Introduction

Since the first pioneering electoral analyses carried out in the United States and Europe at the beginning of the 20th century, the commonality and simultaneity of electoral change has been considered an important indicator of the degree to which a political system is integrated. Common shifts of votes occurring between elections in all territorial areas of a political system mean that the system's parts move together and that everywhere its electorates respond similarly to common factors and issues. Such general simultaneous change of votes towards or away from party families points to an advanced degree of integration of political systems. This classical indicator of vote shifts allows assessing the presence of “waves” or “tides” of electoral change across Europe.

While in the past uniform change has been investigated across constituencies at the national level, this chapter uses this conceptual tool to analyze the formation of a Europe-wide electorate. The detection of waves can be interpreted as an indicator of a “Europeanized” electorate that refers to common supra-national issues, is affected by similar vote determinants, or undergoes similar structural change. When used at the level of national electorates and party systems, shifts of votes are considered an important indicator of the degree of “nationalization” of electoral politics. This chapter extends the use of this indicator to assess the degree of “Europeanization” of electoral politics. Insofar as one can detect common waves of change that take place simultaneously everywhere in Europe – or, at least, in most European political systems – we can speak of a Europeanized or integrated dimension of electoral politics.

To what extent is – in a long-term perspective – electoral change common and simultaneous in Europe? To what extent do national electorates move together in waves of change over the past century and a half? What are the most important waves in the past and how do recent waves compare to those fundamental transformations? Are the waves of the past mostly along the left–right axis while since the de- and re-alignment in the 1970s such waves take place along other dimensions of contestation?

Introduction

Empirical evidence presented in the five chapters of Part II provides ample support for the thesis of the Europeanization of politics. The crucial phases in the late 19th and early 20th centuries of the formation of party systems witness steady homogenization processes and simultaneous waves of change across the continent – in both Western and Central-Eastern Europe. Similar alignments are put in place which – in the past 30 years – have also permeated the EP. The ideological cohesiveness within Europe-wide party families has been high since World War II and it is likely that this indicator, too, has its roots in early periods of party systems’ formation. Executive politics has undergone cross-national convergence under the impact of the recent events of European integration and the end of the Cold War.

The main line of interpretation that transpires through the evidence points to the role of the left, that is the role of socialists and social democrats: “class politics” and the left–right cleavage. This is the dimension that, by structuring national party systems, is the fundamental feature of their similarity. Its emergence in the late 19th century represents the main swing which, up to the present time, has remained unparalleled. In its wake territorially differentiating factors of a cultural nature and pre-industrial age are replaced by a functional alignment cutting across territories. Social democrats appear as the most integrated party family and it is their ideological shift in the early 1990s that causes the convergence of cabinet politics. Europeanization is to a large extent the story of the left and its relationship to the market economy.

The goal of this chapter is to investigate more systematically the sources of Europeanization along the explanatory scheme sketched in Chapter 1.

In 1774 Edmund Burke, addressing the electors of Bristol, said:

Since its first use in the English parliament, representation was based on a territorial, geographical, indeed local principle. Burgesses and knights summoned by the king were representatives of towns and counties (the Commons). For long representation was imposed from above and with democratization the local geographical definition of representation did not change immediately (Morgan, 1988: 43–6). With Burke, however, representation acquires a new meaning: not the representation of particular interests from various territorial constituencies but, as famously stated in the same quote, the vision that “Parliament is not a congress of ambassadors” but rather the site where the same interests across territorial units – say, trade interests in places such as Bristol or Birmingham – are represented. Further, as Pitkin discussed in her seminal treatise on political representation, even towns or regions not enfranchised or represented territorially through a member of parliament may nonetheless be “virtually” represented in a communion of interests and feelings (Pitkin, 1967: 173). More recently, Manin too has written about the importance of representation being “national” rather than local in its historical development (Manin, 1997: 163). In the terminology used in this book, the functional dimension of representation, as opposed to the territorial one, predominates (see Chapter 1).

Historically, in national political systems the actors through which nationwide cross-local interests are represented since the earliest phases of democratization are political parties. Political parties have the role of aggregating the different interests in coherent programmes and ideologies but also of linking the different territorial bits of a country together – in other words, making national representation possible. This is done through the cross-district “linkage” or “coordination” of parties (Cox, 1999) or, what in a different terminology is referred to as their “nationalization, that is the progressive replacement of territorial divisions in the representation process through functional dimensions of contestation.

Introduction

The chapters so far have analyzed processes of cross-country convergence that the literature seen in Chapter 1 mostly traces back to within- and trans-country processes. This chapter introduces aspects that may be ranged under “top-down” effects, namely factors located at the supra-national level that affect countries in a similar way and thus create convergence. At the same time this chapter extends Chapter 6. While including an analysis of long-term trends since 1945, it focuses on Europeanization in the more usual definition of the impact of European integration on member-states. The hypothesis about the impact on cross-country convergence is addressed in an explanatory perspective. Beyond this impact, however, this chapter addresses further supra-national factors such as the breakdown of communist regimes in 1989 and the impact of such a momentous juncture on cross-country convergence.

Furthermore, while this book has so far focussed on the Europeanization of electorates and party systems, this chapter turns to the arena of cabinet and coalition politics. The move to the executive arena is needed first and foremost because the top-down effect from supra-national factors such as European integration is felt primarily in cabinets’ format and policy. It is at the level of cabinets and governmental coalitions that one observes programmatic changes that affect policy making. While small, extreme protest opposition parties are not called to translate their platforms into actual policies, programmes of parties with cabinet responsibility provide the basis for executive action.

Indeed, it is the programmatic action of cabinets that has been most affected by Europe-wide changes such as the end of the Cold War and European integration. On one hand, the breakdown of communist regimes in Central and Eastern Europe had the effect of de-radicalizing leftist politics and reducing the alternatives in foreign and economic policy. This has led to claims about the policy convergence through liberalizations and privatizations, as well as calls for a reduction of the welfare state and of general state intervention in the economy. On the other hand, the acceleration of the process of European integration since the Single European Act of 1986 and the Maastricht Treaty of 1992 has constrained the margin for action on the part of national governments.

Introduction

Following three chapters on the Europeanization of party systems from the perspective of their “format”, the analysis in this and the next chapter turns to the perspective of their “contents”. One of the critiques made against studies of the national integration of parties and party systems is precisely that – in spite of similar levels of support and change over time – the content of the ideology and policy programmes of local “branches” are too different to consider them the “same” party. The names and symbols may be the same, but the local context imposes its constraints so that, say, the French Communist Party in Bretagne is a very different thing from the French Communist Party in the eastern banlieue of Paris.

The critique has been formulated against nationalization approaches, especially for some countries. In most cases, the strong centralization of party organizations at the national level and the fact that competition is national rather than local (or, more precisely, becomes national in the process of nationalization), makes cross-regional ideological variation rather unproblematic. At the European level the problem of “sameness” of parties of different countries bundled together in European party families, however, needs to be addressed in a systematic manner. The classification of national parties into families is the first step (seen in Chapter 2). The analysis of the ideological cohesiveness of parties belonging (or classified as belonging) to the same party family is the second step. This is the goal of the present chapter.

With Europeanization one observes homogenization of electoral support over time, similar cleavage structures and party systems across countries and across levels of governance, and simultaneity of change.

At the turn of the 20th century politics in Bulgaria was dominated by a large liberal party representing the democratically oriented bourgeoisie and independent entrepreneurs who favoured constitutionalism, parliamentarism, and social policies while the main opposition was constituted by conservatives of the richer upper classes and clergy. These elitarian parties ruled until World War I when the challenge coming from the parties of the “masses” – the socialist BRSDP and the agrarian BZNS – became majoritarian among the enfranchised male electorate, only to be in turn overtaken as soon as 1920 by the Bulgarian Communist Party. An unstable party system consisting of socialists, agrarians, and communists on the left and liberals and conservatives on the right managed to survive until 1931 – the last democratic election until the fall of the Berlin Wall in 1989.

Such early party system development is typical of Europe and appears in classical accounts of political parties from Ostrogorski (1902), to Michels (1911), Duverger (1954), and Rokkan (1970), among many others. In fact, it is the story of most European party systems whether of Britain and its early parliamentarism; of Germany, Italy, and Switzerland as they formed from the struggle for national unification; of Austria and Hungary and the other nations emerging from the break-up of multi-national empires in Central and Eastern Europe; or of the Nordic countries. Indeed, a contemporary commentator simply described the Bulgarian party system as “European”. The early opposition between liberals and conservatives during the periods of restricted suffrage, the rise of parties of mass mobilization with the extension of franchise to workers and peasants, the division of the left into socialists and communists after the Soviet Revolution of 1917, and even the breakdown of democracy under totalitarian ideologies in the 1930s are common features of electoral history during the constituent phases of European party systems. Such commonality persists in later phases of electoral development after World War II with the rise of Christian democracy and the welfare state, the new politics of emancipatory values eventually leading to green parties, and the radical-right politics of anti-globalization and anti-immigration which have recently manifested in populist parties throughout the continent.

In 2004 The Nationalization of Politics was published, a book on the formation of national electorates and party systems from the 19th century until the present time. In concluding that study I hinted at one possible, natural continuation of that line of work – namely, the analysis of the incipient supra-national integration and formation in Europe of electorates and party systems; their “Europeanization” as it were.

The present volume is a step in that direction. One of the lessons, perhaps the main one, from the nationalization study was that – within the boundaries of the nation-state, the most important political unit over the past two centuries – politics transformed from territorial to functional, with non-territorial dimensions of contestation (first and foremost the left–right class cleavage) overwhelming and replacing territorial ones. With the blurring of national boundaries through European integration, the question is whether today “nationalization at the European level” is taking place along the same lines of nation-states at an earlier stage of political development. Research for this volume, however, soon showed that, rather than “incipient”, electorates and party systems have long been “Europeanized”.

This study thus investigates in a long-term and historical perspective the extent to which electoral politics (voting behaviour, party systems, ideological families) transforms from national, that is centred on national-territorial specificities, to European, with alignments cutting across national units. Similarly to the work on the nationalization of politics, it is an analysis of democratic integration. Europeanization is operationalized as homogeneity among nation-states in Europe and uniform, simultaneous electoral shifts. Further, it is defined in terms of ideological cohesion among parties of the same family across Europe and of vertical correspondence between national and European electoral arenas. Also similarly to the work on nationalization, this study challenges the conventional view that such integration began after World War II. It shows, on the contrary, that the Europeanization of electoral politics is a much longer process, starting in the 19th century, and that a European electorate has long existed and manifested itself through common and simultaneous “waves” of change across the continent – both Western and Central-Eastern Europe.

The analysis presented in this book is based on four types of data:

• Aggregate data (electoral results): For national elections, the source is a new updated database compiled for the present study based on previous efforts in historical data collection (for the election of sources on Western Europe see Caramani, 2000: 1017–53; on Central and Eastern Europe the main source is Nohlen and Stöver, 2010). For elections to the European Parliament (EP) since 1979 the main source is the European Database at NSD in Bergen.

• Individual data (surveys): The sources of data are Eurobarometers (EB, Mannheim Trend File) and European Social Surveys (ESS).

• Text data (party manifestos): The source of data is the Comparative Manifesto Project (CMP). See Budge et al. (2001) and Klingemann et al. (2006), as well as online updates.

• Membership data (cabinet composition): The source of data is the ParlGov database (Döring and Manow, 2010).

Introduction

Correspondence refers to the overlap or, conversely, the distinctiveness between the two levels of elections in Europe, national and supra-national. This chapter investigates the extent to which there is a distinct electorate and party system at the EU level, independent from national issues, party platforms, and alignments. The vertical correspondence of electoral behaviour between national and European elections is analyzed through a comparison of national elections and elections to the EP. Is there, and, if so, to what degree, a differentiated European party system or, on the contrary, is electoral behaviour in European elections a reflection of national political contests? And is there a trend in the progressive “emancipation” of the European party system from national politics over 30 years of European elections? This chapter investigates the origins of distinctiveness based on swings towards and against party families based on the previous chapter.

Since the first direct elections to the EP in 1979, European elections have been described as “second-order elections” (Reif and Schmitt, 1980). This label implies not only that European elections are (perceived as) less important than national elections in the member-states of the EU, but also that European elections are dominated by national factors. European electorates vote for the EP according to national criteria, campaigns are run on national issues, voters are guided in their voting choices by national party affiliations and electoral alignments, and leaders seek support on national policies and platforms. In sum, European elections are national contests “by other means”.

Over the past 30 years the EP has acquired new competences and strengthened those it already had. Decision making is systematically affected by European party federations, and groups consolidated in the EP. More generally, the process of integration has moved at an accelerated pace and public debates have gained strength in the wake of referenda on accession, common currency, and treaties and with the process of the EU's “constitutionalization”.

Introduction

The task of this chapter is to identify the dimensions of Europeanization from a broad comparative and historical perspective. Concepts from various integration theories are relevant for an encompassing view of Europeanization. This chapter attempts to bring them together to derive indicators for assessing the degree to which Europeanization processes have taken place over the past 150 years. As “Europeanization” is analyzed historically and intended as referring to “nationalization” processes on a vaster scale, the definition given in this book as long-term convergence between national electorates and party systems in Europe is unavoidably broader as well as focussed differently than most definitions.

This definition is broader first of all in temporal terms as it views Europeanization as a process not necessarily linked to European integration. Convergence between national electorates and party systems may be independent of the constitution of a supra-national level of governance (institutions and policies) and therefore may have taken place before European integration since the 1960s. This definition is broader also insofar as convergence is not necessarily related to EU issues or to a cleavage on European integration. It is therefore distinct from “EU-ization”. The long-term convergence of electorates and party systems may be detected in issues other than EU integration, for example in how environmental issues structure publics in European countries, or in similar anti-immigrant reactions to globalization. This definition is broader, finally, because the causes of Europeanization are identified in factors other than EU integration only. While centralization of competences and coordination of policies at the European level clearly have an impact on the degree to which electorates and party systems homogenize over time, this book includes other factors as well.

Additionally, this definition of Europeanization has a different focus. The focus of this study is on the structuring of mass politics and views Europeanization from an electoral perspective. It is therefore different from work on the impact of EU governance on national policies and administrative structures. While there is work on electorates and party systems in a “European integration perspective”, it is usually a literature limited to the European question, that is favourable versus unfavourable views of integration as expressed by voters either in referenda or through anti-Europe parties, or in the position established parties take on Europe (i.e. Europeanization as politicization of the cleavage on European integration).

Introduction

The analysis that follows in the five chapters in Part II of this book is based on a data collection covering 713 national and seven EP elections spread across 30 countries over roughly 150 years of electoral history in Europe. This chapter addresses general issues of cross-sectional and cross-temporal comparability that such a broad comparative and longitudinal analysis entails. This chapter further addresses other general methodological issues to clear the way as much as possible for the analysis in the rest of this book.

As seen, the analysis of Europeanization of electoral politics is multi-pronged in the type of data it relies on. Yet – whether analyzing electoral results, party manifestos, cabinet data, or surveys – the basic units of analysis are political parties and party families. The homogeneity of electoral support across countries is based on a definition of party families. Similarly, the uniformity of electoral swings in Europe is based on simultaneous gains or losses of equivalent parties in different countries. The level of ideological cohesiveness and convergence among European parties of the same family (based on manifestos and voters’ preferences) also requires a classification of national parties into Europe-wide families. Finally, the analysis of similarity of cabinet composition across countries has party families as its basic unit. In all cases party classification is crucial for the analysis in this book.

The first task of this chapter is, therefore, to build party families and a classification useful for diverse countries and over a period of more than a century (first section of this chapter). To do this, what henceforth is referred to as “the cleavage model” is used, and various aggregations of families are proposed. This section also addresses aspects relating to the data file, codes, and overall indices as the number of parties, fractionalization, volatility, and disproportionality. Second, this chapter provides an overall picture of the “European” party system (second section of this chapter). This serves as a reference for the analysis of “deviation” of single countries or areas in Europe. It is the overall “model” against which variation of national systems is measured in Part II. Time periods covered for each country and issues relating to different electoral years are also addressed in this section.