The second decade of the third millennium was marked by growing support for so-called populist parties (PPs) in Western democracies. As the embodiment of people-centrism and anti-elitism, populism is often described as an imminent threat to the foundations of liberal democracy (Mounk Reference Mounk2018). With scholars debating whether it is a permanent phenomenon, on the cusp of the third decade of the 2000s the extremes in some European democracies already seem to be contained. Italy, however, is still one of the countries with the highest support for PPs, such as the 5 Star Movement (Movimento 5 Stelle – M5S). Founded in 2009 by former comedian Beppe Grillo and web strategist Gianroberto Casaleggio, the movement increased its popularity until it became the largest party in the 2018 national elections. The fact that the M5S continues to play a crucial role in Italian politics makes Italy's case particularly relevant.
The M5S fits well into the definition of a populist challenger party. These parties challenge the status quo of a political arena dominated by the mainstream parties by introducing ‘programmatic and ideological thinking that is less consistent with the traditional left–right dimension’ (Müller-Rommel Reference Müller-Rommel1998: 193). In terms of Catherine De Vries and Sara Hobolt's (Reference De Vries and Hobolt2020: 17) definition of challenger parties as ‘parties without government experience’, the M5S is a classic example: (1) it does not come from the traditional party families; (2) it focuses on issues such as the environment and the EU that do not overlap with the classic left–right dimension; and (3) the M5S is classified as a PP by the Populism and Political Parties (POPPA) dataset (Meijers and Zaslove Reference Meijers and Zaslove2020), where it ranks first among Italian parties in the populism index (based on the categories Manichaeism, indivisibility of the people, people-centrality, general will, and anti-elitism). Recognition of the populist character of the M5S is shared by many international classifications (e.g. Rooduijn et al. Reference Rooduijn2019).
PPs' growing presence in contemporary European democracies poses questions about what fuels this ‘populist explosion’ (Judis Reference Judis2016). The literature highlights cultural (Norris and Inglehart Reference Norris and Inglehart2019) and economic (Kriesi and Pappas Reference Kriesi and Pappas2015; Kriesi et al. Reference Kriesi and Kriesi2008) explanations and, more recently, the role of political communication and how PPs fuel affective polarization (Aalberg et al. Reference Aalberg2017; Blassnig et al. Reference Blassnig, Davis and Taras2019; Harteveld et al. Reference Harteveld, Mendoza and Rodujin2021). However, understanding the electoral success of PPs requires further focus on how their entry alters the policy space of party competition and their interactions with mainstream parties. William Riker's (Reference Riker1986) principle of dominance suggests that new parties may stress issues on which they take a divergent position from other parties but have significant electoral support. Bonnie Meguid (Reference Meguid2005) argues that ‘niche’ parties are successful if they have ownership of salient issues often downplayed by mainstream parties. We follow this argument by focusing on the salience of issues in the Italian policy space when the M5S entered the political arena in 2013. We then examine how the entry of the M5S affected changes in existing parties' positions and reshaped the structure of the policy space by focusing on three elections: 2008, 2013 and 2018.
Our central argument is that the entry of new political parties upsets existing equilibria, inducing a realignment of existing parties. We assume policy spaces in contemporary multiparty democracies are multidimensional and change over time (Benoit and Laver Reference Benoit and Laver2006). Contrary to the standard unidimensional model of party competition, multidimensional political environments are characterized by a weaker centripetal draw, allowing new parties to adopt radical positions to gain votes rather than locate themselves at the centre of voters' distributions (Schofield and Sened Reference Schofield and Sened2006).
We use consistent expert survey data collected employing a format developed by Kenneth Benoit and Michael Laver (Reference Benoit and Laver2006) for each national election between 2008 and 2018 to build three-dimensional maps of the policy space in Italy. We complement two-dimensional models of the policy space by adding a third policy dimension. We then apply the solution concept of the uncovered set (UCS) to examine the extent to which new entrants bring about political change (Bianco et al. Reference Bianco, Jeliazkov and Sened2004, Reference Bianco, Heckelman and Miller2015; Jeong et al. Reference Jeong2011; Miller Reference Miller1980). The UCS can be used as a measure of legislative preferences and outcomes (Bianco et al. Reference Bianco, Jeliazkov and Sened2004), allowing us to bypass much of the complexity we could have encountered if we tried to measure the effect of those new entrants by following actual legislation. We find that no new dimension was introduced, but the M5S cleverly manipulated existing dimensions. Our analysis shows the extent to which the M5S's entry affected existing parties' positions and the set of feasible outcomes in the Italian parliament.
Theoretical background
The spatial approach to the study of elections and parliamentary politics assumes that political parties in liberal democracies compete over policy issues and voters select the candidate/party whose policy positions are closest to their own (Downs Reference Downs1957). Thus, a spatial analysis of party competition requires an empirical assessment of party positions and their movement in the policy space (Benoit and Laver Reference Benoit and Laver2006). A single left–right dimension is often used to map the positions and movements of political parties in Western democracies (Castles and Mair Reference Castles and Mair1984). However, a unidimensional account of policy spaces has proven theoretically inadequate to capture the complexity of multiparty democracies (Laver and Shepsle Reference Laver and Shepsle1996; Schofield Reference Schofield, Barnett, Hinich and Schofield1993). Moreover, empirical research has shown that the link between parties' policy positions and voters' perceptions of parties' positions on the left–right scale has weakened since the 1970s in Western Europe (Hooghe et al. Reference Hooghe, Marks and Wilson2002), conceding grounds for a new and independent cultural divide pitting postmaterialist cosmopolitan libertarians against conservative nationalists (Kriesi et al. Reference Kriesi and Kriesi2008). This prompted a debate on how many dimensions effectively structure party competition in Europe and what they contain (Bakker et al. Reference Bakker, Jolly and Polk2012; Bornschier Reference Bornschier2010; De Vries and Marks Reference De Vries and Marks2012). Sara Hobolt and James Tilley (Reference Hobolt and Tilley2016) and Liesbet Hooghe and Gary Marks (Reference Hooghe and Marks2018) provide evidence that a new dimension of conflict, centred on preferences on nationalism, immigration and European integration, has reshaped party politics in Western European countries since 2008.
Multidimensional spatial models enable us to assess how many dimensions structure the policy space at a given time. Moreover, they allow us to see how dimensions change over time and how the policy positions of political actors evolve (Laver and Benoit Reference Laver and Benoit2007). Changes may arise from exogenous events such as policy and agenda shocks (Laver and Shepsle Reference Laver and Shepsle1998; Lupia and Strøm Reference Lupia and Strøm1995) or endogenously due to party competition. Changes may affect the salience of dimensions or shifts in party positions or both. The salience of issue dimensions may change as new actors enter the electoral arena (De Vries and Hobolt Reference De Vries and Hobolt2020; Meguid Reference Meguid2005). Moreover, the dimensions can be strategically manipulated by political actors to influence political outcomes (Riker Reference Riker1986; Stokes Reference Stokes1963).
The spatial approach has examined political entry as a strategic choice, showing that it has a considerable impact on politics not only because it restructures political spaces but also because it upsets the balance of power among existing parties and increases party-system volatility. In unidimensional policy spaces, parties seeking to maximize their electoral appeal are expected to enter at the median voter position (Feddersen et al. Reference Feddersen, Sened and Wright1990). In response, existing parties may choose positions that diverge from the median voter's ideal point to hinder the electoral success of new entrants (Greenberg and Shepsle Reference Greenberg and Shepsle1987; Palfrey Reference Palfrey1984). Effects of entry are more complex to detect in multiparty systems when modelled as inherently multidimensional. In such settings, party positions can, in equilibrium, skew off the centre of the political sphere and be far away from the median on any dimension. Moreover, political entry affects policy outcomes by moving them closer to the entrant's ideal point (Schofield and Sened Reference Schofield and Sened2006). This presents new parties with policy benefits even when chances of electoral success are modest (Kselman et al. Reference Kselman, Powel and Tucker2016).
Research on the entry of challenger parties has explored when and how new issue dimensions are successfully introduced (Bélanger and Meguid Reference Bélanger and Meguid2008; Carmines and Stimson Reference Carmines and Stimson1989; De Vries and Hobolt Reference De Vries and Hobolt2012, Reference De Vries and Hobolt2020). De Vries and Hobolt (Reference De Vries and Hobolt2020: 98) stress that while mainstream parties' strategic appeal lies within the boundaries of the left–right dimension, divisive ‘issues are not easily integrated into a left–right party brand’. This allows challenger parties to become issue entrepreneurs – that is, to mobilize the electorate around these overlooked issues. They show that both immigration and European integration were selected as ‘highly appropriable’ issues by a few challenger parties. Moreover, empirical findings support the hypothesis that challenger parties improve their electoral performance by adopting extreme or radical positions on those issues (Adams et al. Reference Adams2006; Ezrow et al. Reference Ezrow2010). Our analysis will show how the austerity measures imposed by the EU in the wake of the eurozone crisis brought the EU to the fore and became the M5S's ticket to the Italian parliament in 2013.
The interplay among mainstream and challenger parties can also be examined from the perspective of issue yield theory, according to which parties choose issues carefully so as not to divide their supporters and, therefore, lose votes (De Sio and Weber Reference De Sio and Weber2014). Consequently, mainstream parties with an established electorate would refrain from addressing divisive issues but focus their campaign efforts on issues most voters agree with. This ‘strategy of issue avoidance’ is exactly what benefits the challengers (i.e. new parties without a well-formed core electorate), who can ‘afford’ to introduce or capitalize on issues that drive a wedge between and within the dominant parties who strategically avoided these issues. In other words, the issue yield model of party competition seems to explain why mainstream parties refrain from addressing specific salient issues. However, the challenger parties' strategy is best explained through issue salience, issue entrepreneurship and using wedge issues rather than betting on the safe, unanimously agreed-on issues.
Few empirical studies have considered multiple dimensions of political conflict when analysing the positional strategies of political parties (Krause Reference Krause2019). Indeed, moving from unidimensional to the multidimensional spatial representation of the political space poses a challenge. Multidimensional models rarely converge on a clear expected outcome, as in the case of the unidimensional median voter, which almost always persists (Downs Reference Downs1957). With the final outcomes ‘wandering all over the place’, leading scholars described the prospects of multidimensional political environments as ‘chaotic’ (McKelvey Reference McKelvey1986; McKelvey and Schofield Reference McKelvey and Schofield1986). To tackle these problems, a solution concept, the uncovered set (Miller Reference Miller1980), emerged as a powerful predictive tool that describes the set of possible outcomes in multidimensional environments (McKelvey Reference McKelvey1986).
The uncovered set (UCS) is a solution concept that predicts electoral and legislative outcomes in majority-rule multidimensional settings lacking clear and stable equilibria (Miller Reference Miller1980). Alternative x beats y if more voters prefer x to y. Alternative x covers y if x beats y, and every alternative beaten by y. The UCS is a set of alternatives not covered by any other alternative. Therefore, strategically, politicians should bet on uncovered points rather than covered ones that are bound to lose to the points that cover them (see Bianco et al. Reference Bianco, Jeliazkov and Sened2004 for a more comprehensive explanation). Following this logic, analytical and empirical studies establish that electoral and legislative outcomes are more than likely to fall within the UCS (Bianco et al. Reference Bianco, Jeliazkov and Sened2004, Reference Bianco, Heckelman and Miller2015; Cox Reference Cox1987; Jeong et al. Reference Jeong, Miller and Sened2009, Reference Jeong2011; Kam et al. Reference Kam2009). These possible outcomes can be predicted by assessing the location and size of the UCS.
In stable political environments characterized by a fixed number of parties occupying stable positions on a fixed number of dimensions, the UCS will show no change in its location or size. Few political systems in contemporary democracies, however, can boast of such stability. On the contrary, as politics in multiparty democracies is becoming inherently dynamic and multidimensional, the UCS can be expected to change its location and size. Since these characteristics of the UCS are determined by party positions, the entry of a new party can, in theory, shift the set of feasible outcomes towards the newcomer and/or enlarge the size of the UCS. A small UCS means that the set of feasible outcomes is small and clear to all political actors involved and that the distances between the epicentre of the UCS and parties' positions are small and relatively easy to compromise over, ensuring a smooth policymaking process. A large UCS, on the other hand, would create considerably larger distances and require greater compromise, fostering turbulence and uncertainty about future policy.
In this article, we focus on the entry of the M5S. We seek to answer two questions: what accounts for the M5S's electoral success? and how did the M5S's entry affect the Italian policy space? We hypothesize that:
Hypothesis 1: The M5S used a centrifugal electoral strategy on a salient dimension.
Hypothesis 2: The M5S's entry upset the existing equilibrium, causing party realignment in its favour.
To substantiate these hypotheses, we rely on a three-dimensional representation of the policy space of party competition in Italy. Multidimensional models of party competition typically provide a two-dimensional account of the policy space, as solution concepts of spatial games of coalition formation were typically explored in two-dimensional spaces (Laver and Shepsle Reference Laver and Shepsle1998; Schofield Reference Schofield, Barnett, Hinich and Schofield1993). However, the two-dimensional representation of the political space is not the most adequate when trying to capture the changes induced by the entry of new parties. Moreover, we assume that the dimensionality of the policy space is an empirical matter (Benoit and Laver Reference Benoit and Laver2006) and follow Daniela Giannetti et al. (Reference Giannetti, Pedrazzani and Pinto2017), who reported the growing salience of the EU and immigration dimensions in Italian politics. Consequently, our approach complements traditional spatial models, adding a third relevant dimension to better fit volatile voter preferences, changing party positions and policy dimensions that are rarely static.
Method and data
To estimate party positions, scholars rely on different techniques (Budge et al. Reference Budge2001; Laver Reference Laver2001). Among those, expert surveys are a relatively quick and costless way of collecting data. Moreover, this technique allows researchers to assess confidence in the accuracy of estimates (Benoit and Laver Reference Benoit and Laver2007). Our study uses expert survey data collected via the format proposed by Benoit and Laver (Reference Benoit and Laver2006). It adopts an a priori approach:Footnote 1 policy dimensions/scales are predefined, and parties are positioned on these scales by country experts. The scores used to estimate party positions are the aggregated results of expert judgements. Benoit and Laver's expert survey has a fixed set of dimensions (see Online Appendix 1 for the exact phrasing): taxes versus spending, deregulation, social policy, immigration, environment, and decentralization. It also includes three policy scales on specific aspects of European politics: EU authority, EU security, and EU accountability. For each of these nine policy dimensions, experts were also asked to locate each party on a scale measuring the salience of the dimension for that party. This scale ranges from ‘1’ (not important at all) to ‘20’ (very important).
Expert survey data are often collected at one time point, making it difficult to use them for the analysis of political change. This limitation does not apply to our study, as the Benoit–Laver format has been consistently applied by a team of researchers (Di Virgilio et al. Reference Di Virgilio2015; Giannetti et al. Reference Giannetti, Pedrazzani and Pinto2017, Reference Giannetti, Pedrazzani and Pinto2021) in subsequent waves of the expert survey administered to Italian experts in a series of elections in Italy (2001–2018). The number of valid responses remained high in all surveys, never falling below 40. Moreover, our data allow us to conduct validity and replicability tests on the results.Footnote 2 We compare our estimates on a selected number of dimensions with those of the Chapel Hill Expert Survey (CHES) and the Comparative Manifestos Project (MARPOR/CMP). The results, summarized in Online Appendix 2, indicate a high degree of correlation.
We complement our primary dataset by using the Eurobarometer data about main public concerns in the period under consideration. Although coming from a different source, these data are essential to provide information on issue salience in the eyes of the electorate.
We build three-dimensional maps of the policy space by identifying the relevant dimensions based on their salience as derived from our data. Three-dimensional maps require us to apply the notion of UCS. Parties' positions and weights determine the location, size and shape of the UCS. Therefore, it is appropriate to estimate new entrants' effect on a policy space by assessing their impact on the location, size and shape of the UCS in a spatial context. Thus, the UCS is a way of depicting party realignment following the entry of the M5S. We apply an algorithm used to estimate the UCS for any dataset to gauge the effect of the M5S's entry on the Italian policy space.Footnote 3
After calculating the UCS for each election, we use a simple quantitative method to measure the change in the size and location of the UCS. The difference in size is calculated by computing the size of the UCS relative to the total three-dimensional space of a 1.003 cube. The change in the location of the UCS is measured by calculating the centre of the distribution of the results of each election and then measuring the distance between the centres of the distribution of the outcomes of each of the three elections. Thus, we get a clear indication of where the UCS has moved in the relevant three-dimensional space and how far it has moved in each direction between the three successive elections.
Italy
Our study offers theoretical and measurement tools to analyse how entry alters the space of party competition, using Italy as a case study. Italy is particularly relevant as the rise of the M5S had far-reaching consequences for the entire political system. Following the 2018 elections, the party was able to form a ‘fully fledged populist government’, often seen as a potential threat to the EU.Footnote 4
We start our analysis with 2008, as these elections exemplify the configuration of the policy space until then and the predominant pattern of party competition among centre–right and centre–left pre-electoral coalitions. Between the 2008 and 2013 elections, the M5S entered the political arena. Its platform was based on environmentalism, identifying the main issues or the so-called five stars as water, environment, mobility, development and energy (Passarelli et al. Reference Passarelli, Tronconi, Tuorto, Corbetta and Gualmini2013). In the 2013 elections, the M5S became the second-largest party in the Chamber of Deputies, with 25.6% of the vote and 109 seats. In the 2018 elections, the M5S became the largest party in the chamber, with 227 seats (Table 1).Footnote 5
Results of the Italian General Elections in 2008, 2013 and 2018 (Chamber of Deputies)
Source: Ministry of Interior (https://elezionistorico.interno.gov.it/).
Notes: +EU (Più Europa – More Europe); FDI (Fratelli d'Italia – Brothers of Italy); FI (Forza Italia – Go Italy); IDV (Lista di Pietro–Italia del Valori – List Di Pietro–Italy of Values); LeU (Liberi e Uguali – Free and Equal); LN (Lega Nord – Northern League);Footnote 10 M5S (Movimento 5 Stelle – 5 Star Movement); MpA (Movimento per l'Autonomia – Movement for Autonomy); NcI (Noi con l'Italia – Us with Italy); PD (Partito Democratico – Democratic Party); PDL (Il Popolo della Libertà – People of Freedom); SC (Scelta Civica – Civic Choice); SEL (Sinistra Ecologia Libertà – Left Ecology Freedom); UDC (Unione di Centro – Union of the Centre).
The success of the M5S is often explained by its populist rhetoric, which fuels voters' distrust of established political parties (Albertazzi and McDonnell Reference Albertazzi and McDonnell2015; Bobba and Legnante Reference Bobba, Legnante and Aalberg2016). According to the POPPA dataset, the M5S ranks first among Italian political parties in terms of populism score.Footnote 6 However, to explain how the M5S garnered voters' endorsement in 2013 and 2018, our analysis focuses on policy issues (Pirro Reference Pirro2018; Tronconi and Mosca Reference Tronconi and Mosca2019).
Before focusing on the supply side, we examine the main public concerns in the period under consideration. Our theoretical premise suggests that challenger parties exploit unattended public concerns for issue entrepreneurship. We use the Eurobarometer data as a complementary source of information, as our dataset lacks data on public opinion. According to Figure 1, the two biggest public concerns throughout 2008–2018 were unemployment and the economic situation. Taxation gained salience before the 2013 election but declined by 2018. Instead, following the 2015 European ‘refugee crisis’, concerns about immigration scored high in 2016 and remained the second-largest public concern after unemployment in 2018.
Top Public Concerns in Italy, 2008–2018
Source: Eurobarometer Interactive (2008–2018).
Note: The figure includes survey data collected at the closest time point to each election.
Eurobarometer wording: ‘What do you think are the two most important issues facing [country] at the moment? Answers: Crime; Defence/foreign affairs; Economic situation; The education system; Energy; Energy-related issues; The environment, climate and energy issues; Government debt; Health and social security; Healthcare system; Housing; Immigration; [Country]'s external influence; Pensions; Protecting the environment; Public transport; Rising prices/inflation/cost of living; The state of Member States’ public finances; Taxation; Terrorism; Unemployment; Other (spontaneous); None; DK – don't know; Other (specify).’
The Eurobarometer has a separate question on political trust, which we use as a proxy to capture attitudes towards the EU. Political trust is crucial for understanding the nature of PPs, as their rhetoric often features anti-establishment sentiments (Mudde and Rovira Kaltwasser Reference Mudde and Rovira Kaltwasser2017). Voters concerned with economic crises and social change often have less trust in political institutions if they perceive these issues as insufficiently attended to by the government (McLaren Reference McLaren2012), creating fertile ground for challenger parties. In Italy, levels of political trust are initially extremely low (Figure 2). Therefore, it is highly plausible that anti-establishment rhetoric would resonate with the electorate.
Net Public Trust in Political Institutions, 2008–2018
Source: Eurobarometer Interactive (2008–2018).
Note: Net trust was calculated by subtracting the percentage of mistrust from that of trust.
The Eurobarometer wording: ‘I would like to ask you a question about how much trust you have in certain institutions. For each of the following institutions, please tell me if you tend to trust or tend not to trust it: the [nationality] parliament; the [nationality] government; the European Union.’
Figure 2 shows that net public trust in the EU plummeted below zero in 2011 and remained negative despite a subsequent fluctuation. The reason for this was the newly introduced austerity measures by the EU, which created much tension between the national and supra-national levels (Di Virgilio et al. Reference Di Virgilio2015). As we will show in the next section, the political climate during the 2013 election campaign created a window of opportunity for anti-EU parties, which the M5S eagerly seized.
Results
This section analyses the structure of the Italian policy space between 2008 and 2018. Table 2 illustrates the extent to which the parties attributed importance to the most salient issues according to the experts: EU authority, which corresponds with a sharp decline in public trust in the EU (Figure 2), immigration, and taxes versus spending, the last matching the public concern with the economy (Figure 1). Calculating the relative salience of each issue against the backdrop of the average salience of all issues in each of the three elections allows for comparison over time. The relative salience of EU authority and immigration outweighs that of the economic dimension, which corroborates previous findings that indicate the growing importance of the EU and immigration dimensions in Italian politics (Giannetti et al. Reference Giannetti, Pedrazzani and Pinto2017). Interestingly, the environment – one of the main pillars of M5S ideology – is a low-salience policy domain. The party's entry into the Italian policy space failed to increase the relative salience of this issue in both 2013 and 2018 due to a lack of public concern with it (Figure 1).
Relative Issue Salience in Italy, 2008–2018
Note: Relative salience was calculated as a ratio between the salience of each issue and the average salience of all issues in each election. The most salient issues are indicated in bold.
We construct a three-dimensional picture of the Italian policy space based on the most salient issues in the eyes of voters and parties: the EU, immigration and taxes versus spending.Footnote 7 Party positions in the policy space are presented in a three-dimensional chart (Figure 3). Values of the dimensions run from positive at the bottom left side, through 0 (neutral) at the intersection to negative at the top right side of the cube. Figure 3 illustrates how polarized the Italian policy space is. The left-leaning bloc in the lower-left part of the cubes includes the Democratic Party (Partito Democratico – PD) and Left Ecology Freedom (Sinistra Ecologia Libertà – SEL), later joined by the Civic Choice (Scelta Civica – SC), Free and Equal (Liberi e Uguali – LeU) and the pro-European More Europe (Più Europa – +EU). These parties favour immigrant integration, endorse the EU and support taxation. The upper-right corner hosts right-leaning parties: The People of Freedom (Il Popolo della Libertà – PDL), Brothers of Italy (Fratelli d'Italia – FDI), Go Italy (Forza Italia – FI) and the Northern League (Lega Nord – LN).Footnote 8 These parties oppose immigration, advocate reduction of taxes and are often Eurosceptic. The polarization is particularly evident in the immigration dimension, with parties such as the SEL in 2013 and LeU in 2018 taking a very pro-immigration stance and LN an explicitly anti-immigration stance, as well as in the EU dimension, with the M5S and SC taking extreme anti- and pro-EU positions, respectively, in 2013.
Party Positions on the Most Salient Policy Dimensions in 2008, 2013 and 2018
Notes: The lower-left side of the cubes represents a positive, the upper-right side a negative, and the intersection a neutral position. Thus, the parties in the lower-left quadrant are pro-EU and pro-immigration, while the parties in the upper-right corner are anti-EU and anti-immigration. The UCS positions are marked with an ‘x’-like shape accordingly for each election.
To illustrate changes in the distances between parties, parties' positions are connected by a dotted line in 2008, a dashed line in 2013 and a solid line in 2018 (Figure 3). Moreover, the positions of the UCS for each election are indicated. Parties' positions and the number of seats they are allocated determine the position of the UCS. The UCS is often positioned next to the party with the largest share of seats in each election. The largest party tends to form the winning coalition, not only because of its size but also because of its proximity to the UCS.Footnote 9 The dotted shape connecting the parties' positions in 2008 is significantly smaller than the dashed shape connecting the parties' positions in 2013, indicating smaller distances between parties and a more consolidated decision-making process. Indeed in 2008, the policy space was dominated by the largest centre-right PDL – the winner of the elections – and the centre-left PD. A few smaller parties were modestly represented in the chamber. The enlarged dashed shape in 2013 can be explained by the entry of new parties: the SEL, SC and M5S. The new entrants held radical positions on the dimensions of immigration and the EU, increasing the scope for political action. The entry of the left-leaning SEL and the pro-European SC, as well as the electoral victory of the PD, contributed to the UCS moving from the right-leaning bloc in 2008 (upper-right corner in the upper cube) to the left-leaning bloc (lower-left corner). Public support for both the PD and PDL decreased in 2013 compared to 2008, but the PDL only came third, while the right-wing FDI and LN did not have enough power to offset the electoral success of the left-leaning bloc (Table 1). However, there was no enlargement of the UCS in 2013, suggesting that the change did not bring much uncertainty into the Italian policy space.
As an independent contestant in the 2013 election, the M5S placed itself on the middle ground on the dimensions of immigration and taxes versus spending. However, as an anti-EU party, its position on the EU dimension is extreme and comparable only to the LN. The analysis of distances between the M5S's position vis-à-vis other contestants shows that the party entered the Italian policy space holding a radical position on the issue of EU authority. This finding supports our first hypothesis that the M5S used a centrifugal entry strategy on a salient dimension. Moreover, the M5S pursued a centrifugal strategy on the environmental dimension, where it was surpassed only by the left ecological SEL (Table 3). However, the issue was not salient in the eyes of voters (Figure 1) or other parties (Table 2). Therefore, despite being exemplified by the five ‘stars’ in the party's name, the M5S’s radical position on this dimension cannot explain its triumphant entry into Italian politics.
Parties' Positions on the Environment, 2013–2018
The upper cube in Figure 3 shows why the M5S, although the second-largest party, refused to join the PD-led coalition: It was too far from the UCS. In 2013, the M5S had little chance of affecting policymaking in Italy. Thus, an important strategic move of the M5S between the 2013 and 2018 campaigns was a slight moderation of its position on the EU dimension and a notable shift in a negative direction on the immigration dimension (Table 4 and Figure 3). With the M5S taking advantage of the highly polarized and volatile political space in Italy, the pendulum-like movement of the UCS from the right-wing bloc in 2008 to the left in 2013 continued in 2018, shifting the UCS back to the right and upwards. However, there was no change in the size or shape of the UCS despite the dramatic shifts in its position. Small uncovered sets indicate less uncertainty regarding the set of feasible outcomes, meaning that the winning coalition is constrained to a specific policy by the structure of parliament determined by the size and positions of all the parties that passed the entry threshold. The strategic move of the M5S allowed it to move the feasible policy outcome of the Italian parliament back to the right and upwards (lower cube in Figure 3) towards the M5S’s anti-EU position. The party's proximity to the UCS helps explain its pivotal role in the subsequent coalition formation.
Shifts in Parties' Positions on the Most Salient Issues, 2013–2018
Notes: Party positions were rescaled to fit the cube's dimensions (−1, +1). The table includes only parties that entered parliament in both elections.*PDL in 2013, FI in 2018.
Changes in party positions and their weights shifted the set of feasible policies to a more negative stance on immigration and the EU in 2018. Along with the strategic shift in its position, the M5S found itself at the very centre. Several factors may have contributed to this. First, following the 2015 ‘refugee crisis’, immigration became a more salient issue for the electorate, calling for parties to re-evaluate their positions in the 2018 election. Concern with the EU-imposed immigration and asylum policies brought the question of EU authority to salience. The FDI and M5S moved their position closer to the anti-immigration Eurosceptic League, and even the left-leaning PD favoured immigration less (Table 4). Second, voters' concern with the EU's open-door immigration policy affected their voting preferences: the four parties with the largest vote shares were the M5S, LN, PD and FI, with only the PD belonging to the left-leaning bloc. In 2018, the League became the second-largest party in the Chamber of Deputies, gaining 17% of the popular vote. This success cannot be attributed only to the increased salience of immigration – an issue the party has traditionally dominated. Following a change in leadership, the party under Salvini appealed to broader constituencies, as manifested in its more inclusive name, ‘the League’ rather than the Northern League, which contributed to the party's electoral success (Passarelli and Tuorto Reference Passarelli and Tuorto2018). Thus, the UCS in 2018 moved to the right not only because of the M5S's strategic manoeuvres but also due to the electoral success of the M5S, the League and FI, while the left-wing bloc, comprising PD and smaller parties, did not have enough weight to counterbalance it. However, the UCS's movement upwards towards the M5S's position on the EU dimension, rather than just rightwards towards the League, suggests that the 2018 party realignment was more related to the M5S than to the League's rebranding since the latter hardly changed its position in the three-dimensional space.
Thus, our second hypothesis that the entry of the M5S would upset the equilibrium and move the set of feasible outcomes in its favour received significant support, but our conclusion is cautious. While the UCS did move towards the M5S on the EU dimension in 2018, the magnitude of this movement can be ascribed not solely to the M5S's strategy but also to other changes intrinsic to the Italian sphere of politics in that election.
Conclusion
Our study aimed at expanding the scholarly understanding of challenger parties' entry into contemporary multiparty democracies by focusing on the case of the M5S in Italy. Its electoral success in 2013, amplified in 2018 when the party formed a government, presents a clear point of entry and allows for assessment of its effect on the Italian political space. This research focused on two main questions. The first examined the factors that facilitated the M5S's electoral triumph. To this end, we combined data on salient public concerns with data on the dimensionality of the policy space based on salience at the party level. Our empirical analysis distilled three main dimensions of importance: economy – proxied by taxes versus spending in our data – immigration, and the EU. These dimensions corroborate previous findings reporting the importance of the EU and immigration dimensions in addition to the traditional economic dimension. Our multidimensional account shows that – consistent with our expectations – in 2013 the party adopted a centrifugal strategy on one of the most salient dimensions of the policy space – the pro-/anti-EU dimension. The issue of the environment, although prominent in the M5S platform, could not be exploited as a successful entry strategy as it received little attention from the electorate and other parties. These findings contribute to the literature on party competition by showing how the multidimensionality of political space affects parties' entry strategies in terms of both salience and position. Our results provide evidence to those theoretical accounts that stress mainstream parties' avoidance of divisive issues and strategic exploitation of those unattended issues by challenger parties. Moreover, they confirm that challenger parties adopt radical positions to gain electoral ground.
The second question was whether the M5S's entry made any difference in Italian politics. To answer it, we combined our data on party positions with the calculation of the UCS based on these estimates. Not all theoretical models are created equal, and therefore, not all of them can be tested in the same way (Lenine Reference Lenine2018). Estimating the effect of new entrants on multidimensional multiparty policy spaces is a complex undertaking that requires following existing legislation. One of the primary contributions of the analytical framework presented in this study is that it allows us to overcome this complexity by focusing on changes in the location of the UCS.
Our visual three-dimensional representation of the Italian policy space showed that it became more polarized in 2018, with parties shifting their positions both rightwards and towards a more anti-EU stance. Our primary focus was on the EU dimension, which the M5S used strategically to enter Italian politics. The movement of the UCS in this dimension towards the Eurosceptic M5S and the corresponding shifts in the positions of the right-wing FDI and the left-wing PD seem to confirm our second hypothesis that the M5S caused party realignment in its favour. However, we assert this conclusion with some caution, as the reshaping of the policy space could also have been caused by parties adjusting their positions in the wake of the European ‘refugee crisis’ and dissatisfaction with the EU's asylum policy. Overall, our analysis offers tools to assess the volatility and vulnerability of policy spaces in modern liberal democracies in the face of challenger parties' strategic manoeuvres.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/gov.2022.38.
Acknowledgements
We thank Andrea Pedrazzani and Luca Pinto for sharing the dataset on the positions of the Italian parties. We thank the anonymous referees for their insightful comments and suggestions that helped improve the original manuscript.
Open access
