The Evolution of Party Systems: A Global Perspective

Abstract

The evolution of party systems: a global perspective ECPR Research Sessions 2016, Nijmegen proposal Contact – Holger Döring, University of Bremen, Research Session group category A – “very early stage” _ECPR standing groups endorsements_ Political Parties – email by Kurt Richard Luther (Keele University) on 21. January 2016 Elites and Political Leadership – email by Patrick Dumont (University of Luxembourg) on 22. January 2016 # Summary We aim at providing a systematic global assessment of the evolution of party systems by studying about 70 party systems that have experienced longer periods of democracy since 1900. Our theoretical basis is anchored in Kitschelt’s theory of party linkage strategies (Kitschelt 2000, 2014). We aim to add a temporal assessment about the evolution of party systems by comparing the durability of parties, their profiles as well as electoral and leadership dynamics. The major long-term goal is to publish *two edited volumes* about the evolution of party systems in the world since 1900, one descriptive account and an analytical assessment in a second volume. As a *pilot study*, we have prepared respective sources for three countries in three regions of the world (Europe: Germany, Sweden, Poland; Latin America: Brazil, Chile, Ecuador; Asia: Japan, Indonesia, Philippines). The workshop allows us to discuss the major issues in mapping party system evolution in a global perspective, to develop a draft for the codebook and to prepare an agenda for funding applications. The group combines experts on party systems in particular regions of the world and experts in the measurement of party profiles. # Workshop _Preparation_ -- To prepare the research session meeting, all project participants will assess the questions from the Kitschelt (2014) expert survey and evaluate what information can be coded for all political parties since 1900 in two countries. We have collected the respective information (election data, pdf-copies of country chapters in political handbooks) for nine countries in three regions of the world. _Structure_ -- The workshop combines presentations on subjects based on the expertise of the project participants (literature review party systems in particular regions, party expert surveys, online data infrastructures) and discussions about the coding of party profiles (linkage strategies, organizational base, leadership information). _Format_ Day 1a – presentation of participants profiles (regional and/or measurement expertise) Day 1b – discuss linkage literature and party system dynamics in all regions of the world Day 1c – measuring party positions – what have we learned Day 2a – core concepts of mapping party profiles – evaluation feasibility of coding on the basis of compiled sources Day 2b – online conference with Simon Bornschier (Zürich), Allen Hicken (Michigan), Kevin Deegan-Krause (Wayne State), Tom Mustillo (Purdue) Day 2c – compile questions for draft of codebook Day 3a – planning of funding lines and book structure Day 3a – coordinating next meeting and team member responsibilities _Deliverables_ -- An *agenda* for the project, *funding plan* and an initial draft of the *codebook* forms the output of the research sessions. ## Profile participants The workshop does mainly combine younger scholars that have consolidated their academic career after receiving their PhD. The participants are recognized experts on party competition in a particular region of the world and/or experts on measuring party profiles and data infrastructures. Holger Döring, University of Bremen, _Expertise_ – Europe, measurement, data infrastructures (parties, elections, cabinets) _Related work_ – ParlGov database (www.parlgov.org), Party Facts data infrastructure (www.partyfacts.org) Fernando Casal Bértoa, University of Nottingham, _Expertise_ – Europe (esp. Central-, Eastern- and Southern Europe), party systems _Related work_ – Party Systems and Governments Observatory (www.whogoverns.eu), co-editor of the Party Politics and Party Systems book series at Routledge. Elin Bjarnegård, Uppsala University, _Expertise_ – Asia, party politics, gender and politics _Related work_ – Country expertise about Thailand, Myanmar and Cambodia. Lars G Svåsand, University of Bergen, _Expertise_ – Africa, Europe, party politics _Related work_ – Extensive country expertise about democratization in Malawi. Nina Wiesehomeier, Swansea University, _Expertise_ – Latin America, party politics, measurement _Related work_ – Latin America expert survey (ninaw.webfactional.com/ppla), Team Populisms member (populism.byu.edu) Sebastian Fust, University of Zürich, _Expertise_ – Latin America, party politics, measurement _Related work and comment_ – Latin America fieldwork in 2014. PhD student of Simon Bornschier, a project partner (see below), who cannot attend due to scheduling conflicts. ## Project partners We have already cooperated with other colleagues from Europe and the US on particular aspects of the project. Four have been involved in the pilot study and have committed to contribute to the project and a research session meeting. They would join the meeting in a *video conference on the afternoon of 30 June* on Skype or Google Hangouts. Simon Bornschier — University of Zürich — Europe, Latin America Kevin Deegan-Krause — Wayne State University — Europe Allen Hicken — University of Michigan — Asia Tom Mustillo — Purdue University — Latin America # ECPR research network and contribution The idea and identification of the need for our project is based on several ECPR activities. Project members have met at ECPR conferences and the joint sessions and discussed the feasibility of combining our research expertise on particular party systems of the world and the data we have collected. Holger Döring and Kevin Deegan-Krause have harmonized party information from a set of core datasets about political parties that have been maintained by ECPR members (EJPR-PDY, Manifesto project, ParlGov, CHES, CLEA etc.) in the Party Facts project . The workshop participants are active members of ECPR standing groups (Political Parties, Latin American Politics, Elites and Political Leadership). Fernando Casal Bértoa is the co-director of the ECPR Summer School on Political Parties and Democracy in 2016 at the University of Nottingham. Future ECPR events are the focal point for upcoming group meetings. # Project description ## Studies of party systems are focusing on particular regions only Political science work on political parties and the evolution of party systems is still mainly confined to the study of particular regions of the world with little exchange across the different communities. We have an extensive and established work on Western *Europe* (Bartolini and Mair 1990; Bolleyer 2013) and about democratic consolidation of party systems in Central-/Eastern Europe (Kitschelt et al. 1999; Tavits 2013) increasingly within a comparative framework across these two regions (Rohrschneider and Whitefield 2012). There is also a large body of work on party systems in *Latin-America* often on the basis of similar theoretical approaches and indicators (Kitschelt et al. 2010; Mainwaring and Scully 1995). This literature has added detailed accounts of clientelism as an alternative linkage mechanism (Stokes et al. 2013). Scholars working on *Asia* have looked at the nexus of stable party systems and electoral authoritarian regimes (Hicken and Kuhonta 2014). Finally, work on *Africa* has focused on the impact of elections on democratization (Lindberg 2006) and the authoritarian origins of party systems (Riedl 2014). Recently, there have been empirical attempts to provide a comparative assessment of contemporary political parties in democratic regimes across the world through expert surveys (esp. Kitschelt 2014). ## Political parties and representation: Programmatic vs. clientelistic party systems A contemporary perspective on party systems has been developed by looking at linkage strategies. Kitschelt (2000) distinguishes three types of linkage strategies: programmatic, clientelistic and charismatic. Most of the succeeding work has focused on the first two linkage strategies only. Programmatic political parties are the democratic ideal: parties provide distinct and cohesive electoral platforms and implement a broad policy agenda if they win office. Contrary, clientelistic strategies rely on providing selective material benefits to voters after an election that are contingent on electoral support. The linkage strategy has significant implication for the organizational structure of a party and its potential durability. Recent work on linkage strategies has helped to better understand a lack of party system institutionalization in many democratic countries across the world. Today, we have developed an excellent knowledge about the evolution of party systems in particular regions of the world. This work takes as a starting point a common understanding about the role of parties in political regimes by looking at linkages. However, it differs significantly with respect to the analytical focus by putting different emphasis on programmatic profiles, nationalization, volatility, populism and clientelism in particular regions. Due to the different emphasis that is put on the study of party competition by scholars of the regions, we have still not been able to develop a comprehensive understanding about the evolution of party systems across different political regimes and time. There has been significant progress in our understanding about democratization and regime dynamics (Przeworski 2010). This work has also thrived and looked for new measures that allow us to assesses all political regimes across time, in particular recently with the V-Dem project (Coppedge et al. 2015). A similar drive to assess party systems evolution in a global and historical perspective is still lacking. ## The need for a global assessment of party system evolution Scholars of party politics have the theoretical framework, the analytical tools and measures to assess party competition across the world and time. To analyze the global evolution of party systems, we are in need of coherent data about political parties. This must include key information about electoral performance, linkage strategies, organizational characteristics and leaders. Most of this information is already available through political handbooks and election archives but has yet to be transformed into a modern data infrastructure. For example, the yearly editions of the Political Handbook of the World that have been published since 1927 allow us to derive the respective information about the origins of parties, their programmatic profiles over the years and key leadership information about prime ministers, presidents and party leaders. Developing a coherent definition of parties and measuring their linkage strategies over time is the true challenge. Both, Kitschelt’s theoretical work and his recent empirical data provide a valuable point of departure (Kitschelt 2000, 2014). The handbooks about political parties and events are a rich source of descriptive information that forms the basis of such a theoretically guided mapping of political parties. We aim to define and gather a core set of information about political parties across the globe that allows us to analyze the evolution of party systems. The set of countries is limited to about 70 chosen by their size and with at least one longer period of democratic rule (see also Kitschelt 2014; Mainwaring and Zoco 2007). For example, selecting countries with at least one democratic period of 10 years and 5 million inhabitants in 2010 would lead to 67 countries (Europe 21, Americas 18, Asia/Oceania 16, Africa 12). For the total set of countries selected at the workshop in Nijmegen, we later collect and combine data about the electoral performance of parties (based on Nohlen), harmonize party definitions and add information about the profile of parties (linkage strategies), their access to government and leadership information in an integrated data infrastructure. From this data, we can derive standard indicators about party system dynamics such as the effective number of parties, polarization, volatility, access to government and leadership accountability. On the basis of this information we will be able to develop a more coherent analytical understanding about the evolution of party systems. How long do parties last? What are the conditions for programmatic parties? Which levels of electoral volatility stabilize a party system? Are leaders held accountable for electoral and economic performance? Most of the information that we need for our study is already publicly available. Nevertheless, it is scattered, based on different definitions and hard to combine. ## An agenda to map party system evolution across the globe Our analytical goal is to assess different linkage strategies of parties in the world across time. For the later analysis, we collect data in three steps. First, we create a dataset about election results for relevant political parties, mainly on the basis of the Nohlen series. Second, we harmonize the party information from the election results on the basis of a coherent definition of political parties and party change (cf. Barnea and Rahat 2011). We distinguish genuinely new parties and link predecessor/successor parties in the Party Facts data infrastructure (Döring and Regel 2014). Finally, we record a set of key information about each party (esp. programmatic profile, organizational structure, party and executive leadership) on the basis of existing handbooks. To map party profiles, we aim to combine recent innovation in measuring party profiles through expert surveys (esp. Kitschelt 2014) and earlier document based approaches to code party information (Coppedge 1997; Janda 1980). The initial set of information will be based on a coding of existing descriptive accounts of political parties in handbooks. Later, an online survey will allow experts to provide their evaluation of party profiles. In our coding and the survey, we focus only on the main linkage strategies, the programmatic profile, key organizational characteristics and leadership information. This is due to the limited amount of information that is available for some parties, especially smaller parties from the first half of the 20^th^ century. A similar approach has been successfully used to develop better measures of democracy through the V-Dem project (Coppedge et al. 2015). In terms of scale, our goal is more modest: develop a core set of indicators for relevant political parties in all major countries with at least one longer spell of democracy. We also differ from other major efforts that aim to provide more detailed information about a particular group of countries such as the Political Party Database Project ([www.politicalpartydb.org](http://www.politicalpartydb.org)) for advanced high income democracies. ## Feasibility of the project and preparatory work We have evaluated the feasibility of the project by collecting a systematic set of information about party system evolution for *nine countries* in three different regions (Europe: Germany, Sweden, Poland; Latin America: Brazil, Chile, Ecuador; Asia: Japan, Indonesia, Philippines). For these countries we have collected *election results* for the main parties (see Table 1). These election results form the basis for a coherent definition of political parties and to link predecessor parties. We have also collected an initial set of information (pdf-copies) from party handbooks and almanacs for these countries. This information is used as the basis for our discussions about measuring party profiles and linkage strategies and to develop an initial version of the codebook in Nijmegen. Table 1 – Summary information pilot project for nine countries first year parties elections ------------- ------------ --------- ----------- Brazil 1945 22 19 Chile 1915 28 22 Germany 1898 19 31 Ecuador 1947 31 20 Indonesia 1955 16 11 Japan 1902 34 41 Philippines 1907 25 30 Poland 1919 38 24 Sweden 1890 16 39 Finally, one of the proposal participants has developed Party Facts an online database that combines information about political parties from existing data sources (Döring and Regel 2014). It includes more than 3000 parties and combines more than 10000 parties from external datasets among them prominent sources about political parties as the Manifesto data, the Parliaments and governments database (ParlGov), the Constituency Level Election Archive (CLEA), the European Journal of Political Research Political Data Yearbook (EJPR PDY), the World Value Survey (WVS) and various expert surveys (Benoit/Laver, CHES, Kitschelt, Rohrschneider/Whitefield, Wiesehomeier/Benoit). This source is valuable for the project for three reasons: first, it provides an online platform to harmonize information about political parties; second, it gives an interface to data that is of relevance for our data collection; third, it allows us to validate our document based measures with existing information. ## Publication plan The long-term goal of the project is to write *two edited volumes* about the global evolution of party systems on the basis of harmonized measures about political parties in the world since 1900. A first volume would give a *descriptive account* about the evolution of party systems in the world by comparing the different regions. A second volume would take a more *analytical perspective* and assess different dimensions of party system evolution such as electoral volatility, economic voting, government access, the life cycle of parties and leadership accountability. The data collected for these publications allows for a more fine grained assessment about the evolution party systems and will be publicly available. The respective information will also form the basis for peer-reviewed publications by members of the group. ## Funding We aim to find financial support for a *mid-size funding line* given the mix of expertise needed, the heterogeneous network of political scientists from Europe and the US as well as the junior to mid-level career status of many project participants. We plan to support the project with funding through a European research organization. Ideally, we would succeed in funding the project through Open Research Area (ORA), a funding scheme between the ANR (France), DFG (Germany), ESRC (UK), and NWO (Netherlands). One project participant has suggested a specific national funding scheme and demonstrated his commitment to submit a grant application to fund the network. Alternatively, we may fund the project through smaller funding lines from our home institutions. A large scale funding such as a Horizon 2020 project is not needed to successfully implement the project. Our initial data collection for the nine countries, gathering election results and harmonizing political parties, has provided us with a budget estimate for the first two steps of the data collection. To evaluate the potential costs of the last step in the data collection (party profiles) further discussions among project participants are needed. Joint work at the Research Sessions would provide an agenda for preparing a grant application and more detailed cost estimates. A meeting in Nijmegen would also allow us to advance the pilot study through our preparations and discussions at the workshop. # References _removed -- word limit ECPR online form_