Why Are There so Many or so Few Parties?

Abstract

Prior to the 2002 electoral reform, political parties in Russia’s regional legislative elections showed poor performance. Since December 2003, however, all regions have been obliged to elect no less than half of the members of their assemblies by proportional representation. As a result, party competition at the sub-national level became unavoidable. This study tests three kinds of hypotheses dealing with the institutional, sociological, and political factors in the fragmentation of party systems within Russia’s regions. The analysis demonstrates that political factors, especially the activity of the Kremlin and the heads of regional executives, have played the primary role in shaping regional party systems.

Notes

1. This article labels as “regions” the sub-national units comprising the Russian Federation.

2. Only nine regions had previously used a proportional component for legislative elections (Golosov 2003, Table 1).

3. Values were obtained based on the approach for counting the effective number of parties proposed by Golosov (2010).

4. Clark and Golder (2006), however, warn against assuming that Duverger was a pure institutionalist, as many scholars do (Amorim Neto and Cox 1997, 151). They point out that along with electoral laws, Duverger clearly believes that social forces also influence the multiplication of political parties (Clark and Golder 2006, 681).

5. In December 2004 the Kremlin abolished popular gubernatorial elections. When they were restored in May 2012, they were circumscribed by a number of limitations on the participation of challengers.

6. NISP 2016. Nezavisimyi institut sotsial’noi politiki, at http://atlas.socpol.ru/indexes/index_democr.shtml, accessed June 18, 2016.

7. The two-tailed 90 percent confidence intervals around the bold line indicate the conditions under which the degree of a governor’s strength influences the effective number of electoral parties. These intervals have a statistically significant effect until their upper and lower bounds are both above and below the zero line (Brambor, Clark, and Golder 2006, 75–76).

8. The curious case of ethnic heterogeneity that begins under the electoral threshold of 10 percent to facilitate party system fragmentation refers to the Republic of Kalmykia, which has an effective number of ethnic groups equal to 2.47, a legal electoral threshold equaling 10 percent, and, at the same time, a relatively high number of the effective number of electoral parties, namely 3.19.

9. The Mann–Whitney test’s comparison of the values of effective number of electoral parties over the course of regional elections within the second cycle before and after 2011 indicates that these values were lower in the period up until 2011 than afterward (p < 0.01).