The factors of geographical representation in a closed list PR system: The case of the Republic of Serbia

Abstract

The author maps the patterns of geographical representation in the National Assembly of the Republic of Serbia over the last nine electoral cycles, since the year 2000. During that period, a closed-list PR system is implemented in Serbian elections, within a single nationwide constituency. According to the theoretical expectations, such a system is unsuitable for equal geographical representation, because voters choose national lists as a whole, without affecting the order of the candidates, so the system does not provide incentives for parties to offer regionally balanced lists. However, even within that framework, a significant number of elected representatives come from different local subdivisions. Within districts as this paper's level of analysis, the author is determining underrepresentation and overrepresentation patterns, as well as potential demographic, economic, or geographical factors affecting the representation ratio, i.e. the share of elected representatives in relation to the share of the district's population.