Volume 10, Issue 4
Alternative Electoral Systems and the 2005 Ethiopian Parliamentary Election
Abstract: What if an alternative set of electoral rules had been used to govern elections when an authoritarian regime introduces its first real competitive elections? Would this alter the trajectory of democratic transition, after the introduction of political competition? In this paper, I conduct a set of electoral simulations with different electoral systems using the results from the 2005 Ethiopian parliamentary election. Would the results have been different had something other than the single member district plurality system been employed in the 2005 election? Would the opposition parties have attained more seats and if so, how many more? I find that had certain electoral rules been employed (particularly the Block Plurality system), the opposition parties would have fared much better in the 2005 parliamentary elections, and this would have had an important impact affecting the course of events that immediately followed the 2005 election. This has important implications for the negotiations over the rules governing future Ethiopian elections.
As many scholars have noted, electoral systems exert a powerful influence on the process of democratization. Indeed first competitive or "transitional" elections are crucial moments for newly democratizing countries. Although these elections mark only a beginning point in an often arduous journey, their outcomes crucially affect the future course of democratic transition and democratic consolidation. Whoever wins the transitional election often has the opportunity to re-write the rules of the game to their advantage and, hence, significantly influence future political developments.
Would the rewriting of such electoral rules have changed the outcome of the first real competitive parliamentary election in Ethiopia in 2005? The 2005 election was a crucial moment in the history of the transition period in the country. First, it produced a real electoral opposition that could challenge the ruling Ethiopian Peoples' Revolutionary Democratic Front (EPRDF). Second, the results of the election set the stage for a violent confrontation between the opposition and the regime in the summer of 2005 that ultimately led to the "arrest" of further democratization. In part, the violent repression of the opposition on the part of the EPRDF was a direct consequence of the election itself. On the one hand the opposition parties had won enough seats to challenge the regime (especially in Addis Ababa and the region of Oromiya) but not enough seats across the country (even in those areas that were not marred by electoral fraud) to force the regime to recognize them as legitimate coalition partners in a new government. Could the results have been different if the 2005 election had been governed by a different set of electoral rules, such as the adoption of a list Proportional Representation system that the opposition had called for prior to the election?
This paper utilizes a set of electoral simulations using the 2005 Ethiopian parliamentary election results and asks the question of whether the results would have been different had different electoral rules been employed (as opposed to the single member district plurality system)? Would the opposition parties have attained more seats if some variation of Proportional Representation (PR) system or Mixed Member District (MMD) system (similar to Germany's) been employed, and if so how many more? This paper takes the existing electoral returns (using only the results from districts which were not in dispute following the 2005 election) and subjects them to a variety of electoral systems (national PR list, PR list aggregated to the regions, a Mixed Member District system and a Block Plurality System). Finally, it examines the possible impact of the use of alternative electoral rules on the course of post-election Ethiopian politics.
There has been a long tradition in the literature of simulating the effects of different electoral systems using existing electoral data in order to ascertain whether the use of such alternative systems might significantly alter the results. By re-running the national parliamentary elections using alternative electoral systems one can speculate as to whether changing the electoral formulae would have changed the composition of government and parliament. Much of the extant literature, however, has focused on electoral systems simulations in fully consolidated democracies in Europe, Asia, and Latin America, as opposed to systems in transition, or even "semi-authoritarian" regimes.
Assessing the effects of alternative electoral systems for countries in transition would provide important insights as to the potential effects of such systems on the course of democratization. Electoral results in founding elections are critically important because who wins such elections has the opportunity to fashion the constitutional order to advantage their political interests. As Andrew Reynolds notes "the concept of a 'loyal opposition' is a difficult one to entrench when one segment of society sees that losing an election is equivalent to being completely shut out of governmental power. whether to constitute parliament by a plurality, majoritarian, or proportional representation type electoral system, become(s) critically important to the prospects for democratic consolidation in a divided society."
This cultivation of a loyal opposition may be particularly difficult in a semi authoritarian regime, especially for what Steven Levitsky and Lucan Way refer to as "competitive authoritarian" regimes. Such regimes are not democratic in that incumbents routinely manipulate politics to their advantage. However, such regimes are also not full-scale authoritarian regimes. They do not eliminate formal democratic rules or reduce them to a mere façade or ban opposition parties altogether. Rather, incumbents are more likely to use bribery, co-optation, and subtle forms of suppression to "legally" harass, persecute, and extort cooperation. Further, in competitive authoritarian regimes opposition forces often pose significant challenges to the regime. So even though democratic institutions are seriously flawed in competitive authoritarian regimes, unlike in full-fledged authoritarian regimes, the incumbents must take the opposition seriously. Unlike "façade" electoral regimes which characterize full-blown authoritarian systems, in competitive authoritarianism, democratic institutions (such as electoral systems) offer an important channel through which the opposition may seek to pressure the regime into making democratic concessions. Indeed, such competition allows the potential for openings for the opposition to make significant inroads, at least enough to institutionalize their position, and from that point as a potential springboard for full-blown democratic transition. Whether that opposition remains "loyal" when an opening occurs may depend heavily on whether it "feels" included rather than excluded from the political process.
Electoral systems and democracy in Africa
There has historically been a rich literature on elections and political parties in Sub-Saharan Africa. Much of this past literature has centered on the formation of political parties in the shadow of decolonization and the early years of independence. Generally, the literature focused on either the emergence of ethnic parties (or parties that appeal to a particular ethnic group), which were often seen as divisive organizations, or on the emergence of nationalist parties that became the foundation for the later one-party states.
However, this past literature on African parties really did not tap into the existing theories on electoral systems and political parties developed in Europe, Latin America and Asia--- little effort was made to link the African experience with the broader literature on these topics. As the "third wave" hit sub-Saharan Africa, there was a sharp rise in interest in elections and political parties as more and more countries experienced political liberalization and began holding elections.
Since then, several studies have examined party development as a function of electoral systems in Africa. Much of this new literature however has focused either on electoral systems effects on the development of party systems or the electoral performance of individual parties.
Perhaps some of the best recent work (and most relevant given the focus of this paper) is that of Andrew Reynolds. In particular, in his 1999 book he analyzed the election results in the five countries (South Africa, Namibia, Zambia, Zimbabwe, and Malawi) examining the actual results as well as alternative results under a variety of different electoral systems simulations. Arguing that political institutions are important in affecting the development of an "inclusionary" or "exclusionary ethos" in the first competitive elections (and that this "ethos" is crucial in explaining the stability of the democratization process) he reruns elections for each country under alternative electoral systems. A key assumption he makes about voter choice is that voter preferences would have remained the same regardless of the electoral system-although he acknowledges some systems provide powerful incentives for minority party supporters to vote tactically where they believe their first choice party has no feasible chance of winning, particularly in a single member district-i.e. voters will behave differently under different electoral systems. Despite this possible problem, Reynolds argues that in the five countries in his study party identification and voting preferences are very strongly held, which reflect polarized ethnic, linguistic, cultural, ideological and regional lines of cleavage. Thus, the probability is higher that voters in these countries will tend not to vote tactically.
The simulations he conducts provide a persuasive argument in favor of more proportional electoral systems, in that they better promote an "inclusionary" ethos early in the process of democratic transition. For example, he notes that the South African ANC would have been better off with single-member districts, although by supporting proportional representation during the constitutional negotiations this provided enough "voice" to the opposition to keep the varying political interests attached to the incipient political system.
Regarding Ethiopia more specifically, there has been some work on the historical evolution of the political parties in the country and several very good recent studies on the elections and the general process of democratization in Ethiopia. There has also been some work that has recently appeared specifically evaluating the 2005 election and its immediate aftermath. However, as far as the recent work on Ethiopia is concerned, there has been very little systematic work done on the relationship between the electoral systems and the political parties in the country and, to date, no work that has employed the technique of electoral simulations.
What if Ethiopia had employed a different electoral system to govern the 2005 election? Would the adoption of proportional representation (as had been advocated by the Ethiopian opposition) have significantly altered the results? Would the opposition have been better represented (which may have helped avoid the violence and difficulties the country has faced since May 2005)? Before addressing these questions, I first turn to brief discussion of the historical background to the 2005 election.
Background to the 2005 Election
The overthrow of the Communist Derg regime in Ethiopia in 1991 by the Tigrayan Peoples' Liberation Front (TPLF) and its allies marked a new era in Ethiopian politics. Following the collapse of the Derg regime, the victorious TPLF (led by Meles Zenawi) moved quickly to establish its political dominance. The original aim of Tigrayan independence was abandoned by the TPLF when it formed the Ethiopian Peoples' Revolutionary Democratic Front (EPRDF) in order to depose the Derg regime. The EPRDF was an alliance of four other groups, including the Oromo Peoples' Democratic Organization, the Amhara National Democratic Movement, and the Democratic Officers Alliance (in 1993 the South Ethiopian Peoples' Democratic Front was added to replace this group) and the Tigrayan Peoples' Liberation Front. All four regional-ethnic parties were created by the TPLF. In reality, members of parliament from these parties consistently vote with TPLF and have no real independence outside the direction of TPLF.
An important structural change that the EPRDF has instituted under its rule is the reconfiguration of the country into an ethnic federation. Many powers have been 'devolved' to the regions, and the right to use the local language in official dealings has been guaranteed. According to Prime Minister Meles Zenawi, it is only through such constitutional guarantees for ethnic autonomy can the country be held together and future secessions be avoided. Critics, however, point to the fact that despite the federal arrangement, power is concentrated in the hands of the EPRDF (made up of four constituent parties). As a result, similar to the Soviet federal practice of 'democratic centralism,' regional governments are, in practice, implementers of policies adopted by the EPRDF. Others have bemoaned both the financial costs of implementing parallel political and economic institutions across Ethiopia's reorganized ethnic 'states' and are concerned about the threat such ethnic divisions present to Ethiopian unity.
In the 1990s the EPRDF regime emerged as a semi-authoritarian regime and although an opposition was tolerated, it was quite circumscribed. It was within this context that the current Ethiopian opposition parties emerged. Although there are many parties registered at both the national and regional levels, only in 2005 did the opposition coalesce into a viable force that could challenge the EPRDF. In 2003, the United Ethiopian Democratic Forces (or UEDF, sometimes known as Hibrit) was formed in Washington, DC and comprised five Ethiopia-based and nine exile opposition groups. These included widely disparate groups in ideological terms ranging from socialist, to liberal, to secessionist. The principal parties in the UEDF were the Ethiopian Social Democratic Federal Party (ESDFP - formerly the Coalition of Alternative Forces for Peace and Democracy in Ethiopia), the Oromo National Congress (ONC), the United Ethiopia Democratic Party (UEDP), the Southern Ethiopia Peoples' Democratic Coalition (SEPDC), and the All-Amhara People's Organization (AAPO). The UEDF chairman was the political scientist Dr. Merera Gudina, a member of the faculty at Addis Ababa University and chair of the Oromo National Conference. The UEDF vice-chair was Beyene Petros, a leading figure in the Council of Alternative Forces for Peace and Democracy (CAFPDE) and later the United Ethiopian Democratic Party. The UEDF has campaigned to shift greater power to the various ethnic groups and the UEDF insists that the ethnic-confederation model (which other opposition parties, most notably the CUD oppose) should not only be retained, but actually followed more faithfully.
The other major opposition group in current Ethiopian politics is the Coalition for Unity and Democracy (CUD-Kinijit). Kinijit was formed by four new political parties in 2004. It consists of the All Ethiopia Unity Party (AEUP), the Ethiopian Democratic League (EDL), the Ethiopian Democratic Unity Party-Medhin (EDUP-M), and the Rainbow Alliance/Movement for Democracy and Social Justice. The organization contains constituent groups with differing views regarding economic and political management ranging from social democrats to economic liberals. The CUD is highly critical of the EPRDF led government's policy of promoting ethnic federalism. Its leaders challenge the EPRDF's definition of the Ethiopia as an ethnic confederation and contend that it is a recipe for the disintegration of the country. The CUD had strong support in the Amhara region and Addis Ababa. It was led by intellectuals like Hailu Shawil and Berhanu Nega.
These three parties (EPRDF, UEDF, and CUDP) were the main contenders in the May 15 2005 parliamentary elections, the third such elections since the adoption of the EPRDF- sponsored constitution of 1994. The Ethiopian elections were conducted using a SMDP system with 548 districts in the country. The 1995 election had been largely boycotted by the opposition groups, and the EPRDF won an overwhelming number of seats in the parliament. In the 2000 election, the EPRDF again won an overwhelming number of seats (472), with the opposition parties (which competed as individual parties) winning only twelve seats out of the total 547.
Under considerable pressure from the West (particularly after Meles Zenawi's prominent inclusion in British Prime Minister Tony Blair's Commission for Africa), the ruling EPRDF took measures to reform the nomination and election procedure for the 2005 election. Earlier elections in 1995 and 2000 were marked by government harassment of opposition parties and a boycott of the polls by the most influential opposition organizations. In the lead-up to 2005, the EPRDF indicated that it wanted to run an election that was perceived as free and fair by the international community and that included greater participation by opposition parties within Ethiopia. The government agreed in October 2004 to meet some of the demands put forward by leading opposition groups, notably allowing international election observers and ensuring opposition access to state-run media. However, the opposition parties' demand that the electoral system be changed to a proportional representation system was not accepted.
On May 15, 2005, Ethiopia held its third general election for seats in national and regional parliamentary elections. Turnout was around 90 percent. The National Election Board (NEBE) announced that official results would be released on June 8. The initial returns indicated that the ruling EPRDF won over 300 seats, although the opposition parties won all 23 seats in the capital city, Addis Ababa. In addition, the EPRDF could count on their 'affiliated parties' (often referred to in Ethiopia as satellite parties or "Quisling" or even "Condom" parties). Beyond the official EPRDF parties (TPLF, OPDO, ANDM and SEPDO) the EPRDF has its affiliates also in the other states and smaller ethnic groups, such as the Afar National Democratic Party, the Benishangul Gumuz People's Democratic Unity Party, the Gambela People's Democratic Front, the Somali Peoples Democratic Party and others are all members of the EPRDF block and closely controlled by TPLF. Their candidates are selected by the EPRDF's agents, and these parties govern the remaining federal states on behalf of EPRDF. Thus the official results reported underestimate the true dominance of the EPRDF..
Following the election, the CUD and UEDF claimed massive electoral fraud and demanded an investigation of nearly 300 district elections. Anti-government demonstrations erupted in the capital in early June, and were met with violent suppression by security forces, resulting in the death of over 30 student protesters. Both the CUD and UEDF, which had agreed in June to a truce with the EPRDF, continued to allege that massive electoral fraud had stolen the election. The opposition parties had decided to boycott the related August 21 elections in the Somali Region. On September 5, the NEBE released its final results, in which the EPRDF retained its control of the government with 327 seats. Opposition parties won 174 seats (up from 12 in 2000), with the CUD winning 109 total seats, and the UEDF winning 52, and minor parties and independents taking the remainder.
Alternative Electoral Systems
As mentioned above, the intention of this paper is to assess whether the 2005 parliamentary election results would have been different if different electoral systems had been employed. Indeed, if the results had been different, and had the opposition's political strength been better represented in the results, perhaps, using Reynolds' term, an "inclusive ethos" could have been established in Ethiopia, and perhaps compromise would have replaced the violence that followed the election.
Like Reynolds, I assume that voter preferences would have remained the same regardless of the electoral system- in Ethiopia as is the case elsewhere in Africa, party identification and voting preferences reflect polarized ethnic, linguistic, cultural, ideological and regional lines of cleavage. However, even with this assumption, there are certain limitations on what can be tested given the fact that this analysis can only be done post facto - i.e., only on the basis of votes that have already been recorded. Thus, for instance, since the ballot in Ethiopia was not ordinal (meaning no rank ordering was afforded to voters) one cannot assess popular alternative systems such as the Single Transferable Vote, the Approval Vote, or the Alternative Vote, all of which require a rank ordered ballot.
There are at least four alternative electoral systems that can be assessed using the available electoral data. The first is a plurality system called the Block Vote, which uses multi-member districts in which electors have as many votes as there are candidates to be elected. Counting is identical to a First Past the Post/Single Member District Plurality system (as was the system used in Ethiopia) with the candidates with the highest vote totals winning the seats. In this case I use as the natural multimember districts the "zones" which made for districts as large as twenty seats and as small as one seat. The second system is the Mixed Member/Parallel system in which the choices expressed by the voters are used to elect representatives through two different systems - Party-list Proportional Representation and a plurality/majority system. Unlike in some countries (such as Germany) where the PR list compensates for the disproportionality in the results from the plurality/majority system, the Parallel system is a mixed system in which the two components are separated from one another (as is the case in Russia from 1993-97). In addition there are two kinds of Party-List Proportional Representation systems. Generally under this system, each party or grouping presents a list of candidates for a multi-member electoral district. The voters then vote for a party, and parties receive seats in proportion to their share of the vote. However, there is a difference between systems where seats are aggregated nationally (such as in the Netherlands and Israel) or aggregated at the regional level (as in Belgium and Slovakia). Generally these types of systems are accompanied by a minimal threshold in terms of the percentage of the vote in order to qualify for seats (5 percent). There are several ways in which to allocate the seats (and remaining seats from the votes for parties that did not pass the threshold) but the most common is the D'hondt method.
It is important to mention here that one system which is not included in the set of electoral simulations presented in this paper is the widely touted Alternative Vote which comes in both single seat and multi-member district versions. Although not widely used (it has been employed in Australian senate elections, as well as in local elections in Canada and elsewhere) it has been touted as an institutional remedy for politics in ethnically divided societies (as is the case in Ethiopia). The Alternative Vote system allows for voters to express their preferences for candidates in a single member district. If there is not a single candidate who wins a majority (50 percent) of the first preferences votes, then the lowest polling candidate is eliminated and that candidate's second preferences are redistributed to candidates remaining in the race, until a single candidate surmounts the 50 percent threshold.
Scholars have used the Alternative Vote in their simulations, but they were able to do this by inferring second preferences of voters via pre-election voter surveys which identified the party that was the second choice of most voters. Such data, however, do not exist for the 2005 Ethiopian election-hence it would be highly speculative as to the second preferences of voters and thus not appropriate for the simulations in this paper.
Finally, there is another limitation in using existing voting results to estimate alternative possible outcomes from the 2005 election--- there remains considerable doubt as to the veracity of the election results. Indeed at least 139 districts were investigated for election irregularities, and there was some question of the results from the Afar region. Further, the Somali region did not hold its parliamentary elections until August 21, 2005. In order to take the veracity of the election results into account (to some extent) I remove the voting results from 172 electoral districts from consideration that were either protested because of irregularities by the opposition, the governing party, or were from the Somali region. This left the voting tallies from 375 election districts.
In addition, however, there were also questionable results in other parts of the country. For instance, the overwhelming voting returns for the EPRDF in Tigray (in some districts the returns for the EPRDF reached or exceeded 100 percent) can be viewed as somewhat questionable as well. Nonetheless, the results from the 38 districts in Tigray were not officially challenged, so they were not automatically subtracted from the sample of 375 districts. However, I do examine the simulation results with and without the Tigray region seats included.
Table 1 presents voting results by percentage for the EPRDF (and its allies) the CUDP and the UEDF (as well as an aggregated column for "others" generally smaller regional parties not generally aligned with the EPRDF and independents) based upon the vote tallies from the non-challenged results from the 375 electoral districts. The EPRDF dominated in Tigray (93 percent) and did well in Benshangul (65 percent), Gambela (74 percent) and Oromiya (64 percent). The CUDP did well in the cities of Addis Ababa (74 percent), Dire Dawa (42 percent) and the Amhara Region (48 percent). The UEDF had its best showings in Oromiya and Southern Nations, Nationalities, and People's Region (SNNPR). Overall the EPRDF and its allies (based upon the results from the 375 non-challenged districts) won 59.5 percent of the national vote, the CUDP 21.9 percent, UEDF 10.9 percent and others (both independents and smaller parties 7.7 percent).
What if a proportional representation system had been used? Table 2 compares seat allocations based upon the actual results from the 2005 election (from the 375 non challenged electoral districts) with a system where seats are aggregated and allocated by region, with a 5 percent threshold using the D'hondt method. Further, the seat totals are reported for all 375 districts in the sample, as well as the totals if the results from the Tigray region are removed. As indicated, when considering the sample of 375 seats, the EPRDF would have received six fewer seats than it actually did, but the CUDP would have received fifteen fewer seats, and the UEDF eleven fewer seats. The biggest beneficiary from using a PR list system with seat allocations at the regional level would be the small regional parties that were unaffiliated with the EPRDF that would increase their allocated seats from 12 to 44. When subtracting out the seats from the Tigray region, the EPRDF would lose four seats, the CUDP fifteen seats and the UEDF eleven seats. These results do not differ significantly from the seat allocation results from those of a national PR list system (see Table 4 below).
Table 3 reports results from a Block Plurality system redrawing the single member districts into multi-member districts based upon the zones for each region. As indicated in the table, the number of seats for the EPRDF is reduced considerably to only 195 seats (to 158 seats when the Tigray region seats are separated out). The CUDP seat total amounts to 94 (93 when the Tigray region seats are removed) under the Block Plurality method (down from 99 in the actual results). The greatest beneficiary from the Block Plurality system is the UEDF which increases its seat total under this system to 61 seats (from the actual total of 48 seats). Minor regional parties that were not affiliated with the EPRDF would increase their share of seats from 12 to 25 (both with and without the Tigray districts included).
Table 4 below summarizes the seat allocations for each party by electoral system, comparing the SMDP system, the National PR list (with D'hondt with a 5 percent threshold), the Regional PR list system (with D'hondt and a 5 percent threshold), a Mixed Member District-parallel system with the 375 seats and 100 National PR list seats (for a total of 475 seats) and finally the Block Plurality system based on the country's zones. Table 5 summarizes the seat allocations subtracting out the seat totals from the Tigray region.
As indicated in Table 5, the EPRDF does best under a national PR list system (59.5 percent of the seats). This is because the EPRDF received an overwhelming majority in Tigray, and substantial majorities in Benshangul, Gambela, but also Oromiya. Even when separating out the districts from Tigray, the EPRDF performs best under the national PR list system (with 55.5 percent of the seats). The CUDP victories and the UEDF victories were by smaller margins, hence they did better in systems that used plurality competitions (the actual system used, the MMD/parallel system, and especially the Block Plurality system). The UEDF did particularly well under a Block Plurality system, especially since the lost several close single member district competitions (but had relatively high vote totals), and the EPRDF won some seats without a majority of the vote. This meant that in a multi-member plurality competition, the UEDF might have come in fourth or fifth in a five seat district and won seats, although in the actual election they may have lost each of the individual district competitions.
Thus, despite the opposition's call for the adoption of a proportional representation system prior to the 2005 parliamentary elections, the above results suggest that the opposition would have performed better under some other form of district plurality competition (either the SMD plurality system, the MMD-parallel system, or the Block Plurality system). The CUDP performed best under the SMDP system (the one actually used) and the UEDF would have performed best under the Block Plurality system. The opposition overall performed best under the Block Plurality system. Indeed, if the results from the Tigrayan districts were removed from consideration, the EPRDF and its allies would have failed to reach a majority of the seats allocated (winning only 46.9 percent of the contested seats). This may be due, perhaps, to the regional concentration of electoral support for both the major opposition parties, the CUDP in Amhara and Addis Ababa, and the UEDF in Oromiya and SNNPR. The EPRDF and its allies, on the other hand, seemed to perform generally well across all regions, even when excluding the votes from districts where the results were challenged. Thus, if the opposition really wanted to adopt an electoral system that would give them an advantage, the Block Plurality System would appear most attractive in contrast to either a national list or regional list proportional representation system.
These results, however, should also be taken with a large 'grain of salt.' Although this study did not include districts whose results were officially challenged (or from the Somali region), it is likely that the voting results in many of the remaining districts and regions were still highly questionable. Nonetheless, even when taking that into consideration, the opposition would have done quite well in the Block Plurality system, well enough to potentially form an alternative coalition government that excluded the EPRDF, and at the very least, be close enough to potentially lure EPRDF defectors to the side of the opposition.
The above paper sought to reexamine the 2005 parliamentary election results using electoral systems other than the Single Member District Plurality system that was actually employed. Although one must be careful to note that the results did not include all of the districts, given that questionable seat competitions are excluded from the analysis, the results are nonetheless suggestive. As indicated above, and contrary to the opposition's general expectation that they would perform better under the conditions of a proportional representation system, the opposition parties would have performed best in terms of seat allocations under the existing SMD system or a Block Plurality system. This is largely due to the regional concentration of voter support for the CUDP and the UEDF and the more diffuse electoral support for the EPRDF and its allies. Thus, PR would not have been the panacea for the opposition that it was made out to be prior to the 2005 election.
Although I am not in a position to evaluate other highly touted electoral systems for ethnically or culturally divided societies (like Ethiopia) such as the Single Transferable Vote system or the Alternative Vote system (which may produce very different results in the Ethiopian context), of the systems considered in this paper a most promising alternative appears to be the Block Plurality system (at least from the perspective of the opposition). This system, if employed, would simply involve the amalgamation of existing single member district systems into multi member plurality systems, thus making it potentially easier to implement than completely overhauling the existing system via the institution of PR. If the Block Plurality system had been employed this may have produced, at the minimum, a closely divided legislature after the 2005 elections-and perhaps even an opposition-led coalition government. At the very least, it would have created what Andrew Reynolds referred to as an "inclusionary ethos" that may have provided a powerful incentive for the Ethiopian opposition to not take their grievances regarding the election outcome to the streets. Further, such an outcome may have provided an opportunity for both the EPRDF and the opposition to come to an accommodation that would have gone much further in promoting democracy (and pleasing Meles' western backers) than what actually transpired since 2005.
While this paper has examined only one case and one election, it does suggest that the use of such "simulations" allows us to understand the possible consequences of reforming the electoral system in Ethiopia. It is an issue which will undoubtedly gain greater attention as the 2010 legislative election approaches. Perhaps at that point another opportunity will arise in Ethiopia to put the country on the track towards democracy, and at that time the choice of electoral system will be crucial factor in affecting the course of that transformation of the country.
 Taagepera and Shugart 1989; Horowitz 1985; Ishiyama 1997.
 Olsen 1993; Schmitter 1992; Bermeo, 1987.
 See for instance Baker and Scheiner, 2007; Jansen, 2004; Diaz-Caveros, 2005; Danache, Mondak and Cabrera 2005; Bissey, Carini and Ortona, 2004; Dunleavy , Margetts and Weir, 1992.
 An important exception is the work of Andrew Reynolds (1999) who has run such simulations using data from five African countries.
 Levitsky and Way, 2004.
 For further explication of this common concept regarding the emergence of ethnic parties, see Ishiyama and Breuning (1998) and Chandra (2002); for one-party states, see
Young 1966; Collier 1982; Liebenow 1986; Decalo 1990.
 Ottaway 1999; Bratton and van de Walle 1997.
 Mozaffar and Scarritt 2005; Posner and Simon 2002; Reynolds, 1999;1995a; 1995b; Barkan 1995.
 Mozaffar and Scarritt 2005; Bogaards 2004; van de Walle 2003; Mozaffar et al 2003; Kuenzi and Lambright 2001; Lumumba-Kosongo 2001 deal with the development of party systems. Ishiyama and Quinn 2006; Vengroff 1993 focus on individual parties.
 Reynolds.1999; 1995.
 For the evolution of political parties, (see Pauswang, Tronvoll and Aaleniegfried,) 2002; Clapham 1985; Ellingson 1977. Recent studies include Abbink 2006; Harbeson 2005; Harbeson 1998; Henze 1998.
 Ishiyama, 2007; Abbink 2006; Harbeson 2005.
 An exception is a recent piece by John Harbeson who does discuss the evolution of the party organizations in Ethiopia, but not their nomination behavior (Harbeson 2005).
 The constituent parts of the Ethiopia Federation include 9 ethnically-based administrative regions and 2 chartered cities: Addis Ababa, Afar, Amhara, Benishangul/Gumaz. Dire Dawa, Gambela, Harar, Oromiya, Somali, Southern Nations, Nationalities, and Peoples Region and Tigray.
 Other parties that joined to create the UEDF are: Afar Revolutionary Democratic Unity Front (ARDUF), All-Ethiopia Socialist Movement (MEISON), Council of Alternative Forces for Peace and Democracy (CAFPDE), Ethiopian Democratic Union - Tehadiso (EDU Tehadiso), Ethiopian National United Front (ENUF), Ethiopian People Federal Democratic Unity Party (HIBREHIZB), Ethiopian People's Revolutionary Party (EPRP), Gambella People's United Democratic Front, (ONC), Oromo People's Liberation Organization (OPLO - IBSO), (SEPDC), and Tigrayan Alliance for Democracy (TAND).
Beyene Petros and Merera Gudima alternated as chair and vice chair of the UEDF in half year cycles.
 Harbeson 2005.
CUD Electoral manifesto at http://www.kinijit.org/static/KINIJIT-MANIFESTO-English-ver-1.0.pdf accessed March 1, 2006.
 Harbeson 2005: 149.
 Surafel 2001:1.
 Harbeson 2005.
 The D'hondt method is a commonly used "highest averages" method for allocating seats in party-list proportional representation electoral systems. The method is named after 19th century mathematician Victor D'hondt. For a discussion see Taagepera and Shugart, 1989, chapter 3.
 Reynolds 1999.
 One of its strongest proponents of the Alternative Vote system is Donald Horowitz (1991). Using the case of South Africa, Horowitz argued that unlike other electoral systems arrangements that promote post election coalitions (such as proportional representation) the Alternative Vote provided the incentive for compromise, which is the key to accommodation in ethnically divided societies. The Alternative Vote, encourages such preelection compromise because it encourages vote pooling and party appeals across ethnic groups, first, because of the need to obtain a majority of the vote, and also because it is in a party's interest to appeal as a "second choice" to voters, meaning toning down ethnically particular voter appeals.
There have been several critics of the Alternative Vote. Arend Lijphart (1991) and Reynolds (1993) have been particularly critical, arguing that it produces no better results than other electoral systems. Jansen(2004) found that in Canada AV had little impact on proportionality and voter turnout, and did little to facilitate the cooperative behavior among competing parties in provincial elections in Alberta and Manitoba, and on balance, differed little from the single member plurality system.
 See for instance Reynolds, 1999.
Abbink, J. "Discomfiture of Democracy? The 2005 Election Crisis In Ethiopia and Its Aftermath," African Affairs 105, no. 419 (2006):173-199.
Baker, A., and Scheiner, E. "Electoral system effects and ruling party dominance in Japan: A counterfactual simulation based on adaptive parties," Electoral Studies, 26, no. 2 (2007):477-491.
Barkan, J.D. "Elections in Agrarian Societies," Journal of Democracy, 6 , no. 4 (1995): 85-99.
Bermeo, N. "Redemocratization and Transition Elections: A Comparison of Spain and Portugal," Comparative Politics, 19, no. 3 (1987): 213 231.
Bissey, M., Carini, M.and Ortona, G. "ALEX3: a Simulation Program to Compare Electoral Systems," Journal of Artificial Societies and Social Simulation, 7, no.3(2004): http://jasss.soc.surrey.ac.uk/7/3/3.html.
Bogaards, M. "Counting Parties and Identifying Dominant Party Systems in Africa," European Journal of Political Research, 43, no. 3(2004):173-197.
Bratton, M. and N. van de Walle. Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. Cambridge: Cambridge University Press, 1997.
Canache, D., Mondak, J. and Cabrera, E. "Voters and the Personal Vote: A Counterfactual Simulation," Political Research Quarterly, 53, no. 3 (2000): 663-676
Clapham, C.S. "The Workers' Party of Ethiopia", Journal of Communist Studies 1, no.1 (1985): 76-77.
Collier, R.B. Regimes in Tropical Africa: Changing Forms of Supremacy 1945 - 1997. Berkeley and Los Angeles: University of California Press, 1982.
Decalo. S. Coups and Army Rule in Africa. New Haven:Yale University Press, 1990.
Diaz-Caveros, A. "Endogenous Institutional Change in the Mexican Senate," Comparative Political Studies, 38, no. 10 (2005): 1196-1218.
Dunleavy, P., Margetts, H. and Weir, S. "How Britain would have voted under Alternative Electoral Systems in 1992," Parliamentary Affairs 45, no. 3 (1992): 640-655.
Duverger M. "What is the Best Electoral System?" In Choosing an Electoral System: Issues and Alternatives, eds. Arend Lijphart and Bernard Grofman, 30-40. New York: Praeger, 1984.
Ellingson, L. "The Emergence of Political Parties in Eritrea, 1941-1950," Journal of African History, 18, no. 3 (1977): 261-81.
Harbeson, J.W. The Ethiopian Transformation: The Quest for the Post Imperial State. Boulder: Westview Press, 1988.
______. "A Bureaucratic Authoritarian Regime," Journal of Democracy, 9, no.4 (1998): 62-69.
______. "Ethiopia's Extended Transition," Journal of Democracy, 16 no. 3 (2005): 144-158.
Henze, P.B. "A Political Success Story," Journal of Democracy, 9, no. 4 (1998): 40-54.
Horowitz D. A Democratic South Africa? Constitutional Engineering in a Divided Society. Berkeley: University of California Press, 1991.
______. Ethnic Groups in Conflict. Berkeley: University of California Press, 1985.
Ishiyama, J. "Transitional Electoral Systems in Eastern Europe," Political Science Quarterly, 112, no. 3(1997):95-116.
______. "Nominations and Party Development in Ethiopia: The Opposition and the 2005 Parliamentary Election," African and Asian Studies, 6 nos.1& 2(2007): 81-105.
Ishiyama, J. and J.J. Quinn. "African Phoenix? Explaining the Electoral Performance of the Formerly Dominant Parties in Africa," Party Politics, 12, no. 3 (2006): 317-340.
Jansen, H. "The Political Consequences of the Alternative Vote: Lessons from Western Canada," Canadian Journal of Political Science/Revue canadienne de science politique, 37, no 3 (2004): 647-669
Kuenzi, M. and G. Lambright. "Party System Institutionalization in 30 African Countries," Party Politics, 7, no. 4 (2001): 437-468.
Lakeman, E. "The Case for Proportional Representation." In Choosing an Electoral System: Issues and Alternatives, eds. Arend Lijphart and Bernard Grofman, 41-51. New York: Praeger, 1984.
Levitsky, S. and Way, L. "The Rise of Competitive Authoritarianism," Journal of Democracy, 13, no. 1(2002): 51-65.
Liebenow, J. G. African Politics: Crises and Challenges. Bloomington and Indianapolis: Indiana University Press, 1986.
Lijphart, A. "The Political Consequences of Electoral laws, 1945-1985," American Political Science Review, 84, no. 3 (1990):481-496.
______. "The Alternative Vote: A Realistic Alternative for South Africa?" Politikon, 18, no. 1(1991): 91-101
Lumumba-Kosongo T. "Political Parties and Ruling Governments in Sub-Saharan Africa" in R. Dibie (ed.), The Politics and Policies of Sub-Saharan Africa. Lanham, MD: University Press of America, 2001.
Mozaffar, S., J.R. Scarritt, and G. Galaich. "Electoral Institutions, Ethnopolitical Cleavages and Party Systems in Africa's Emerging Democracies", American Political Science Review, 97, no. 3(2003): 379 - 390.
Mozaffar S. and J.R. Scarritt. "The Puzzle of African Party Systems", Party Politics, 11, no. 4 (2005): 399-421.
Olson, D.M. "Compartmentalized Competition: The Managed Transitional Election System in Poland," Journal of Politics 55, no. 3 (1993): 300-415.
Ottaway, M. "Ethnic Politics in Africa: Change and Continuity," in R. Joseph, (ed.) State Conflict, and Democracy in Africa. (Boulder: Lynne Rienner Publishers, 1999).
Pauswang, S., K. Tronvoll and L. Aaleniegfried (eds.): Ethiopia Since the Derg: A Decade of Democratic Pretension and Performance. London: Zed Books, 2002.
Posner, D.J. and. Simon, D.J. "Economic Conditions and Incumbent Support in Africa's New Democracies," Comparative Political Studies, 35,no. 3 (2002):313-336.
Rae, D. The Political Consequences of Electoral Laws. New Haven: Yale University Press, 1967.
Reynolds, A. Voting for a New South Africa. Cape Town: Maskew Miller Longman, 1993.
______. "Testing Voting Systems in South Africa," Voting and Democracy Report. Washington, D.C.: The Center for Voting and Democracy, 1995.
______. "The Case for Proportionality," Journal of Democracy, 6, no.4 (1995a): 117-124.
______. "Constitutional Engineering in Southern Africa", Journal of Democracy 6, no. 2(1995b):86-99.
______. Electoral Systems and Democratization in Southern Africa. New York: Oxford University Press, 1999.
Schmitter, P.C. "The Consolidation of Democracy and Representation of Social Groups," American Behavioral Scientist, 35, no. 4 (1992):422-449.
Surafel G. "Dr. Berhanu Nega: An Evening with the Economist" Addis Tribune. March 23, 2001 p. 1.
Taagepera, R. and Shugart, M. Seats and Votes. Berkeley: University of California Press, 1989.
van de Walle, N. "Presidentialism and Clientelism in Africa's Emerging Party Systems," Journal of Modern African Studies, 41, no. 2(2003): 297-321.
Vengroff, R. "Governance and the transition to democracy: Political parties and the party system in Mali," Journal of Modern African Studies, 31, no. 4 (1993): 541-562.
Young, C. Politics in the Congo. Princeton: Princeton University Press, 1965.
John Ishiyama is a professor of Political Science at the University of North Texas. He is the author or editor of four books and 101 articles (in such journals as the American Political Science Review, Perspectives on Politics, Comparative Political Studies, Comparative Politics, Political Science Quarterly, Political Research Quarterly, Social Science Quarterly, Europe-Asia Studies, Communist and Post Communist Studies, and Party Politics) and book chapters on political parties and ethnic politics in post communist Russian, European, and African politics and political science education. He is a research fellow at the University of Kansas' Center for Russian and East European Studies and is Editor-in-Chief of the Journal of Political Science Education.
Reference Style: The following is the suggested format for referencing this article: John Ishiyama, "Alternative Electoral Systems and the 2005 Ethiopian Parliamentary Election" African Studies Quarterly 10, issue 4: (Spring 2009) [online] URL: http://www.africa.ufl.edu/asq/v10/v10i4a2.htm