Ethnicity, Economic Conditions, and Opposition Support: Evidence from Ethiopia's 2005 Elections

Introduction

What do the 2005 parliamentary elections reveal about the nature of opposition support in Ethiopia? Although the final election results have been disputed by the major parties, these results do suggest that the race between ruling party and opposition party candidates was highly competitive in most electoral districts. However, there is no clear consensus among students of Ethiopian politics to explain the evident variation in opposition support across the country and especially across its largest regions. Some stress the role of ethnicity in determining support for the various opposition coalitions, while others point to such factors as nationalism and neopatrimonialism.

The puzzle of opposition support in Ethiopia can be generalized to the rest of Africa, where ruling parties manage to win reelection by relying on a set of familiar strategies—distributing patronage, exploiting ethnic cleavages, and employing violence (van de Walle 2003; Adejumobi 2000; Diamond and Plattner 1999). While ruling parties' deliberate manipulation of the electoral arena is well established in the Africanist literature (Takougang 2003; Makumbe 2002; Crook 1997), we still lack a clear conception of the factors that enable opposition parties to build popular support in countries where democracy has yet to be consolidated. We have no adequate explanation for why voters in some electoral districts are more willing to take a risk in opting for an opposition party's candidate over the ruling party's, even when it is clear that the government of the day has no intention of leaving office.

To explore this question in the Ethiopian context, I examine the economic conditions that may have influenced opposition support at the level of the electoral district. Two specific aspects of Ethiopia's political development make the study of opposition support particularly interesting. First, the 2005 elections mark the first time that the majority of Ethiopian voters had a real choice between government and opposition parties. Unlike many other African states, which enjoyed a limited period of multiparty politics after independence in the 1960s, Ethiopia did not experience multiparty competition until the 1990s. The notion of legal and democratic opposition was simply absent from previous regimes and the dominant political culture (Gebru Tareke 1991; Clapham 1988; Levine 1974). Nevertheless, while no alternation was achieved through the 2005 parliamentary elections, Ethiopia was transformed from a de facto single-party system to a party system with at least two effective electoral parties.¹

Second, Ethiopia's current institutional organization as an ethnic federal system has primed ethnicity as the principal basis of political mobilization. If we are to see exclusively ethnic voting anywhere, it should be in Ethiopia: the boundaries of the federal states have been drawn to conform to ethno-linguistic settlement patterns in order to produce relatively homogeneous units, and the country's single-member electoral districts (mercha kilil) are based on preexisting, relatively homogeneous administrative districts (woredas). Both electoral laws and party rules virtually guarantee that voters at the district level will choose between candidates from the same ethnic background.² But despite these incentives to coalesce behind a single ethno-regional party, in the 2005 parliamentary elections, voters in the country's largest ethnic regions—Amhara and Oromo—essentially split their votes between the ruling Ethiopian People's Revolutionary Democratic Front (EPRDF) and the two opposition coalitions, the Coalition for Unity and Democracy (CUD) and the United Ethiopian Democratic Forces (UEDF).

I focus in this article on understanding the district-level determinants of opposition support: What kinds of electoral districts were more likely to choose either the CUD or UEDF over the ruling party? Is ethnicity enough to explain the variation we see in these parties' vote shares across districts in the country's largest regions? To answer these questions, I use the 2005 electoral data to estimate a regression model of opposition party vote share at the district level. The results of this analysis confirm the importance of ethno-regional identity in determining patterns of opposition support, but they also show that economic conditions play a role in shaping the opposition parties' vote shares within a district. What is more, I...