Election Trends 2000-2014: A Numerical Analysis of Participation and Representation

30.07.2015 Leon Malazogu

With 10 sets of elections in a period of 15 years, there is now sufficient information for evaluating Kosovo’s understanding and practice of democracy. Given the absence of upcoming elections, the next two years present a great opportunity to reflect upon and improve the electoral experience. This paper looks at a wide range of trends, draws inferences, and presents analysis in a neutral fashion to inform policy-makers who may be deliberating about electoral design in the near future.

A longer perspective is necessary to make sense of evolutionary shifts and to evaluate the slow rate of democratization. For any reforms to be successful, there should be a baseline upon which past electoral history can be judged. This study aims to evaluate that history. Although still in its infancy, Kosovo’s democracy has already begun to show some signs of consolidation. This is indicated by the reduction in the number of parties contesting elections, by recent mandates starting to last longer, and by the completion of three peaceful half-transitions.

In the first half of the paper, we compared official figures released by the CEC (pertaining to the voters list, turnout, party performance, and individual votes) to determine long-term trends and what can be inferred as a result. The paper begins with an analysis of the reasons behind a surge of invalid and spoilt ballots, which would not have happenned if Kosovo had electoral districts. We cautioned against organizing several sets of elections on the same day, especially if the ballot is not simplified. It is also baffling to note that although all voters were given three ballots, around a sixth of the votes were not cast for municipal councils, which is a strong indicator of potential fraud.

The voters list has improved recently, which is an encouragement that more can be done. Making the voters list more accurate could be in the interest of popular politicians who would be elected with a higher proportion turnout. Of every three registered voters, two vote, hardly the crisis that official figures yield. The lack of figures and the inconsistent manner in which they are published, makes analyses with official figures difficult and costly.

The second half of the paper looks at the results of three public opinion polls. Trends (such as how Kosovars perceive their leadership, institutions in charge of elections, and party membership) are compared over time and then cross-tabulated by a range of demographic factors. Voters are increasingly changing their favorite party. From 2007 to 2010, about half of respondents changed their party choice. Middle-aged respondents, those with higher education, and women form the greatest share of swing voters and should be targeted by political campaigns for this reason.

One of the growing trends is that the higher a person’s income is, the more likely they are to have met a public official. Employed respondents, those who are educated, and those between the ages of 35 and 44 are more likely to be party members. The percentage of those who feel very close to their party has dropped from 25% to 15%, while the rate of Kosovars who do not feel close to politics has increased.

The paper was prepared with contributions and research assistance by: Ngadhnjim Halilaj and Rezarta Delibashzade.