Abstract
This Special Issue aims to (1) theorise party strategies in multi-dimensional policy spaces; and (2) apply the theory to party competition in multinational democracies characterised by a salient territorial dimension alongside a more established economic dimension. The introductory article brings together recent contributions treating spatial and salience theories as compatible and policy spaces as two-dimensional to propose four party strategies that can be ranked from one- to two-dimensional competitive behaviour: uni-dimensionality, blurring, subsuming, and two- dimensionality. The remaining contributions operationalise these strategies and draw on a variety of data sources ranging from manifestos to parliamentary bill proposals and expert surveys to describe when and explore why parties use these strategies in competition, focusing on patterns of party competition in multinational democracies, selected as typical cases of multi-dimensional competition.
Keywords
1. Introduction
A number of recent developments have taken place in the party competition literature that broaden our understanding of party strategies in political competition. Firstly, the assumption that political spaces are one-dimensional has been challenged since a single left-right dimension “is steadily diminishing in its ability to summarize party behaviour” (Albright, 2010: 714). While economic issues constitute the most important dimension along which parties compete in most countries (Huber and Inglehart, 1995; Wagner 2012), social and cultural issues have gained in importance since the 1970s (Inglehart and Flanagan, 1987). Moreover, it has been argued that in many countries, the cultural and the economic dimensions do not neatly correlate with each other anymore so that the political space cannot be characterized by a single left-right axis but has to be depicted as two-dimensional (Kitschelt, 1994; Kriesi et al., 2006, 2012; Marks et al., 2006). Following cleavage theory (Lipset and Rokkan, 1967), dimensions of competition reflect the cleavage structures underpinning the party system. Since political parties are rooted in particular cleavages, they are more vested in the dimensions that reflect these cleavages, i.e. they have a core dimension. At the same time, they sometimes cannot afford to ignore other dimensions within the same party systems, even though they are secondary to them.
Secondly, spatial and salience theories of party competition are increasingly seen as complementary rather than competing approaches (Alonso, 2012; Basile, 2012; de Sio and Weber, 2010; Meguid, 2005, 2008; Rovny, 2013; Rovny and Edwards, 2012; Wagner, 2012). Parties can challenge each other by altering their position on an issue dimension and their emphasis of political issues. These recent advances help us understand how parties behave strategically within a two-dimensional space where the two dimensions are not equally important to them. Spatial theory on its own would expect parties to confront each other by positioning themselves on both dimensions. Salience theory on its own would expect parties to selectively emphasise issues favourable to them, but would fail to acknowledge that there are dimensions rooted in societal conflict that parties may not be able to ignore. Together they help us explain how parties both respond to and try to shape political spaces (Rovny and Edwards, 2012: 53) where different dimensions are not equally important to them.
This special issue therefore seeks to combine insights from these recent theoretical advances to present a more fine-grained theory of how parties rooted in one dimension deal with a second dimension in competition. It seeks to answer the following two questions:
What kinds of strategies are available to political parties when dealing with a second dimension in competition?
When and why do parties choose which strategies?
This introductory article focuses on the first question. We follow Basile (2013: 69–75) in arguing that when challenging their competitors, political parties choose from a tool box consisting of positioning, selective issue emphasis, and issue framing. We then propose a set of four strategies resulting from using elements of this tool box that parties can use to deal with what is their secondary dimension in competition: a uni-dimensional strategy, where parties selectively emphasise and position on their core dimension while ignoring the second dimension; a blurring strategy (Rovny, 2013: 5–6), where parties adopt “vague, contradictory or ambiguous positions” on the second dimension; a subsuming strategy, where parties frame issues associated with the second dimension in core dimension terms; and a two-dimensional strategy, where parties position themselves on both dimensions. The empirical contributions to this special issue then draw on a variety of methods and data sets to describe the strategies, and build hypotheses explaining why parties choose each of them.
The aim of this contribution is to suggest a set of strategies that can in principle be used by parties dealing with a second dimension in competition, whatever the substance of the dimension in question. However, the contributions to this special issue have a common empirical focus, in that they examine patterns of party competition in multinational democracies where multi-dimensional competition is typical. In multinational democracies such as Spain or Belgium, parties have to navigate in a policy space constituted by a territorial dimension alongside a left-right dimension. To date the more general party competition literature has paid surprisingly little attention to multi-dimensional party competition arising from the presence of an ethnonational (Szöcsik and Zuber, 2015) or territorial (Alonso, 2012) dimension alongside the more commonly analysed economic and social value dimensions. Scholars have often subsumed territorial issues or issues related to national identity under a cultural or a ‘new politics’ dimension (Spies and Franzman, 2011: 1053). Alternatively, ethno-regionalist parties’ electoral success has been explained as a function of the strategic reaction of mainstream parties on the single issue of decentralization (Meguid, 2005, 2008). By contrast, scholars of territorial politics have documented the increasing salience and electoral/political implications of territorial issues in many places, but have only recently begun to pay attention to the role of party strategies and the consequences for patterns of party competition (Elias, 2015; Hepburn and Detterbeck, 2013; Libbrecht et al, 2009; Massetti and Toubeau, 2013).
Focusing on multinational contexts implies theorising and empirically analysing the strategies of two main types of parties. They have been referred to under different labels: state-wide/ethnic majority/mainstream parties on the one hand, and regionalist/ethnic minority/ethno-regionalist/autonomist/secessionist/minority nationalist/stateless nationalist and regionalist parties parties on the other hand. The starting point for this Special Issue is that whereas the exact terminology used to refer to these parties may differ depending on the meaning of the label in a given context and the definition preferred by authors, what unites situations of party competition in multinational contexts is that they involve strategic interaction between one type of party for which the core dimension of competition is the economic dimension and one type of party for which the core dimension is the territorial dimension (see also Alonso, 2012: 40). For ease of analysis, this article refers to the first type of party as “mainstream party” and the second type of party as “ethno-regionalist party”.
The rest of this article is organised as follows. In the next section of the paper, we present our theoretical framework. We discuss positioning, selective emphasis, and issue framing as the strategic tools available to political parties competing in a two-dimensional space consisting of an economic and a territorial dimension. In Section Three, we define the four strategies and illustrate them with examples from multinational contexts, as well as other settings where there is a second dimension of competition and where we believe an analogous logic to be at play. Finally, in Section Four, we present the other contributions to the Special Issue.
2. Theoretical framework: Position, selective emphasis and framing
Conceptualising party competition
We define party competition with Franzmann (2011: 320) as follows: “Party Competition is an institution in which parties strategically cooperate or contest as political actors to gain political power.” The definition implies that upon deciding its overall strategy in competition, a party has to (1) distinguish the parties it seeks to cooperate with (e.g. by forming an electoral alliance or a coalition government) from the parties it seeks to contest, and in the latter case, (2) choose which strategy it wants to apply in contest. 1
It is this second choice that we seek to conceptualise in more detail in this article. To this end, we draw on recent contributions that have begun to combine elements from two prominent theories explaining the strategic choices parties make when contesting others: spatial theory in the tradition of Downs (1957) that expects parties to contest each other by positioning themselves on dimensions, and salience theory in the tradition of Robertson (1976), Riker (1986), and Budge and Farlie (1983) that expects parties to contest by selectively emphasising political issues.
It should be noted that this definition of party strategies does not limit us to parties’ strategic behaviour during electoral campaigns. Rather, we follow Benoit and Laver (2006: 37) in treating political competition as a continuous process that is structured into two distinct phases by two institutional automata – an electoral system and a legally mandated maximum inter-election period. […] They distinguish what we might think of as the ‘electoral’ phase of political competition from the ‘inter-electoral’ phase.
The excellent classification provided by Steenbergen and Scott (2004) summarises the main differences between positional and salience theory as follows: the positional account states that parties contest each other by changing their positions on issues, moving within the political space. This account takes issue salience to be exogenous and party positions to be endogenous to competition. the salience account states that parties contest each other by selectively emphasising issues favourable to them and de-emphasising issues that are unfavourable to them, thereby defining the political space. Here issue salience is endogenous, whereas party positions are exogenous to competition.
However, the authors also note that “in reality, party competition involves strategic choices on both issue salience and issue positions, and a fully specified analysis should treat both of these elements [position and salience, authors’ note] as endogenous” (Steenbergen and Scott, 2004: 167, Note 1; see also Budge and Farlie, 1983: 270, 303–304). More recently, a number of authors have also acknowledged this, and have started combining insights from both accounts.
A first case in point is the work by Meguid (2008), who proposes a Position, Salience, and Ownership (PSO) theory of party competition. Acknowledging that issue salience is not stable during campaigns and that “voters are not indifferent between parties promising the same policy positions” (Meguid, 2008: 274), she proposes to add the tactic of altering the salience of issues to the standard tactics of policy convergence and policy divergence emphasised by spatial theorists. Meguid (2008), however, still treats the competitive space as one-dimensional. She merely adds the possibility that parties may selectively emphasise niche issues that are by definition unaligned with the left-right dimension of competition.
The literature on niche parties stands in the spatial, Downsian (1957) tradition of viewing party competition as one-dimensional and consequently treats the emergence of new, unaligned issues as a temporary upheaval rather than an indication of genuine multi-dimensionality. In contrast, Alonso’s (2012: 42) theoretical work is inspired by European multi-party systems where political spaces are often multi-dimensional to begin with. She argues that parties use both positioning and selective emphasis within a two-dimensional space, consisting of a left-right and a centre-periphery dimension. Importantly, she shows that the dimensions do not have equal weight for all parties alike, and that, as a consequence, their behaviour on what they consider to be their primary dimension differs from their behaviour on their secondary dimension. A party’s reputation and its goal to gain and defend issue ownership limit its strategic flexibility on its primary dimension. In contrast, parties have more strategic flexibility on their secondary dimension, “moving from a catch-all to a positional tactic and even leapfrogging between issues” (Alonso, 2012: 42).
Rovny (2013) also theorises party strategies in a multi-dimensional context, and adds to Alonso’s insights about parties being more concerned about their primary dimension the possibility that parties may deliberately avoid positioning (or they will blur their position) on their secondary dimension. Rovny’s (2013) position avoidance or position blurring is therefore different from Meguid’s (2008: 274) dismissive strategy. Rovny assumes a multi-dimensional space where parties can choose to position themselves on one dimension, but avoid positioning themselves on another in order not to divide their voters. By contrast, Meguid assumes a one-dimensional space where mainstream parties respond to niche parties that emphasise new issues, unaligned to the left-right dimension. The dismissive strategy then consists in trying to keep the salience of a new, unaligned issue low.
Finally, Rovny and Edwards (2012) have argued that parties can deliberately try to subsume new issues into their primary dimension, thereby aiming to alter the dimensional structure of the political space itself. Issue subsuming is neither the same as positioning, nor is it the same as selective emphasis. It essentially requires rhetorically re-framing a new issue in terms of the dimension for which a party is perceived to be competent by the voters: When mainstream parties engage newly introduced issues, they do so with the ultimate aim of translating them into their ideological issue-bundle. This strategic dynamic eventually leads to the absorption of the new issue and to the realignment of party positioning along an updated political continuum, effectively amounting to the ‘turning’ of the competition axis. (Rovny and Edwards, 2012: 61)
Taking both issue salience and party positions to be endogenous and exogenous to party competition involves no contradiction, as long as we differentiate precisely what is exogenous and endogenous for whom and when and between individual party strategies at the micro level and the structure of the political space at the macro level. 2 A single party P at a given point in time t does not choose its strategy in a vacuum, but finds itself in a society where public opinion polls show voters to already have certain preferences, where public discourses and party ideologies are already linking positions on issues in particular ways, and where some issues are perceived to be more important than others. To party P at t, the structure of the political space (in terms of the correlations between party positions on issues that define dimensions and in terms of the systemic salience of individual issues) is indeed exogenous and we as researchers can seek to explain this choice taking voter preferences and issue correlations and issue salience as parametric. However, rather than limiting itself to moving along issue dimensions already defining the space at t, P may selectively emphasise a new issue I, that has so far not been salient. Let us assume that P manages to successfully plant its programmatic statements receiving a lot of media coverage and creating an impression of importance of I among voters. As a consequence, I may have turned into a salient agenda issue by the time an election is held at t+1, forcing other parties to respond. For P at t, the structure of the space (dimensionality) is thus exogenous, whereas the salience of I for P is endogenous. By contrast, at t+1, for both P and all other parties, the salience of I is exogenous. If several parties then try to additionally link I to already existing dimensions, or correlate their stances on I in a way that upsets the previous issue-correlation pattern defining the dimensions of the political space, the structure of the space itself also becomes endogenous.
We can therefore conclude with Rovny and Edwards (2012: 53) that [w]hile citizen preferences underlie the issue composition of political space, it is political parties that—partially and strategically—translate these issues into political conflict. Political competition becomes a struggle over issue linkages, that is to say, over the dimensional configuration of political space.
Issues and dimensions in party competition
We now turn to conceptualising the political space in the multinational democracies this special issue focuses on. In principle, the political space is n-dimensional once we allow for the capacity of parties to compete not only within, but also about the definition of the political space (which can change its structure as a result of strategic interaction between political parties). However, following Benoit and Laver (2012: 196), mapping political spaces is always a theoretical effort and can never be achieved in a fully inductive way, since political spaces “are ultimately metaphors and both the dimensions spanning these spaces and agents’ positions on these dimensions are fundamentally unobservable”. As such, our interest is in the two-dimensional space constituted by an economic and a territorial dimension, since this is the most appropriate for theorising and empirically analysing party strategies in multinational democracies. 3
To establish whether and to what extent individual issues in a given context belong or do not belong to either of these dimensions, we follow Robertson (1976: 70 and 2006: 168) who analyses dimensions as correlations between issues. Following Robertson, we can further talk of a two-dimensional space if parties have positions on at least four issues ABCD, and their stances on A and B and their stances on C and D correlate, but their stances between A and C and between A and D and between B and C do not correlate. 4
The economic dimension is a constitutive part of the left-right axis which is the centre of attention in a uni-dimensional conceptualization of party competition (e.g. Adams et al., 2006; Budge, 1994; Huber and Powell, 1994; Laver and Schofield, 1990; Powell, 2000). However, economic issues do not exhaust the meaning of left-right. Many scholars have also identified a social dimension to the left-right axis encompassing issues such as sexual lifestyle or religious values, and that spans from libertarian or alternative politics to authoritarian or traditional politics (Hooghe et al., 2002; Marks et al., 2006; Rovny and Edwards, 2012: 62; Wagner, 2012). The aggregate view is therefore problematic because it obfuscates the fact that it is highly contingent whether in a given society (1) party positions on social values and economic issues will indeed correlate in a way that allows us to collapse the two dimensions into a single left-right axis of competition (or whether we rather find a truly two-dimensional space with parties positioning themselves in all four quadrants); and (2) if positions do correlate, in which direction (Marks et al., 2006; Rovny and Edwards, 2012). Acknowledging the context-dependent meaning and composition of left-right, we therefore focus on the economic dimension that is cross-contextually meaningful and is also the most important dimension of competition in contemporary democracies (Wagner, 2012). The economic dimension is constituted by parties’ stances on the role of the state in managing the economy. Parties on the economic left emphasize a large role for the state in managing the economy, while on the economic right, parties prioritize a lean state and individual economic freedom (Marks et al., 2006: 156–157).
The second dimension this Special Issue is interested in is the territorial dimension. In multinational democracies, a second set of issues on which political parties may position themselves relate to the centre-periphery cleavage. In their seminal work from 1983, Rokkan and Urwin (1983: 30) ascribe the emergence of this cleavage to “long sequences of migration, centre-building, cultural standardization and the imposition of boundaries” which prompted resistance from historically, economically, culturally, or linguistically distinct communities on the peripheries of states. However, peripheral resistance to processes of state modernisation has not been uniform. The drivers of peripheral mobilisation have varied considerably from place to place. Scholars have paid particular attention to culture, identity, language, and ethnicity as key markers of territorial difference within the state (for a recent example, see Szöcsik and Zuber, 2015), but these factors have not been equally important everywhere. In some places, economic or fiscal interests have been just as, or more, significant in providing a basis for differentiation within the state (Alonso, 2012: 25; Hepburn, 2009: 484–485; Rokkan and Urwin, 1983). The political/institutional demands articulated by political actors in the name of a distinct national or regional group have also differed, ranging from full secession to far more limited demands for cultural or linguistic protection (De Winter, 1998; Massetti, 2009; Szöcsik and Zuber, 2015).
One consequence of this diversity in peripheral interests and aspirations is that scholars have struggled to find a single label for political actors rooted in the centre-periphery conflict (for an overview, see Hepburn, 2009: 480–485). And yet what all these actors have in common is a shared desire for territorial empowerment, whereby empowerment involves seeking to represent and advance the particular interests of the stateless territory—be it referred to as a region, nation, people or Heimat—and where territorial interests may be economic, political, social, cultural or symbolic in nature. (Hepburn, 2009: 482)
3. Party strategies in two-dimensional space
Political parties are assumed to be rational, strategic actors. This may seem an obvious assumption to make, and indeed it is taken for granted in much of the party competition literature. However, initial work on the conceptualisation of niche parties has asserted that such actors (which include ethno-regionalist parties) have little incentive to behave strategically (Meguid, 2005, 2008). On the contrary, as niche parties advance a very narrow set of issues and receive electoral support exclusively on this basis, they will be unable to boost their support by emphasising non-niche issues in their programmes (Adams et al., 2006; Ezrow, 2008; Meguid, 2008: 14–15). However, more recent work has challenged this proposition, and has provided substantial empirical evidence of the strategic capacity of ethno-regionalist parties (Alonso, 2012; Elias, 2009; Gómez-Reino, 2006; Hepburn, 2009; Zuber, 2012). This literature highlights the fact that ethno-regionalist parties often choose to position themselves on issue dimensions beyond their core business, and that such choices are usually driven by a desire to compete with more established partisan rivals for votes. There are therefore strong grounds for assuming that both mainstream and ethno-regionalist parties are strategic actors. This implies that political parties will choose those strategies that best further their goals which, following Strøm (1990), are three-fold: parties can prioritise vote-maximisation, policy impact, or office incumbency (though the pursuit of one goal may undermine another goal, and hence there are trade-offs involved that parties often cannot avoid). From such a perspective, even pure niche party behaviour can be strategic, as parties may consider it in their best interest electorally to focus exclusively on one niche issue in order to not divide their voters (Rovny, 2013). Accordingly, ‘nicheness’ is better understood as a characteristic of a party’s programmatic offer, rather than an intrinsic feature of some party families and not others (Wagner, 2012: 848).
Which strategies are available to parties seeking to advance their goals of policy, office, and votes? In line with the combination of salience and spatial arguments discussed above, we assume that at any given point in time, parties position themselves on already defined, existing dimensions of competition, but may also choose to alter the dimensionality of the space itself in the long run, by selectively emphasising certain issues over others and by framing issues associated with certain dimensions in new ways. Parties thereby choose not only which dimensions to compete on (positioning), but also how much emphasis they place on each dimension (selective emphasis) and how they define issues associated with these dimensions in the programmatic offering that they put forward (issue framing) (Basile, 2013: 32; see also Chaney, 2013, 2014). We focus here on how parties use these tools to deal with a second dimension in competition: a uni-dimensional strategy, where parties selectively emphasize and position on their core dimensions, while ignoring their secondary dimension, thus not using any of the aforementioned tools on the second dimension; a blurring strategy, where parties adopt “vague, contradictory or ambiguous positions” (Rovny, 2013: 5–6) on their secondary dimension; a subsuming strategy, where parties deliberately try to ‘erase’ the second dimension by framing the issues associated with it in core-dimension terms; and a two-dimensional strategy where parties position themselves not only on their core, but also on their secondary dimension. The strategies can therefore be ranked from one- to two-dimensional behaviour of a party and are presented in this order in more detail below.
Firstly, parties position themselves on the dimension they are most invested in and that is associated with their core issues; at the same time they may dismiss the dimension that is of secondary importance to them, thus applying none of the tools on the second dimension but selectively emphasising and positioning on their core dimension. We refer to this as a uni-dimensional strategy, illustrated in Figure 1. For mainstream parties, this would mean exclusive positioning on the economic dimension, whilst for ethno-regionalist parties this would involve exclusive positioning on the territorial dimension. The latter example of strategic behaviour would come close to resembling niche party behaviour as conceptualised by Meguid (2005, 2008), though as discussed above, Meguid does not take the existence of a territorial dimension into account, but merely discusses the single issue of decentralisation. With regard to the cases selected for this volume, small ethnic minority parties in particular may find this uni-dimensional strategy appealing. For example, two Hungarian minority parties in Serbia (The Democratic Community of Vojvodina Hungarians and the Movement of Hungarian Hope) demand territorial autonomy for the Hungarian minority in Northern Vojvodina (thereby clearly positioning themselves on the territorial dimension), but consider other issues such as economic development to be beyond the scope of what a Hungarian minority party can and should cover in its programme (Zuber, 2012). Such a strategic choice is not exclusive to such parties however. New parties, such as Green parties when they first emerged in the 1970s and 1980s, also tend to prioritise their core set of concerns, to the neglect of other issues that are less crucial for party identity (Poguntke, 1987; Richardson and Rootes, 2006).
The uni-dimensional strategy: Selective emphasis of and position on core dimension (Dcore), whilst secondary dimension is ignored (Dsecond).
Secondly, whilst political parties may want to focus on issue positions on their core dimension, they may not want to ignore secondary issues completely. Drawing on the agenda-setting literature, Green-Pedersen and Mortensen (2010: 260–264) argue that individual parties both seek to shape and see themselves forced to respond to the “party-system agenda” that “emerges from the continuous political debate among political parties” (Green-Pedersen and Mortensen, 2010: 260). In particular, governing parties find themselves under great public pressure to say something about issues on the party-system agenda. For example, over the last five decades, decentralisation has been an agenda issue in Italy with the result that mainstream parties simply could not afford to dismiss the issue entirely within their public statements (Basile, 2013). In such a situation, parties may talk about issues associated with the secondary dimension, not to signal a clear position on these issues, but to deliberately blur their position on the secondary dimension. Blurring can be done in different ways, as according to Rovny it means adopting “vague, contradictory or ambiguous positions” instead of a clear ideological stance, with the aim of masking a party’s “spatial distance from voters in order to either attract broader support, or at least not deter voters on these issues” (Rovny, 2013: 5–6; see also Somer-Topcu, 2014, and Tomz and Van Houweling, 2009). A blurring strategy is outlined in Figure 2. It has been shown to be a particularly attractive strategy for radical right parties, who blur their position on the economic dimension so that all voters who share these parties’ nationalistic, anti-cosmopolitan, and anti-immigrant values but have heterogeneous economic preferences can be simultaneously addressed. Mirroring this behaviour of radical right parties, large mainstream parties of both the left and the right side of the political spectrum may be tempted to blur their position on immigration, an issue potentially dividing their electorate: “Liberal mass parties face conflicts between unions who favor restrictive policies and liberals and ethnic groups who favor expansionist policies; conservative parties are divided between employers who favor expansionist immigration policies and cultural conservatives who favor restrictive policies […] Since seeking to exploit immigration as an issue is as likely to divide one’s own party as one’s opposition, mass parties are reluctant to do so” (Perlmutter, 1996: 377).
The blurring strategy: Position on core dimension (Dcore) and blurring (vague, contradictory, or ambiguous positions) on secondary dimension (Dsecond).
Thirdly, parties may frame issues belonging to the second dimension in terms of their core dimension; such a subsuming strategy is illustrated in Figure 3. Rovny and Edwards (2012: 70) argue that mainstream parties respond to new issue dimensions by forcing “newer and often disruptive issues into the dominant economic dimension”. The implication of a subsuming strategy, therefore, is that parties will try to frame issues in core-dimension terms (Basile, 2012). This indicates that subsuming is a rhetorical strategy, and can therefore only be identified by analysing political texts in a qualitative way. Thus state-wide parties may frame decentralisation in economic terms, thereby seeking to link it to the economic dimension of competition which is of primary importance to them. For example, the Spanish Partido Popular has recently advocated the re-centralisation of certain policy competencies in the interest of economic recovery and efficiency (Verge, 2013: 329). In contrast, ethno-regionalist parties may frame the economic dimension in terms of a national identity discourse, thereby seeking to link it to the territorial dimension of competition. A similar logic is suggested in Rabushka and Shepsle’s (1972) famous ethnic outbidding model of party competition in plural societies that assumes that ethnic parties will interpret all issues in communal/ethnic terms, once the ethnic dimension becomes salient. Consequently, they model competition between ethnic parties in a one-dimensional space. Another example of the subsuming strategy can be illustrated by the framing strategies used by parties to justify their position on European integration. Communist and socialist parties oppose European integration as they see labour and social security standards endangered by European integration. By contrast, populist and radical right parties justify their opposition to European integration in cultural terms using nationalistic frames (Helbling et al., 2010).
The subsuming strategy: Issues associated with the secondary dimension (Dsecond) are framed in terms of the core dimension (Dcore).
Fourthly, parties may decide to take clear, distinguishable positions on issues that belong to both dimensions (Alonso, 2012: 36). We refer to this strategy as the two-dimensional strategy, as shown in Figure 4. For mainstream parties, the implication is that they assume positions on both the economic and territorial dimensions simultaneously. There is ample evidence that mainstream parties have indeed been forced to take the latter dimension seriously, for example in order to stave off an electoral challenge from ethno-regionalist parties and/or as a response to decentralisation processes that have fostered territorially differentiated electoral preferences. Hopkin and Bradbury (2006), for example, document the attempts of British mainstream parties to adopt more distinctively Scottish and Welsh discourses in response to the creation of new devolved institutions in these territories in the late 1990s. Similar dynamics have been identified on a case study basis elsewhere in Western Europe (Fabre and Martínez-Herrera, 2009; Maddens and Libbrecht, 2009), and Toubeau and Wagner (2015) show in the first large-N study that the stances that state-wide parties adopt on decentralisation are systematically linked to their positioning on the economic and cultural dimensions. For ethno-regionalist parties, Massetti (2009) has shown that the vast majority of these actors in Western Europe combine positions on their core territorial dimension with centre, left, or right stances on the economic dimension. Such a multidimensional strategy is not confined to multinational settings. In German politics, for example, party competition is also two-dimensional, which is not due to the presence of a territorial dimension, 5 but is due to the fact that the economic and social value dimension cannot be reduced to a single left-right axis of party competition (Huber, 1989; Pappi, 1984). For example, whereas the FDP has a liberal, progressive position on the social value dimension as well as a liberal (i.e. rightist) position on economic issues (Franzmann, 2012), the CDU/CSU combines a centre right economic position with a value conservative position.
The two-dimensional strategy: Position on core (Dcore) and secondary dimensions (Dsecond) without framing one in terms of the other.
4. Contributions
The aim of this article was to conceptualise how parties deal with a second dimension in party competition. We have therefore taken institutional and other strategic incentives and constraints as ceteris paribus conditions. Of course parties do not choose these strategies in a vacuum but under a wide range of constraints that result inter alia from the institutional regime, public opinion, and the nature of strategic interaction with other parties competing within the same system. The empirical contributions to this Special Issue provide important insights into the determinants of parties’ strategic choices in each of the selected cases.
The four articles that follow operationalize the strategies outlined above and explore when and why parties use each of them in the two-dimensional competition spaces characteristic of multinational democracies. In doing so, the authors adopt methodologically innovative ways of measuring the different strategies. With regard to the first two tools, position and salience, systematic data collection for the territorial dimension has only recently begun and the contributors to this Special Issue are at the forefront of this development (Alonso, 2012; Alonso et al., 2015; Massetti, 2009; Massetti and Schakel, 2015; Szöcsik and Zuber, 2015). With regard to the third tool, parties’ framing of issues, and independent of whether these issues belong to the economic or the territorial dimension, only few scholars have thus far systematically measured this important component of parties’ strategies (Basile, 2013; Chaney, 2013, 2014; Helbling et al., 2010). This may be due to the fact that measuring issue framing requires labour-intensive, qualitative assessment of party rhetoric, as becomes evident from the systematic content analysis in the two contributions to the Special Issue that take a closer look at issue framing (Basile, 2015; Hamann and Field, 2015).
The authors also draw on a wide repertoire of sources to measure different strategies. These sources range from national (Basile, 2015) and regional (Alonso et al., 2015) electoral manifestos and bill proposals in parliament (Hamann and Field, 2015), to the systematic collection and coding of case studies (Massetti and Schakel, 2015). On the basis of these sources, the authors present descriptive inferences about the strategies mainstream and ethno-regionalist parties employ during national elections (Basile, 2015; Massetti and Schakel, 2015) and during regional elections (Alonso et al., 2015 and Massetti and Schakel, 2015), but also during the inter-electoral period of political competition when parties fulfil their representative tasks and engage in policy-making in parliament (Hamann and Field, 2015). They test whether mainstream and ethno-regionalist parties use different strategies across different regional electoral arenas within the same state as well as across states (Alonso et al., 2015; Massetti and Schakel, 2015) and across different national electoral arenas (Massetti and Schakel, 2015).
Taken together, the descriptive findings presented in this Special Issue challenge the assumption that ethno-regionalist parties behave as niche parties (Ezrow, 2008, Meguid, 2008). Alonso et al. (2015) show that competition spaces in those regions of Spain and the UK that host relevant regionalist parties are thoroughly two-dimensional and that regionalist parties use both of these dimensions in electoral contest, rather than focusing exclusively on territorial issues. Their explorative regression analyses also show that state-wide parties that face peripheral competitors are more likely to adopt the two-dimensional strategy than a one-dimensional strategy compared with state-wide parties that are not challenged by regional parties. Finally, if regionalist or state-wide parties opt for the blurring strategy in regional elections, then surprisingly both of them tend to blur their economic position. The challenge to the ‘nicheness’ of ethno-regionalist parties receives further support from Massetti and Schakel, who identify a clear position on the left-right dimension for all regionalist parties in Western Europe. They show that left-right and centre-periphery positioning are correlated, as leftist positions are associated with secessionist ones and rightist positions with autonomist ones. Turning to the explanations of parties’ strategic choices, they show that regionalist parties’ position on the left-right dimension is influenced by the relative socio-economic status of the region the regionalist party seeks to represent: regionalist parties in relatively poor regions tend to position themselves to the left, while regionalist parties in relatively rich regions tend to position themselves to the right.
Field and Hamman provide further ample evidence of ‘beyond niche’ behaviour in their study of a Basque and a Catalan party that have been governing their respective regions for most of Spain’s democratic history. During this period, these parties have proposed a wide range of bills in the Spanish parliament that are not related to their core centre-periphery business. Their findings suggest that regionally governing parties have to deal with the full spectrum of relevant political issues and cannot afford to focus exclusively on issues associated with the centre-periphery dimension. For those centre-periphery related bills they do propose, the parties more often than not choose to employ non-territorial frames for justifying the bill proposal. In doing so, the two parties use a much wider range of frames than merely those associated with an economic and a centre-periphery dimension. This shows that, empirically, issue framing is a flexible tool for parties to link issues in new ways that exceed the options theorised in this introductory article.
The Special Issue also sheds further light on mainstream party strategies in settings where there is a salient territorial dimension. As stated above, the contribution by Alonso et al. (2015) provides evidence of the two-dimensional behaviour of such parties in Spain and the UK. Furthermore, Basile (2015) demonstrates that Italian mainstream parties went beyond the predictions made by Meguid’s (2008) PSO theory when responding to the challenge mounted by the Lega Nord. Rather than changing their selective emphasis of, and their position on, decentralisation they blurred their position on territorial issues or modified their rhetoric, making use of the subsuming strategy. This strategic response would have remained obscured in both a classical spatial and a classical saliency analysis and only comes to light as a result of Basile’s analysis of parties’ use of framing alongside positioning and selective emphasis.
The Conclusion to the Special Issue by Rovny (2015) summarises the main empirical findings and theoretical contributions. Taken together, the articles demonstrate that both mainstream and ethno-regionalist parties frequently deal actively with their secondary dimensions, rather than selectively emphasising only their core dimension. They make use of the blurring, subsuming and two-dimensional strategies introduced in this Introduction. Mainstream and ethno-regionalist parties across Western Europe thus employ a wider range of strategies than previously acknowledged in the scholarly literature and rarely resort to a uni-dimensional strategy. The explorative insights of the articles in this collection suggest that these choices are determined by the competitive configuration of the party system, parties’ participation in government, and the socio-economic status of the region in case of ethno-regionalist parties. The Special Issue suggests these hypotheses for future research that could test them on the basis of more cases and using several measurements for party strategies at both the regional and the state level and in both the electoral and the inter-electoral periods of competition. But as Rovny notes, the empirical analyses also point to the structural limits to political strategizing; the Special Issue thus points to the complementarity of strategic vs structural approaches to studying party competition.
Future research should apply our concepts to, and test the emerging hypotheses in, cases where dimensions other than the economic and the territorial one are primary and secondary to parties. By conceptualising the strategic tool kit available to political parties, and arguing and showing empirically that parties’ strategic repertoire for dealing with a second dimension is broader than previously acknowledged by the literature, this Special Issue thus lays the groundwork for a comprehensive theory of party competition in two-dimensional space.
Footnotes
Acknowledgements
The idea for this Special Issue was originally developed at a workshop entitled “Ethnicity, Territory and Party Competition. Toward a Unified Approach” directed by Edina Szöcsik and Christina Zuber at the ECPR Joint Sessions of Workshops in Antwerp, 10–15 April 2012. Previous versions of this article were presented at the Department of International Politics, University of Aberystwyth, 12–13 December 2013 and at a workshop on “The Politics of Identity Adoption and Change”, IBEI, Barcelona, 13 October 2014. We would like to thank Sonia Alonso, Francesc Amat, Linda Basile, Daniel Bochsler, Kanchan Chandra, Bonnie Field, Aina Gallego, Simon Franzmann, Maria José Hierro, Filippo Tronconi, Holger Reinermann, Jan Rovny, Arjan Schakel, Markus Wagner, and two anonymous referees for their excellent comments.
Funding
The work undertaken by Anwen Elias for this Special Issue was funded by the Nuffield Foundation (Ref: NCF/36360). Christina Zuber did part of the work for this Special Issue while being funded by the Fritz Thyssen Foundation.