The Influence of Programmatic Homogeneity on the Formation of New Political Parties

This paper argues that the gap between the programmatic supply of existing political parties and concerns of voters is a crucial factor in explaining the formation of new political parties. Whereas studies on single parties and party families have addressed this point quantitative studies on new political parties miss an adequate distinction between the programmatic supply side of existing parties and the demand side of voters in theoretical and empirical terms – for example when considering the importance of new issues. The main hypotheses of this paper are that the probability of new party formation is an increasing function of the programmatic homogeneity of existing political parties and that this effect gets stronger with the heterogeneity of society. In other words the incentive for party formation goes up when existing parties do not sufficiently fulfill their representational task. The hypotheses are tested by a negative binomial regression using pooled time-series cross-sectional data from 22 OECD countries. A new measure of programmatic homogeneity is used to operationalize the programmatic supply side of political parties. In contrast to measures of polarization it enables a clear-cut distinction between the supply and demand side of the political market. The gap is then operationalized by an interaction effect between this new measure and several indicators of societal heterogeneity. The empirical results support the hypothesized relationships. Moreover this quantitative approach enables a comparison of the explaining power of different factors. It can be shown that the explanatory contribution of the programmatic factors is equally high or even higher than that of factors usually mentioned in the literature on new parties – e.g. electoral institutions or age of democracy. The omitted distinction between supply and demand side and the omission of programmatic factors might explain why previous quantitative studies find insignificant or contradictory results in regard to other relevant variables.