The Legislative Logic of Electoral Reform in Ukraine

Abstract

Since the end of Communist rule, Ukraine has undertaken three major electoral reforms, moving from a single-member district majoritarian system, to a mixed-member system, to a closed-list proportional representation system, and back to the mixed-member system. Some argue that political parties are primarily motivated by the desire to maximise seats or improve their ability to impact on policy. I argue that existing theories of electoral reform often assume that parties are unitary actors during electoral reform. My analysis of electoral reform in Ukraine clearly demonstrates significant intra-party dissonance on electoral system preferences. This result questions the usefulness of the party unity assumption.

Notes

¹ Benoit (2004) presents a more explicit, formal model of seat maximisation.

² There are a handful of studies that do discuss issues of party discipline and cohesion. Renwick et al. (2009) argue that the move to a bonus adjusted proportional representation system in Italy in 2005 was instituted, in part, by the desire of party leaders to control candidate selection. More centralised control would translate into more disciplined parties and electoral coalitions. In the eyes of party leaders, more disciplined parties are more effective ones. Reilly (2007) contends that the adoption of mixed-member majoritarian systems in several Asian states was based on a desire to create more cohesive, stable, and nationalised political parties. Thus, there is some, albeit limited evidence that party discipline and cohesion have mattered in some electoral reform debates.

³ In mixed-member majoritarian systems, voters cast one ballot in a nominal tier and another in a list tier; parties obtain seats independently from each tier (Shugart & Wattenberg 2001). These systems differ from mixed-member proportional systems where the tiers are linked in terms of party seat share.

⁴ Groups of constituents, workplace collectives, and political party conferences all had the right to nominate candidates. The rules for political party nominations were the most onerous (Bojcun 1995). Parties had to convene regional conferences with at least two thirds of its members or 50 delegates and provide the electoral commission with detailed membership lists. For voter groups, they simply needed to provide information on the minimum of ten voters necessary to form the group. Worker collectives only needed to provide a declaration of support for a particular candidate. Once nominated, all candidates had to gather the signatures of 300 constituents in support of their candidacies.

⁵ The Rada uses an absolute majority system, meaning that for a vote to pass it needs to obtain at least 226 ‘for’ votes (more if the issue requires a super majority). Thus, not voting for whatever reason has the same effects as a vote against.

⁶ Information on party membership was taken from the Rada website.

⁷ The probabilities were calculated using Clarify 2.0 (Tomz et al. 2003).

⁸ PZU was taken over by a group of business interests who sought to capitalise on the party’s name recognition, not to push for eco-friendly government policies (Kubicek 2001).

⁹ This bloc included the party that represented both the powerful Donetsk (Party of Regions) and Dnipropetrovsk (TU) clans.

¹⁰ See the Appendix for a list of the votes and bills.

¹¹ Tests did not indicate the presence of overdispersion; therefore, I chose a Poisson regression model.

¹² Given the significant turnover in parliamentary parties during this period, aggregating the party affiliations into these categories avoids having to employ 15 different party variables that do not cover all periods.

¹³ I used Clarify 2.0 to simulate the vote counts (Tomz et al. 2003).