And They Still Matter...: Parties' Preferences on Markets and the Liberalization of Western Societies

In the fields of welfare, tax, and liberalization policies, many scholars have argued that partisan differences have disappeared since the 1980s. Recent evidence points to conflicting results. I conduct a meta-analysis of studies analyzing partisan differences in these fields from 1980-2013. I found that 90 percent of the stated party indifferences are due to improvable measurements of partisan preferences and the negligence of sufficient executive majorities to implement envisaged reforms. But even the use of more sophisticated left-right measures entails a major drawback: the variant meaning of the left-right construct leads to systematic underestimation of partisan impacts. I suggest a new measurement procedure for party preferences applying confirmatory factor analysis. This enables the analysis of the temporal and spatial stability of latent constructs.

Recent discussions about differential item functioning in major political conflict dimensions highlight the necessity to assess the stability of underlying factors. This problem also holds true for expert surveys and limit their genuine function as a validity instrument. My assessment indicates that only the economic dimension has a solid stability over time and space, when conceptualized as a conflict over the application of market mechanisms – the concept of market liberalism. Programmatic proposals (measured in terms of issue salience or word score) can be seen as reflecting the underlying construct of market liberalism. The reflective measurement of partisan preferences is retested within the models of the meta-analysis. The results show robust effects of partisan market preferences over the whole range of welfare, tax, and liberalization policies.