Kato, Koichi
(2008)
Evaluating compulsory voting: Australia in comparative perspective.
PhD thesis, London School of Economics and Political Science.
Abstract
Low turnout is a growing concern among the industrial democracies. Compulsory voting has achieved very high turnouts in several countries, but it has been mostly neglected as a solution to the problem of low turnout elsewhere. This thesis considers the usefulness of compulsory voting for industrial democracies. I argue that, for it to be useful, compulsory voting must be effective on two levels. First, compulsory voting must be effective in increasing turnout. Second, the high turnout resulting from compulsory voting must improve the total utility of the people-defined here as the well-being of the people-otherwise compulsory voting will not ultimately be useful. Rational choice models are constructed and operationalised in order to describe, explain and evaluate compulsory voting. Although data analysis is undertaken for a range of industrial democracies in order to test these rational choice hypotheses, the major focus of this research is on Australia, which has achieved very high turnout levels (around 95% of the registered voters) since the introduction of compulsory voting for federal elections in 1924. Furthermore, by examining the case of Australia, this thesis determines the conditions and necessary adjustments for compulsory voting to work effectively in practice. Finally, compulsory voting is tested with rational choice theory and data analysis on the actual industrial democracies in order to see whether this system is applicable under globally varying conditions. The conclusion of the analysis is that compulsory voting seems to be useful for several industrial democracies in theory and also seems to be workable in practice. However, some subjective judgement needs to be introduced for a full cost-benefit analysis to be made about compulsory voting.
Actions (login required)
 |
Record administration - authorised staff only |
Download statistics
Download statistics