From Party Brands to Party Names

Just as brands differentiate products in market competition, they also serve to distinguish political parties in electoral competition. Brands operate as a shortcut, formed in interaction between firms and consumers (Kornberger Reference Kornberger2010). Branding results from strategic investment by leadership but is not solely top-down; it is also shaped through the experiences and perceptions of consumers (Clifton Reference Clifton2009). In political contexts, brands help parties to communicate both identity and competence (for an overview, see Ishiyama and Rybalko Reference Ishiyama and Rybalko2023). According to Scammell (Reference Scammell2015), brands in party politics are multidimensional constructs that combine substantive policy positions and performance (boundary conditions) with emotional, cultural, and symbolic appeals (brand differentiators) that arise in interaction between party elites and voters.

However, the concept of party brands is often discussed metaphorically, without a clear empirical indicator. Kim and Solt (Reference Kim and Solt2017) suggest party names as an indicator of branding. They examine the extent of party name changes and demonstrate that renaming is not uncommon. Across European democracies, roughly a third of all parties have renamed themselves at least once since 1945. As stated, parties with a shorter history are more likely to undergo renaming, which reflects lower investment in an existing brand and reaction to electoral setbacks, with significant contextual differences between more institutionalized and weaker party systems (see also Avina and Spoon Reference Avina and Spoon2025; Casiraghi, Cusumano, and Chryssogelos Reference Casiraghi, Cusumano and Chryssogelos2024; Ibenskas and Sikk Reference Ibenskas and Sikk2017).

Additional studies considering name change as an independent variable have demonstrated its significant consequences for parties’ organizational dynamics and electoral success. For example, Gouglas and colleagues (Reference Gouglas, Katz, Maddens and Brans2021) show that parties that change their name are more likely to experience legislative turnover and renew the composition of their parliamentary group. Avina (Reference Avina2024) examines the effect of rebranding on electoral performance and finds that parties that rename themselves tend to gain approximately 0.74 percentage points more in vote share compared with their previous results. In contrast, he does not find a similar effect for programmatic changes in party manifestos. Ishiyama and Marshall (Reference Ishiyama and Marshall2017) identify renaming by former rebel groups as a critical stage in democratization in postconflict settings, as previously violent groups transition into electoral parties. Overall, these studies indicate that party (re)naming is not random but rather a strategic choice with meaningful consequences for a party’s trajectory.

While these studies emphasize the external outcomes of name change, they also identify a key mechanism related to internal organizational constraints that shape whether and how such transformations occur. As Ishiyama and Marshall (Reference Ishiyama and Marshall2017) and Ishiyama and Rybalko (Reference Ishiyama and Rybalko2023) argue, the extent to which party leaders must accommodate internal factions and constituencies, following the logic of nested games (Tsebelis Reference Tsebelis1991), significantly shapes the branding strategies parties can adopt. Leaders cannot unilaterally change a party’s brand; instead, they must negotiate with members who may be deeply invested in the party’s existing name and broader identity (Lupu Reference Lupu2016). This focus on organizational constraints underscores that rebranding is not merely a strategic act aimed at external audiences, but a negotiated process shaped by internal dynamics.

Another strand of literature examines party names, including those of newly formed parties, as a central element in communicating a party’s appeal. Parties enter the public sphere through their name, which is how elites and voters refer to them and how they appear on the ballot. Names are often complemented by additional symbols, such as logos and specific color schemes (Avina and Spoon Reference Avina and Spoon2025; Casiraghi, Curini, and Cusumano Reference Casiraghi, Curini and Cusumano2023; Casiraghi, Cusumano, and Chryssogelos Reference Casiraghi, Cusumano and Chryssogelos2024). Dolinsky (Reference Dolinsky2023) provides a dataset of group appeals in party names in Israel and the Netherlands between 1977 and 2015. She finds that 44% of Dutch parties and about a third of Israeli parties include a group appeal in their name, most often a religious reference. Similarly, Ishiyama and Breuning (Reference Ishiyama and Breuning2011) study ethnic parties in postcommunist countries and show that parties that do not include an ethnic group’s name tend to be more inclusive, attract some support from the majority group, and are more likely to encourage their supporters to accept the democratic framework of the nation-state they inhabit.

This body of literature faces three key limitations that we aim to address: (1) the lack of a systematic content-based perspective, (2) an exclusive focus on the electoral arena, and (3) insufficient evidence on the electoral consequences of party names. We address the first two limitations by analyzing references in party names along two dimensions: ideological and organizational. The ideological dimension captures the extent to which party names explicitly reference the ideological traditions of established party families, such as Christian democrats, socialists, liberals, and conservatives. These party families have historically structured political competition and played a key role in integrating European societies after World War II (e.g., Kirchheimer Reference Kirchheimer, La Palombara and Weiner1966). However, as societies become more fragmented and electoral competition intensifies, parties may strategically abandon explicit ideological markers in their name to attract a broader electorate with weaker partisan ties and less attachment to historically dominant ideological traditions.

The organizational dimension, by contrast, concerns the organizational model and identity referenced in a party’s name. As the crisis of political parties deepens, some parties may move away from explicitly calling themselves a “party,” instead opting for alternative organizational labels that may appear more attractive. Building on the emerging literature on party–movement interactions and movement parties (e.g., Castelli Gattinara and Pirro Reference Casiraghi, Cusumano and Chryssogelos2024; Della Porta et al. Reference della Porta, Fernández, Kouki and Mosca2017; Pirro and Castelli Gattinara Reference Pirro and Gattinara2018; Tarrow Reference Tarrow2021), we argue that references to horizontal organizational structures associated with movement politics may be perceived as a more viable alternative to the traditional party model. Rather than considering electoral competition in isolation, we argue that parties position themselves as potential allies of civil society, social movements, and other actors engaged in nonelectoral mobilization.

Building on the distinction between the ideological and organizational dimensions, we develop a novel typology that classifies party names along a continuum from classical to nonclassical. We propose a typology that distinguishes six mutually exclusive categories: (1) classical party names (e.g., the Conservative and Unionist Party in the United Kingdom), (2) ideological nonparty names (e.g., the Christian Democratic Union in Germany), (3) ideological movement names (e.g., the Panhellenic Socialist Movement in Greece), (4) nonideological party names (e.g., the Democratic Party in Italy), (5) nonideological nonparty names (e.g., Law and Justice in Poland), and (6) nonideological movement names (e.g., the Five Star Movement in Italy). From these six categories, we focus empirically in particular on classical party names at one end of the continuum, nonideological movement names at the other, and nonideological nonparty names as an intermediate category within the nonclassical spectrum.

To address the third limitation in the literature, we examine the demand-side consequences of party name choices and investigate the causal effect of party names on electoral appeal using two survey experiments. In line with existing research, we distinguish between newly established party names and name changes among existing parties.

Before outlining our research design, we first theorize how the crisis of representation may shape party name choices, identify the parties most likely to adopt nonclassical names, and discuss the expected electoral consequences of these strategic naming decisions.

Why Shift to Nonclassical Names? The Crisis of Representation

Academic literature and public discourse increasingly argue that liberal democracies are in crisis. In some cases, democratic backsliding is evident, driven by authoritarian forces that weaken liberal checks and balances (e.g., Bohle, Greskovits, and Naczyk Reference Bohle, Greskovits and Naczyk2024; McCoy, Rahman, and Somer Reference McCoy, Rahman and Somer2018). In other cases, the crisis remains latent, yet to fully materialize (e.g., Valgarðsson et al. Reference Valgarðsson, Jennings, Stoker, Bunting, Devine, McKay and Klassen2025). A key aspect of this crisis is the declining ability of mainstream parties to connect citizens with the state (e.g., Katz and Mair Reference Katz and Mair1995; Mair Reference Mair2013; Schmitter Reference Schmitter, Diamond and Gunther2001). As the mass party model declined (Duverger Reference Duverger, North and North1954), parties withdrew from representing distinct societal milieus. Since the 1960s, they have increasingly positioned themselves as catchall organizations, appealing to broad electorates rather than specific groups (Kirchheimer Reference Kirchheimer, La Palombara and Weiner1966). This crisis of democratic representation has both attitudinal (distrust and perceived misrepresentation) and behavioral (declining electoral participation) dimensions (Mainwaring, Bejarano, and Pizarro Leongómez Reference Mainwaring, Bejarano, Leongómez, Mainwaring, Bejarano and Leongómez2006). Kirchheimer’s (Reference Kirchheimer, La Palombara and Weiner1966) diagnosis, that parties are losing their role as intermediaries between citizens and government, has thus only become more relevant since he originally formulated it.

The crisis of representation unfolds through both structuralist and functionalist dynamics. From a structuralist perspective, the crisis stems from mainstream parties’ inability to adapt to shifting cleavage structures, as they face competition from both the radical Right and the new Left (Hooghe and Marks Reference Hooghe and Marks2018; Kriesi et al. Reference Kriesi, Grande, Lachat, Dolezal, Bornschier and Frey2008; Reference Kriesi, Grande, Dolezal, Helbling, Höglinger, Hutter and Wüest2012). A key empirical indicator of this crisis is programmatic convergence, which dilutes party brands, making it harder for voters to distinguish between electoral alternatives. This increases uncertainty and weakens retrospective accountability. Several factors drive this process, including exogenous shocks (Hutter and Kriesi Reference Hutter and Kriesi2019; Lupu Reference Lupu2016), incumbency (Roberts Reference Roberts2014), electoral losses (Janda et al. Reference Janda, Harmel, Edens and Goff1995), and intraparty conflicts (Pasotti Reference Pasotti2010). For instance, Lupu (Reference Lupu2016) argues that brand dilution led to the electoral collapse of established parties in Latin America. Mainstream parties, particularly under economic pressure, converged in their appeals, incentivizing incumbents to deviate from party lines in favor of pragmatic policy making (see also Roberts Reference Roberts2014). However, this inconsistency deepens dissatisfaction with democracy and contributes to declining turnout rates (Dassonneville and McAllister Reference Dassonneville and McAllister2020).

From a functionalist perspective, the crisis is reflected in the increasing volatility and fragmentation of party systems, as well as the growing success of new parties (Casal Bértoa and Enyedi Reference Casal Bértoa and Enyedi2021; Dassonneville Reference Dassonneville2022; Engler Reference Engler2023; Haughton and Deegan-Krause Reference Haughton and Deegan-Krause2020). From this perspective, the crisis of mainstream parties is linked to their organizational brand: long-established governing parties struggle against emerging challengers (De Vries and Hobolt Reference de Vries and Hobolt2020; Haughton and Deegan-Krause Reference Haughton and Deegan-Krause2020; Sikk Reference Sikk2011), which mobilize against the perceived aloofness of established parties (Mudde Reference Mudde2007; Van Kessel Reference van Kessel2015). Electoral volatility, often taken as a key empirical indicator, signals parties’ difficulty in maintaining stable voter bases (Dalton and Wattenberg Reference Dalton and Wattenberg2000). This is particularly concerning when volatility entails ideological shifts (Dassonneville Reference Dassonneville2022) or the rise of new parties (Engler Reference Engler2023; Haughton and Deegan-Krause Reference Haughton and Deegan-Krause2020). High volatility is also associated with weak party system institutionalization (Casal Bértoa and Enyedi Reference Casal Bértoa and Enyedi2021), undermining retrospective accountability when voters cannot reward or punish parties that may not survive until the next election (Borbáth Reference Borbáth2021).

One consequence of this crisis of representation and the broader hollowing out of democracies (Berman Reference Berman2021; Mair Reference Mair2013) is the normalization of protest participation (Giugni and Grasso Reference Giugni and Grasso2019; Meyer and Tarrow Reference Meyer and Tarrow1998). Participation outside political parties, such as attending demonstrations, has become a widely accepted form of civic engagement, influencing norms of citizenship and political involvement (Dalton Reference Dalton2008a). As demonstrations and other protest activities increasingly complement or substitute voting (Oser Reference Oser2022; Rojon et al. Reference Rojon, Pankowska, Vittori and Paulis2025), parties are incentivized to adapt their organizational form and public image. Many new parties, in particular, turn to the protest arena as a source of organizational innovation (Castelli Gattinara and Pirro Reference Castelli Gattinara and Andrea2024; Della Porta et al. Reference della Porta, Fernández, Kouki and Mosca2017; Tarrow Reference Tarrow2021), contributing to the erosion of boundaries between electoral and protest politics. This convergence creates a strategic rationale for new entrants to avoid classical “party” labels. Emphasizing horizontal, flexible, and citizen-oriented organizational models, they seek to distance themselves from hierarchical and elite-driven connotations associated with traditional party identities. In this context, movement names become a tool for signaling proximity to grassroots mobilization. Classical labels, by contrast, may evoke institutional inertia or reinforce the crisis of representation, making them less attractive to voters. As such, the shift toward movement names might reflect a calculated effort to frame the party’s brand in ways that carry electoral appeal.

Drawing on this discussion, we formulate our first expectation: as the crisis confronting political parties deepens, and as many seek to define themselves in less hierarchical or ideologically rigid ways, the composition of party systems increasingly reflects a shift toward parties with nonclassical names. For clarity, classical party names reference both a traditional organizational label (i.e., “Party”) and an established ideological anchor (e.g., “Christian Democrat,” “Conservative,” “Socialist”). In contrast, nonclassical names signal a break with these conventions. In more formal terms, we expect the following:

Who Adopts Nonclassical Names? Challenger versus Mainstream Parties

The long-term transformation of European party systems, unfolding against the background of the crisis of representation, does not affect all parties in the same way. Parties respond differently to these crisis conditions. Challenger parties, less constrained by entrenched organizational and ideological legacies, can adapt their brand more easily, while mainstream parties tend to be slower to respond. Therefore, challenger parties are expected to play a key role as vanguards of this transformation. As already emphasized in the discussion of H1, this party-level dynamic, particularly the strategic behavior of new entrants, drives broader party system change over time. However, conceptualizing the challenger status differs between the structuralist and functionalist perspectives introduced above.

The structuralist tradition points to the importance of ideological factors, highlighting political parties that challenge mainstream parties ideologically and politicize issues previously absent from electoral dynamics (Hooghe and Marks Reference Hooghe and Marks2018; Kriesi et al. Reference Kriesi, Grande, Lachat, Dolezal, Bornschier and Frey2008). The issues that challenger parties mobilize, such as immigration, European integration, and climate change (De Vries and Hobolt Reference de Vries and Hobolt2020; De Wilde et al. Reference de Wilde, Koopmans, Merkel, Strijbis and Zürn2019; Dickson and Hobolt Reference Dickson and Hobolt2025), cut across the preferences of the mainstream party landscape. From this perspective, the challenger status is primarily associated with the radical-right and green/new-left party families, which do not integrate neatly into the traditional left–right dimension. As the issues they politicize have gained importance in the shadow of Europe’s multiple crises (Hooghe and Marks Reference Hooghe and Marks2018; Hutter and Kriesi Reference Hutter and Kriesi2019), radical-right and green parties have increased in popularity, further challenging their mainstream competitors and heightening the fragmentation of European party systems.

By contrast, the functionalist perspective is less concerned with parties’ ideological positions and focuses more on their strategic position in the party system (e.g., Casal Bértoa and Enyedi Reference Casal Bértoa and Enyedi2021). By classifying parties as dominant or challengers based on whether they had previously entered government, De Vries and Hobolt (Reference de Vries and Hobolt2020), for instance, document an electoral shift in Western Europe over time from dominant to challenger parties. While De Vries and Hobolt (Reference de Vries and Hobolt2020) emphasize that ideological positioning also plays a role, Sikk (Reference Sikk2011) proposes that “newness” in itself can be a winning formula for challengers. Through a comparative analysis of four Eastern European parties, Sikk shows that new parties can succeed even when they occupy the same ideological position as previously existing parties or, as Engler (Reference Engler2023) finds, when they do not take a clear position on the main divides structuring the party system. Similarly, Haughton and Deegan-Krause (Reference Haughton and Deegan-Krause2020) distinguish between old and new parties, identifying the emergence of new parties in Central and Eastern Europe as posing the main challenge to established parties.

The literature on the interactions between movement and electoral politics integrates insights from both the structuralist and functionalist perspectives. In terms of ideology, radical-right (e.g., Castelli Gattinara and Pirro Reference Castelli Gattinara and Andrea2024) and green/new-left parties (e.g., Blings Reference Blings2020; Della Porta et al. Reference della Porta, Fernández, Kouki and Mosca2017) have featured prominently in these analyses. Kitschelt (Reference Kitschelt, Katz and Crotty2006), for example, points to radical-right and green parties as archetypal “movement party entrepreneurs,” which invest little in resolving collective and social choice problems. In other words, these parties are less likely to invest in organizational hierarchies that constrain leaders, and are also less likely to adopt encompassing electoral programs. In addition, this research strand also examines factors typically associated with a functionalist perspective. In this regard, studies indicate that opposition status is the main predictor of a party’s engagement with “street” mobilization (Hutter and Vliegenthart Reference Hutter and Vliegenthart2018). Particularly in Central and Eastern European countries, parties that belong to the mainstream or the challenger party families are likely to mobilize when they are in opposition (Borbáth and Hutter Reference Borbáth and Hutter2021).

To summarize, the literature points to several party-level features associated with breaking from classical party names. These include newness, opposition status, and membership in challenger party families (e.g., radical right, green). By omitting ideological labels or established organizational tags like “party,” these challengers signal their alternative approach and outsider status. Accordingly, we formulate our second hypothesis as follows:

How Do Nonclassical Names Influence Voters? A Demand-Side Perspective

From a demand-side perspective, the literature primarily examines how citizens develop interconnected networks of political information around party brands (e.g., French and Smith Reference French and Smith2010; John et al. Reference John, Loken, Kim and Monga2006). The brand labels or names serve as informational cues or heuristics, helping voters to navigate complex political environments and form structured evaluations of parties (Lau and Redlawsk Reference Lau and Redlawsk2001; Popkin Reference Popkin1994). As Snyder and Ting (Reference Snyder and Ting2002) argue from a theoretical perspective (supported by empirical illustrations), party names play a key role in enabling citizens to hold parties accountable.

From an experimental standpoint, Van Elk, Van Schie, and Bekkering (Reference van Elk, van Schie and Bekkering2010) demonstrate that even left–right spatial metaphors embedded in party names can automatically activate political associations. This activation is also sensitive to a voter’s own ideology, suggesting that labels alone can polarize or prime voters’ thinking. Along similar lines, Lau and Redlawsk (Reference Lau and Redlawsk2006) provide evidence that voters rely on party labels to reduce cognitive effort, often substituting these heuristics for more detailed information about policy or ideology. However, recent work by Titelman and Lauderdale (Reference Titelman and Lauderdale2025) suggests that the marginal effect of party labels may be limited when voters already possess clear information about the ideological positioning of the political supply.

When parties adopt nonclassical brands or names, they may do so based on the assumption that classic labels (e.g., “socialist,” “conservative,” “Christian democratic”) have lost their broad appeal, an idea supported by the research on the crisis of representation, declining partisan ties, and voter volatility across Europe (Dalton and Wattenberg Reference Dalton and Wattenberg2000; Katz and Mair Reference Katz and Mair1995; Kirchheimer Reference Kirchheimer, La Palombara and Weiner1966; Mair Reference Mair2013; Valgarðsson et al. Reference Valgarðsson, Jennings, Stoker, Bunting, Devine, McKay and Klassen2025). This crisis creates conditions for new or challenger parties to exploit the perception of traditional elites as aloof by branding themselves in novel ways (Mudde Reference Mudde2007; Van Kessel Reference van Kessel2015). Nonclassical names can thus convey novelty or a break from discredited elites, appealing especially to voters who seek alternatives to perceived establishment politics. Kam (Reference Kam2005) shows, for instance, that some voters are more sensitive to cues than others, potentially making nonclassical names especially effective among those who are not attached to established parties any more.

The case of En Marche! in France illustrates how such naming choices are not accidental, but part of a strategic plan. As Adrien Taquet, the communications advisor behind the name, recalled, the label was designed to evoke movement, disruption, and emotional resonance with collective marches throughout history, from the French Revolution to the civil rights movement, while also subtly embedding Emmanuel Macron’s initials in the acronym (Lorrain Reference Lorrain2017). While voters are not misled about whether such a formation is a party, the name was intentionally chosen by Macron’s camp to frame their initiative as civic, open, forward looking, and unaffected by traditional partisan baggage.

A related development is the rise of expressive voting, in which voters do not necessarily base their choice on programmatic or instrumental evaluations of which party best aligns with their ideological preferences. Instead, they may vote in ways that reflect emotional or symbolic attachments, sometimes prompted by charismatic leadership, populist appeals, or distinctive party labels (Brennan and Lomasky Reference Brennan and Lomasky1997; Hamlin and Jennings Reference Hamlin and Jennings2011; Schuessler Reference Schuessler2000). For instance, Hobolt and Tilley (Reference Hobolt and Tilley2014) demonstrate that citizens’ group-based predispositions toward the European Union can outweigh standard partisan cues, causing them to shift blame or credit in ways that do not align with traditional loyalties.

In combination, parties’ strategic rebranding attempts and citizens’ shifting motivations can amplify the power of nonclassical brands. These alternative brands potentially offer narratives of renewal, authenticity, or outsider status. Consequently, nonclassical party names may outcompete classical labels, which are often associated with the status quo or perceived as disconnected from voter realities. Drawing out the individual-level implications of these trends, we hypothesize the following:

Study 1: Party Names

One reason researchers have not conducted a comparative study on the content of party names is the lack of a comprehensive dataset on party name changes over a long period. Existing datasets, such as ParlGov (Döring and Manow Reference Döring and Manow2024) or the Manifesto Data Collection (Lehmann et al. Reference Lehmann, Franzmann, Burst, Regel, Riethmüller, Volkens, Weßels and Zehnter2023), focus on continuity and track the same parties over the longest time frame possible, but they do not systematically cover differences in the names of political parties.

Therefore, we undertook a major data collection effort. We used ParlGov as our starting point and manually coded changes in the names of all parties in this dataset. During this process, we relied primarily on Wikipedia to trace the evolution of party names over time, consulting both the English-language Wikipedia pages and the typically more detailed national-language pages of the respective parties. Our focus was on the name under which parties appeared on the ballot paper.Footnote ¹ To distinguish between new and old parties, we followed ParlGov’s decisions on whether a party received a new party identification.Footnote ² Although Wikipedia is widely used in political science (Brown Reference Brown2011) and in studies of party competition (Herrmann and Döring Reference Herrmann and Döring2023), offering comprehensive coverage of political parties, we still wished to validate its information with additional sources. Most importantly, we used internal documents from the Manifesto Project (Lehmann et al. Reference Lehmann, Franzmann, Burst, Regel, Riethmüller, Volkens, Weßels and Zehnter2023), as well as secondary literature. The resulting data cover 616 unique parties from 28 European countries between 1945 and 2023. In countries that democratized later, we covered the period starting with the first competitive election.Footnote ³ We used this dataset to test our over-time hypotheses (H1) and to examine which parties are most likely to adopt nonclassical names (H2).

In classifying party names, we considered both ideological and organizational references. Among the former, we singled out nonideological references. Importantly, by nonideological references, we do not mean labels devoid of all value content, but rather labels that do not reference historically rooted ideological antagonisms associated with traditional party families. More concretely, these names do not refer to classical party families or ideologies such as socialism, Christian democracy, liberalism, and so forth. Under nonideological references, we included labels containing an action verb or referencing values, the country’s name, the leader’s name, anti-elitism, reforms, democracy, the people, or even broader statements.Footnote ⁴ Figure 1 in online appendix A includes more information on the distribution of each of these terms.

Regarding organizational references, we coded both the absence of the term “party” in the label and the presence of a movement reference. These two are mutually exclusive: parties that do not call themselves “party” may go a step further and adopt a movement label, though not all do so. While “party” may be associated with a hierarchical, potentially rigid structure, references to movements evoke more horizontal and inclusive forms of organization. Among movement references, we included names suggesting an image of marching in the streets. Accordingly, we coded “rally,” “block,” “front,” “forces,” “force,” “movement,” “action,” “attack,” “fighters,” “march,” and “revolution” as movement references. Figure 2 in online appendix A shows the prevalence of each of these terms. Drawing on these two dimensions, we define a typology of six possible party name types as previously described. We focus in particular on the prevalence of three of these types: classical party names, nonideological movement names, and nonideological nonparty names.

To explain variation in our three dependent variables, we included several types of party- and context-level information. On the party level, we used the eight party families included in ParlGov: agrarian, Christian democracy, communist/socialist, conservative, green/ecologist, liberal, right wing, and social democracy.Footnote ⁵ We classify Christian democratic, conservative, liberal, and social democratic parties as mainstream. To examine differences between new and old parties, we included the logged value of party age.Footnote ⁶ For the strategic position, we used a dichotomous indicator marking whether a party had ever held the prime ministership. To address the role of internal organizational constraints, we also examined how branding choices relate to party cohesion and the concentration of leadership power (taking their role in the candidate selection process as a proxy).Footnote ⁷ Depending on the analysis, we also controlled for current opposition status, vote share, and electoral coalitions.

As contextual information, we added country-year-level indicators typically associated with the crisis of representation and changing party systems. To account for programmatic convergence, we used polarization as the key indicator. Since many studies have shown that electoral competition in Europe is multidimensional (e.g., Hooghe and Marks Reference Hooghe and Marks2018; Kriesi et al. Reference Kriesi, Grande, Lachat, Dolezal, Bornschier and Frey2008; Reference Kriesi, Grande, Dolezal, Helbling, Höglinger, Hutter and Wüest2012), we distinguished between economic and cultural polarization. We measured both dimensions using party manifesto data (Lehmann et al. Reference Lehmann, Franzmann, Burst, Regel, Riethmüller, Volkens, Weßels and Zehnter2023), employing the same issue classification as Hutter and Gessler (Reference Hutter, Gessler, Hutter and Kriesi2019, 67) and the formula proposed by Dalton (Reference Dalton2008b, 906). As functional features, we included electoral volatility based on the updated dataset of Emanuele, Chiaramonte, and Soare (Reference Emanuele, Chiaramonte and Soare2020). We also incorporated turnout in the last national parliamentary elections from the IDEA (2024) dataset and the quality of electoral democracy index from the V-Dem Country-Year Dataset (Coppedge et al. Reference Coppedge, Gerring, Knutsen, Lindberg, Teorell, Altman and Angiolillo2024) to account for variation in exposure to the crisis of democracy. Finally, we controlled for presidentialism (as opposed to parliamentarism) using the indicator from the Quality of Government Standard Dataset (Teorell et al. Reference Teorell, Sundström, Holmberg, Rothstein, Pachon, Dalli, Valverde, Phiri and Gerber2025).

All models included period fixed effects to test our hypotheses on over-time dynamics.Footnote ⁸ Our periodization follows theoretical considerations about the evolution of European party systems and empirical considerations regarding data density. Specifically, we used 1945 to 1969 as the reference category, then grouped subsequent periods into 1970 to 1989, 1990 to 2008, and 2009 to 2023.

Multiple levels of analysis structured the resulting data; party names were nested within parties and years. We relied on a three-level model, with parties nested in country-years and countries.Footnote ⁹ We sought to capture both name changes and continuity, so we included the “negative cases” of no change. To do so, we expanded the dataset so that each party-year combination was present from the first election in which the party participated until its disappearance. Because our dataset comprises the full population of European parties, we focused on substantive effect sizes rather than p-values.

Study 2: Survey Experiment

To capture cross-national variation in Europe’s crisis of representation, we selected four cases for our survey experiment: Italy, Germany, Austria, and Hungary (see online appendix D for a detailed discussion of case selection). In this study, we focus on similarities across contexts, as we are primarily interested in findings that generalize beyond single country cases.

To examine how party names affect voters’ attitudes, we conducted two conjoint survey experiments. The first experiment focused on newly formed parties whose policy positions are ideologically congruent with respondents’ own positions. The second experiment examined the rebranding of currently existing parties supported by respondents. We provide more details on the second experiment in online appendix D and concentrate on the first experiment in the main text. We use these experiments to test our hypothesis on the appeal of nonparty names (H3).

Both experiments were part of a research project on movement brands and also included treatments related to party organizations and mobilization repertoires. We use the results concerning party organizations and mobilization repertoires as a benchmark when assessing the effects of party names. The survey was administered in 2023, with similar sample sizes in all four countries.Footnote ¹⁰ It was conducted online by the company Bilendi, using a quota sample representative of each country’s population regarding age, gender, and education. The ethics committee of WZB Berlin Social Science Center approved the study, and we preregistered both the design and our hypothesis.Footnote ¹¹

We designed the experiment to simulate the entry of a new party into the respective country’s parliament. The fictitious new party advocated policy positions that were ideologically congruent with respondents’ preferences (see online appendix D). We employed these congruent issue positions to isolate the marginal effect of the party’s brand from its ideological alignment, even though this approach may reduce the likelihood of detecting large effects (Titelman and Lauderdale Reference Titelman and Lauderdale2025).

Respondents were then shown two profiles (group A and group B), which both took the following form:

The two profiles were randomly generated from predefined values, including the order of the three bullet points. Table 1 presents the predefined values.

Table 1 Design of the New Party Branding Conjoint Experiment

Each treatment included an empty reference category, which served as a baseline comparison. We used the four name-treatment values—(1) empty, (2) New [country name], (3) Party for a New [country name], and (4) Movement for a New [country name]—to evaluate our hypothesis. Respondents were asked to evaluate two profiles three times.

To measure our dependent variables, respondents answered two items after both experiments. First, they rated their support for the two profiles on a scale from zero to 10 (“How much do you support the described group/scenario ‘A/B’?”). Second, they chose between the two profiles: “Which of the two organizations/scenarios would you be more likely to vote for?” Our results remain robust across both measurement techniques. In the main text, we focus on the choice (trade-off) item; see online appendix E for the results based on the continuous scale. Following our preregistration, we also performed additional analyses of heterogeneous effects by satisfaction with democracy, left–right ideology, strength of party identification, history of participation in demonstrations, internal and external political efficacy, populist attitudes, trust, and the issues on which respondents self-selected into treatment groups (positions and salience) (see online appendix E).Footnote ¹²