Carsten Anckar

Determinants of disproportionality and wasted votes

Abstract The aim of this article is to explain differences in disproportionality. The dependent variable has two components. One is the index of disproportionality created by Michael Gallagher (1991). The other component consists of the amount of wasted votes, i.e. votes that are given to representatives of parties which fail to gain representation in parliament. Three independent variables are used: district magnitude, electoral thresholds, and the combined measurement of these variables, namely the effective threshold. The test reveals that while Gallaghers index of disproportionality is affected by district magnitude and the effective threshold, the amount of wasted votes remains, with the possible exception of a weak link to electoral thresholds, unaffected by changes in the independent variables. Explanations for the findings are introduced and discussed.

Democracies without Parties

The belief that modern democracy is party democracy is widespread. However, the belief may be questioned. A number of small independent island states that subscribe to a high extent to democratic values, standards, and institutions manage without political parties. In all, six such cases exist, namely, Belau (Palau), the Federated States of Micronesia, Kiribati, Marshall Islands, Nauru, and Tuvalu. The analysis of these cases is guided by three general assumptions: (a) the impact of diminutive size on the existence and number of parties, (b) the corresponding impact of geographical noncontiguity, and (c) the impact of culturally defined resistances against party life and party rule. Comparisons with conditions in other small island states suggest that the assumptions are valid given that extreme values are entered into the analysis. Extreme smallness, an extremely archipelagic geography, and an intense cultural resistance all contribute to an absence of political parties in democracies.

Party Systems and Democratization : A Comparative Study of the Third World

This study examines the development of democracy in the Third World in the 1990s. All Third World countries that had at least one competitive election in the 1990s are included. The development of democracy is the dependent variable: did the countries mature into liberal democracy, remain electoral democracies or relapse into authoritarianism? The independent variables measure three dimensions of party systems: fragmentation, one-party dominance, and stability. Presidentialism/parliamentarism and majoritarianism/proportionalism are controlled for as intervening variables. Overall, the results indicate that party system characteristics are of limited explanatory value. However, in certain contexts they seem to play a role. In countries with majoritarian electoral systems, both one-party dominance and a high degree of party system fragmentation are detrimental to the development of democracy. In parliamentary systems, there is a negative association between the development of democracy and the degree of party system fragmentation.

Length of independence and democratic failure

This article takes as a point of departure a well-known but rarely tested assumption in the literature on state-building and democratisation, namely that democratic regimes in newly established states are politically unstable. When states take their first steps as independent entities, the state-building process is often incomplete, the political institutions fragile and democratic routines yet to be established. However, with increasing years of independence, these democratic shortcomings are expected to be remedied. This makes it reasonable to assume that the likelihood of democratic failure decreases with increasing years as an independent state. Based on an extensive empirical data set, the conclusion was reached that there is indeed a negative relationship between the length of independence and democratic failure and that this association is insensitive to the period in time when the countries received their independence. Furthermore, the results suggest that the length of democratic rule has a positive impact on democratic stability. However, this pattern is detectable only in states created after 1946.

Are powerful presidents always popularly elected? Implications for separating semi-presidentialism from parliamentarism

ABSTRACT The traditional conception of semi-presidentialism stipulates that powers are shared by a popularly elected president and a prime minister, who is responsible to parliament. In recent years, an increasingly popular strategy has become to disregard the power dimension and define semi-presidentialism only with regard to whether the president is popularly elected or not. Based on a principal-agent framework the present study sets out to test the relationship between the mode of election and the powers of the president in democratic republics where the government is dependent on the legislature for survival. Findings indicate that although powerful presidents most often are popularly elected there are also instances where a non-popularly elected president shares executive powers with the prime minister. This makes it difficult to define semi-presidentialism only with regard to how the president comes to power.

Nordic Contributions to Comparative Research

Comparative politics has traditionally not been one of the dominant sub-fields of political science in the Nordic countries. Although early pioneering efforts were made by Herbert Tingsten and later by Stein Rokkan, this field of research led an obscure life in these countries until the late 1980s/early 1990s. This article gives a short overview of the development and present state of comparative research in the Nordic countries. It also gives a brief account of the comparative research profiles at the political science departments in Nordic universities. In order to give an empirical assessment of the status of comparative politics among Nordic political scientists, the article follows up an earlier study on publication patterns in five Nordic political science journals. In addition, a bibliographical analysis of Nordic contributions in international journals with a focus on comparative politics is undertaken. The results show, among other things, that there has been a marked increase in the share and number of comparative articles published by Danish, Swedish and Norwegian scholars. In Finland (and Iceland), however, no similar trend can be discerned. The conclusion is that the internationalization of political science is likely to stimulate cross-country research and accordingly comparative politics is likely to consolidate its position as a major political science discipline in Scandinavia.