Electoral Systems, Party Systems, and Ideological Representation: An Analysis of Distortion in Western Democracies

The effects of party system features and election rules on ideological representation can be seen in parliamentary elections in Western democracies over a fiftyyear period. “Distortion” is short-term representation failure—the distance between the median voter and the legislature or government immediately after the election. Electoral choice and left-right positions of parties (from the manifesto data) can be used to estimate median voter positions. The number of parties, party polarization, and the election rules all independently affect ideological distances. But party system polarization seems to be the predominate factor shaping distortion of governments’ relationship with the median voter. Examining the effects of party systems under different election rules helps clarify the causal connections between legislative and government levels. Democracy means government by the people. In modern nation-states policies are predominately made by elected representatives of the people, rather than by the people themselves. Political scientists have invested a great deal of effort in exploring the connections between the people and these representatives. One element that looms large in both theoretical and empirical studies is systematically generated correspondence between the preferences of the people and the commitments of their chosen representatives. Such correspondence is sometimes identified as the definition of democracy, as its best justification, or as essential to its overarching goal. We need not debate alternative definitions or explore the complexities of interests and preferences to accept that agreement between citizens and representatives is “one of the most important notions within the broad family of theories of representative democracy.” If elections systematically fail to generate close correspondence between the preferences of citizens and the stated positions of those they elect, what is here called “distortion,” the quality of the democratic process may be diminished. In comparative studies of representation, two general empirical approaches can be discerned. One of these assumes that what voters want is best summarized by their partisan vote; representation is measured by comparing the distribution of party votes with