EUROCORES_HumVIB: REPCONG_Policy Congruence between Citizens and Elites

Representative democracy is premised on the notion that there should be a reasonable degree of congruence between the priorities of citizens and those elected to represent them. Representatives are mandated to represent electors` interest, values and opinions. This concept of representation is based on the idea of the responsible party (Ranney 1971); parties compete for votes by offering diverse policy platforms, and voters choose the party whose promises come closest to their preferences. Parties in government will then "implement" what they offered and hence, meet or at least come close to the preferences of a majority of their voters. In an ideal world, representatives fully reflect their voters` views, and hence, are considered responsive leading to well represented citizens (Powell 2004). But in the real world, policy priorities and positions of representatives do not always mirror citizens` views leading to a "representation gap" (Whitefield 2006). Surprisingly, we still know very little about the actual extent of policy congruence and the nature of representation in European democracies and its impact on citizens` attitudes towards it. Based on this gap in the literature our proposal focuses both on two sets of questions: 1. What determines the level of policy congruence between citizens and elites (policy congruence as the dependent variable)? In particular, what impact do political institutions and direct democracy have on policy congruence? 2. What impact has level of policy congruence on the perception of and satisfaction with democracy (policy congruence as an independent variable)? Pursuing our first research question, we examine which institutional and political factors have an impact on the degree of policy congruence and in turn on democratic representation. Despite the centrality of representation to democracy our understanding of these effects (institutional, political and contextual) are often country-specific and highly idiosyncratic as cross-country research is plagued by the problem of incomparable data (see, however, Powell 2000). Now that data which allows us to compare countries on an equal footing has become available - both European survey data (individuals/citizens) and European comparative party policy data (parties/elites) - questions like ours can be pursued. All these factors are examined at the national and supra-national level. We anticipate different degrees of policy congruence for each level and we extend our analysis to determine the causes of this divergence. Our second research question, tests the impact of differences in policy congruence on satisfaction with democracy as well as the perception of specific representative institutions, such as national parliaments/governments and European institutions such as the European Parliament. We stipulate that policy congruence between citizens and elites is shaped by institutional, political and contextual factors such as party size or type of government (e.g. Powell 2000). Yet, these factors are located are different levels: party size is at the level of individual parties; it is an attribute that is specific to each party at a given point in time. We take the hierarchical nature of these relationships into account by employing the technique of multi-level analysis and, where necessary, developing new techniques that are appropriate to analyse the complex data structure we investigate (Goldstein 1995; Snijders/Bosker 1999; Steenbergen/Jones 2002). Multi-level models (MLM) are designed to analyse these types of (non-hierarchical) clustered data. In this study, policy congruence is clustered on the basis of policy dimensions, countries and time; citizen perceptions and attitudes are clustered on the basis of parties, countries and time. To sum up, our project constitutes a first attempt at explaining variation in levels of policy congruence and its impact on representation and democracy across European countries, parties, policies and between levels of governance over time by using novel methodological approaches. References: Goldstein, Harvey (1995) Multi-level statistical models, London: Oxford University Press. Powell, G. Bingham (2000) Elections as Instruments of Democracy, New Haven and London: Yale University Press. Powell, G. Bingham (2004) "The chain of responsiveness", Journal of Democracy 15/4, 91-105. Ranney, Austin (1971) Parties in state politics, in: H. Jacob and K. Vines (eds.) Politics in the American States: A Comparative Analysis, Boston: Little, Brown, and Co. Snijders, Tom A. and Roel J. Bosker (1999) Multilevel analysis: an introduction to basic and advanced multilevel modelling, London: Sage. Steenbergen, Marco R. and Bradford S. Jones (2002) Modeling Multilevel Data Structures, American Journal of Political Science 46, 218-237. Whitefield, Stephen (2006) "Mind the Representation Gap: Explaining Differences in Public Views of Representation in Postcommunist Democracies", Comparative Political Studies 39/6, 733-758.

In real-world representative democracy, policy congruence between citizens and those elected to represent them is partial, for various reasons: representatives may deviate from what they originally promised when electoral sanction is unlikely, parties have informational advantages, or the multi-dimensional nature of party competition in mass elections distorts the congruence between citizens and party preferences. Yet, when starting the project we knew little about the actual extent, the determinants and consequences of policy congruence between citizens and elites in Europe, at the national or the European Union levels. We contributed to scientific progress by testing theories of representation and empirically assessing policy congruence between representatives and represented in Europe and the EU: we scrutinized democratic representation in the EU by analysing congruence in a setting of multi-level governance and by specifically examining formerly neglected EU issues. More specifically, drawing on normative models of representation, we established varied levels of policy congruence in different dimensions of political conflict (left-right, European unification, immigration), as well as differences in congruence degrees achieved at distinct levels of governance (regional, national, EU) through several empirical analyses of European countries and the EU as a whole. We did so by utilizing existing data resources (e.g. European Social Survey, European Election Study, Comparative Electoral System Survey), as well as methods of multi-level analysis. We also expanded previous research on representation and policy congruence by looking inter alia at the multiple links between inequality and representation, gender and representation, political trust and party-voter congruence, extreme right and left party-voter congruence across Europe, and voters perceptual agreement regarding party positions.The integration of the project in a collaborative research project had two main positive effects:firstly, the collaboration provided us with a platform for cooperation, allowed sharing expertise and databases, and facilitated the organization of dissemination activities and the co-authorship of articles. Secondly, it helped us to enrich and extend our methodological expertise via the ESF-funded Workshop on Multilevel Modelling and the participation of junior researchers in Summer Schools of Methods and Data Analysis. More specifically, it increased the visibility of our research and enabled us to exchange ideas and expertise as well as collaborate with researchers working in the same area of research.