the issues that determines this factor is the ballot system. In the case of MMP, the electorate can cast their vote either in one or two ballots. In a single ballot system, voters indicate their preferred candidate and party in a single paper. In a two-ballot system, the voters indicate their district level vote and party (countrywide) vote in separate papers. The two have their own advantages and disadvantages. Separating the ballot for district level and party preferences gives the electorate the chance to split their preferences. However, in countries like Ethiopia where literacy rate is low, it can be complicated for the voters. Introducing a single ballot, whereby the voter indicates his/her preferred candidate and party in a single paper, reduces the inconvenience. A single ballot is, therefore, more convenient for Ethiopia.
The electoral system in place has a direct bearing on political participation in the Ethiopian Federation. As per the discussions in this paper, the majoritarian electoral system practiced at all levels throughout the country is not the best in terms of realizing the founding principles and the explicit objectives of the 1995 Constitution, inter alia, accommodation of diversity. A change in the current electoral system will therefore have a direct bearing on the state of political participation in the country. In order to ensure that all significant and genuine political voices are heard, the existing electoral system should be changed. On 10 October 2016, the President of the FDRE, in his annual opening speech at the joint meeting of the HPR and the HoF, confirmed that the government has noticed the loopholes in the existing electoral system. He further announced that, in the years ahead, the government will work to ensure that diverse political voices of the people of the country are represented. He promised changes will be introduced starting in the existing era of parliament without waiting until the next countrywide election scheduled to take place in the year 2020. His speech indicates that the existing electoral law will be changed soon.
No one can be certain about the exact changes to be introduced by the government but the move is definitely towards adopting either a mixed or a proportional system. As the discussions in this paper show, a change that introduces some elements of proportional representation to the existing electoral system will result in legislatures that are more diverse. The difference is almost automatic. Practice shows that the votes of the electorate are diverse. If the system is redesigned in a manner that accommodates such votes, it will result in at least diverse legislative bodies. Diverse legislatures will highly likely result in better checks and balances. Better checks and balances among branches or spheres of governments, on the other hand, result in better popular satisfaction in the governance system, which is essential for political stability.
This author is of the opinion that a complete shift to the PR system is not the ideal option. This is because of the weaknesses of the PR system. A mixed system that combines the strengths of the PR and the Majoritarian system is the better option. The mixed system should be the Mixed Member Proportional (MMP) system whereby the PR system is employed to compensate disproportionality created by the Majority System. This calls for redesigning the electoral districts of the country to 275 (from the current 547). This will result in increased number of people represented in a single district (up to 400,000.00). To ensure their representation, the seats reserved for numerical minorities at the HPR should be expanded further depending on the number of the numerical minorities (probably up to fifty). This requires amending the constitution and the existing election-related laws. The constitutional amendment, in this writer's opinion, should focus on setting principles of election in Ethiopia. Details regarding technical issues should be provided in the Electoral Proclamation and other laws.
Alemseged Abay (2004). Diversity and State-building in Ethiopia. African Affairs 103 (413): 593-614.
Assefa Fiseha (2015). 'Legislative–Executive Relations in the Ethiopian Parliamentary System: Towards Institutional and Legal Reform' in Jaap de Visser, Nico Steytler, Derek Powell, and Ebenezer Durojaye (eds.). Constitution-Building in Africa. Baden-Baden: Nomos.
Beza Dessalegn (2013). The Right of Minorities to Political Participation under the Ethiopian Electoral System. Mizan Law Review 7(1): 67-100.
Ethiopia (1995). Constitution of the Federal Democratic Republic of Ethiopia. Negarit Gazeta, Proclamation No. 1/1995.
Ethiopia (2007). The Amended Electoral Law of Ethiopia. Negarit Gazeta, Proclamation No. 532/2007.
Lijphart, Arend (2008). Thinking about Democracy: Power Sharing and Majority Rule in Theory and Practice. London: Routledge.
Moser, Robert G. (2008). Electoral Systems and the Representation of Ethnic Minorities: Evidence from Russia. Comparative Politics 40 (3): 273-292.
Norris, Pippa (1997). Choosing Electoral Systems: Proportional, Majoritarian and Mixed Systems. International Political Science Review 18(3): 297-312.
Reynolds, Andrew (2005). ‘South Africa: Electoral Systems, Conflict Management, and Inclusion’ in Andrew Reynolds et al. (eds.). Electoral System Design: The New International IDEA Handbook. Stockholm: International Institute for Democracy and Electoral Assistance.
Reynolds, Andrew et al. (2005). Electoral System Design: The New International IDEA Handbook. Stockholm: International Institute for Democracy and Electoral Assistance.
Yonatan Tesfaye Fessha (2008). Institutional Recognition and Accommodation of Ethnic Diversity: Federalism in South Africa and Ethiopia. PhD Dissertation, University of the Western Cape.
[*] By Fiseha Haftetsion. The piece reflects the personal opinion of the writer only. Hence, it is not in any way related to institutions or organizations the writer is affiliated with. This is a brief summary of the main issues of a comparatively broader work in progress and the writer highly welcomes comments. He can be reached at: firstname.lastname@example.org