From elections to outputs: linking party system change and distributive policy change

(to download the first research paper from this project, go to --> publications --> conference papers)
 
The Swiss National Science Foundation has granted us funding for this project (with Prof. Hanspeter Kriesi) on the impact of party system change and party competition on distributive policy change (project duration April 2010 to March 2013). We have Dominik Geering as a doctoral student on board for the entire time and Simone Wasmann as a research assistant until January 2012.

Here is a summary of the project proposal:

People vote for political parties, because they hope for particular policy outputs, which correspond to their preferences and needs. Responsive parties deliver such policies that cater to their constituencies. In the field of social and economic policies (i.e. distributive policies), the political science literature observed a very straightforward relationship between electorates and parties in the post-war period: The working class was the core constituency of the left and the more privileged strata tended to vote for the right. Accordingly, left-wing governments delivered more generous economic and social policies and right-wing governments stood for less generous and more selective policies (e.g. Hibbs 1977, Korpi 1983, Garrett 1998). Individuals knew what to expect when casting their vote for one side or the other.
Since the 1980s, this pattern has increasingly vanished: in many countries, left-wing parties have engaged in retrenchment and liberalization, adopting a more market-liberal agenda. At the same time, many right-wing parties have defended social rights and benefits, rather than enacting the radical cutbacks one might have expected (e.g. Kitschelt 2001, Pierson 2001, Huber and Stephens 2001, Ross 2000, Beramendi and Rueda 2007, Häusermann forthcoming). The traditional left-right distinction seems to dissolve and it has become unclear to both voters and political scientists what parties stand for. Are parties still responsive, and if yes, to whom? Two answers have been developed in the literature, and through this project, we want to propose and explore a third one. The first answer is that parties have indeed become unresponsive, i.e. detached from their socio-structural constituencies (e.g. Mair 2004). They govern in an elitist way, responding to structural and strategic pressure, rather than to voters’ preferences. The second answer is that parties are still responsive, but to the cultural, rather than the distributive concerns of their electorates (e.g. Kitschelt 1995, Kriesi et al. 2008). According to this view, distributive policies have lost their key role in linking parties and voters.
We propose an alternative explanation. We argue that economic and social policies still matter: political parties may still be responsive to the distributive concerns of their constituencies. However, these constituencies have changed profoundly. The left is not simply the party of the working class anymore and the right has lost its monopoly on the highly skilled and high-income voters: party constituencies have become more heterogeneous. Moreover, different constituencies have specific needs, which cannot simply be understood in terms of “more vs. less” welfare: parts of the working class still prioritize income protection, while others are much more in need of jobs or public services such as active labor market policies or child care infrastructure. Consequently, the transformation of party positions with regard to distributive policies may not be the result of unresponsiveness or of a shift to cultural politics, but a consequence of electoral change. Parties do respond, but they respond to different preferences than in the post-war period, and they need to reposition themselves strategically in a deeply transformed party system. So far, research on electoral change has developed separately from research on the transformation of social needs and risks. Hence, we lack an understanding of the links between voters, party system change and policy-making. In this project, we want to investigate these links.
 
Our project will be one of the first to marry research on voter preferences, party system transformation and distributive policy analysis. We include seven countries (Austria, Denmark, the Netherlands, the UK, Italy, Germany and Switzerland) and three policy fields (family, tax and labor market policy) to investigate the effect of electoral change on policy reforms since the 1990s. The project has three stages: in the first stage, we will analyze electoral change, voter preferences and party positions in the electoral arena. We will identify the constituencies of parties, their preferences and the congruence between voter and party positions. In a second stage, we establish the empirical link between the electoral arena and the parliamentary arena, by comparing the party-positions on selected policy reforms with the policy positions these same parties advocate in the electoral arena. Finally, we will analyze policy-making processes in the parliamentary arena and the distributive outputs of the reforms. We will test to what extent the distributive consequences of reforms can be traced back to voter preferences, party preferences and coalitional dynamics within governments. Across the three stages, the project will use a variety of data sources (survey data, coded data on political parties, primary sources from policy-processes) and methods (descriptive statistics and regression techniques, uni- and multi-dimensional scaling, process-tracing).
 
With this project, we want to make contributions of both scientific and practical significance. We want to provide party system research with a more adequate understanding of the issues at stake in today’s distributional policy agenda, and we provide policy researchers with a more adequate understanding of what center left and center right parties want and do. Furthermore, we will investigate the consequences of electoral change by identifying the winners and losers of distributive policy-making in the new party political space. The project also aims at making a strong empirical and methodological contribution: we rely on a newly developed technique of computer-assisted semi-automatic coding (COSA “core sentence annotation”) to generate a database on the preferences of political parties in both electoral campaigns and actual policy-making processes. This new database will link what parties “say” (in elections) to what they “do” (in parliament), and it will be available for further research. Last but not least, the research we plan will also be of very practical interests. We will be among the first projects to actually test whether parties (still) do what their voters want. This will be of high interest not only to political parties themselves, but also to the media and to voters, who must hold parties accountable in democratic politics.