Shaheen Mozaffar and James R. Scarritt, "The Puzzle of African Party Systems," Party Politics, 11 (July, 2005), 399-421.

First Paragraph:
Two puzzling features characterize African party systems. One is the relatively low level of fragmentation reflected in low levels of electoral and legislative competitiveness, high vote-seat disproportionality, and the dominance of small numbers of large parties surrounded by large numbers of small parties. The other is the persistence of low party system fragmentation in the face of high electoral and legislative volatility in both majoritarian and proportional electoral systems. Mozaffar et al. (2003) explain these puzzling features of African party systems as the joint effects of electoral institutions and the distinctive morphology of African ethno-political groups that serve as sources of strategic coordination among voters and parties over votes and seats. In this article, we present an alternative but complementary explanation that emphasizes the joint effects of strategic choice and institutional legacies on the formation and stability of political parties.

Figures and Tables:
Table 1. The structure of African party systems
Table 2. The structure and volatility of African party systems
Table 3. Party system structure and volatility classified by electoral rules
Appendix: Countries and elections

Last Paragraph:
Empirically, our analysis implies, counter-intuitively, that the puzzling features of African party systems may be conducive to democratic consolidation. High electoral volatility can be viewed as a system-clearing device that eliminates inefficient parties, leaving a small number of parties to compete for votes and form governments. Because of the political salience of ethnicity as an important source of strategic coordination, and because no African ethno-political group is numerically large enough to form either a political party or a government on its own, multi-ethnic coalitions tend to be the norm in the formation of political parties as well as in the formation of governing coalitions. And to the extent that elections remain the principal legitimate source of forming and changing governments, increasing information about the effects of electoral institutions and the extent of electoral support engenders a learning process for both voters and candidates that is likely to improve the prospect of strategic coordination among competing ethno-political groups over a small number of winning candidates.12 This process, which is already evident in Ghana, Mali, Mozambique and Senegal, among others, will also help lower the current high rate of electoral volatility and the associated cost of forming political coalitions in each election.