Return to: Search Page or to: Table of Contents Vol. 20, Issue 5

Thomas M Meyer and Wolfgang C Müller, "Testing theories of party competition: The Austrian case," Party Politics, 20 (September, 2014), 802-813. [Available at ]

First paragraph:
Ever since Anthony Downs' (1957) made his powerful assertion that parties in two-party systems converge to the median voter, political scientists have been concerned with spatial analyses of party competition. In the course of that journey, they have developed various models that explain the positioning of parties in the policy space.

All of these models ground in the assumption that voters base their choices on spatial thinking about the parties' proposed policies. According to Downs' proximity model, voters value parties close to their personal policy preferences. Under a normal distribution of preferences, competitive parties then take policy positions at the centre of the policy space (Downs, 1957). In contrast, the directional model of voting (Rabinowitz and Macdonald, 1989) states that voters prefer parties with intense policy positions. Assuming voters follow this theory of voting, parties take policy positions further away from the centre of the policy space (see, e.g., Adams and Merrill, 1999a). Other spatial models assume that voters discount the parties' proposed policy positions (Grofman, 1985) or consider expected policy outcomes rather than individual parties' policy positions (Kedar, 2005, 2009). Moreover, spatial elements of voting have been combined with non-policy-related factors likely to affect vote choices (Merrill and Grofman, 1999).

Table 1. Results of conditional logit models predicting vote choice.
Figure 1. Comparing model predictions to actual (perceived) party policy positions
Table 2. Comparison of predicted vote shares.

Last Paragraph:
(First paragraph of Conclusion This article has tested rival theories of party competition. Our analysis is based on four spatial models of voting. Our main goal is to study the observable implications of these models for party policy positions. Here, we bring in the Austrian case. It adds to the empirical body of literature predicting party position-taking that is still restricted to a rather narrow set of countries (Adams et al., 2005; Calvo and Hellwig, 2011) and has not kept pace with the extensive theoretical debates on spatial models of voting.

Last updated November 2014