Introduction

Although the study of political parties and the study of protests and social movements mainly developed independently (McAdam and Tarrow 2010), their relationship is re-examined by the emerging literature on party–movement interactions (e.g., Hutter et al. 2019; Pirro and Gattinara 2018; Tarrow 2021). This line of literature demonstrates that a wide range of parties belonging to the far-right (Caiani and Císař 2019; Castelli Gattinara et al. 2022), economically left-wing (Della Porta et al. 2017), and mainstream families (Borbáth and Hutter 2021) organize protest events on the streets. The previously entrenched assumption of mobilization in the electoral and protest arenas being isolated and following distinct logics is increasingly being challenged by several case studies (e.g., Borbáth and Paxton 2024; Pirro and Gattinara 2018; Weisskircher et al. 2022) and comparative analysis (e.g., Caiani andí Císař 2019; Della Porta et al. 2017; Borbáth and Hutter 2021; Císař and Vráblíková, 2019; Castelli Gattinara et al. 2022; Tarrow 2021). Existing debates relate to the endurance of parties’ movement character, with Kitschelt (2006) famously arguing for the transitional nature of the phenomena (see Anria 2018 for a counter-perspective). Missing from this line of literature is an investigation of the mechanisms driving the dynamic of party’s presence in protest.

In this paper, we examine the question of what mechanisms are driving parties’ involvement in protests before and after entry to parliament. To do so, we follow a relational approach (Hoffmann et al. 2023; McKeever 2021; Pirro et al. 2021) and identify the type of networks that parties in protest politics mobilize. We make at least four contributions to the study of movement parties. Firstly, we revisit the theory of increasing differentiation between the electoral and protest arenas against the background of over time (Anria 2018; Kitschelt 2006) and cross-national (Borbáth and Hutter 2021) variation in parties’ involvement in protests. Secondly, we consider the expectations regarding the type of parties more likely to protest based on theories that expect protest to be driven by niche and left-wing actors (Císař and Vráblíková 2019; Hutter et al. 2019, p. 325; Nonnemacher 2023). Thirdly, we analytically introduce and empirically distinguish three different mechanisms of parties' reliance on protest as (1) a signaling device of issue ownership, (2) an alliance-building tool, and (3) a means to strengthen the local presence of the party. Fourth, we combine the literature on issue ownership in electoral competition with the literature on protest politics and adapt the Hobolt and De Vries (2015) indicator to apply it to parties' involvement in the protest arena.

We take a dynamic perspective and study movement party mobilization over time (see the introduction to this special issue: Borbáth and Paxton 2024) in Hungary. We consider Hungary a case representative of the Central and Eastern European (CEE) region, characterized by a comparatively high involvement of protest parties (Borbáth and Hutter 2021), low nonelectoral participation (Borbáth and Gessler 2020), and a protest issue agenda detached from party politics (Císař and Vráblíková, 2019). We compare protests by two parties that entered the Hungarian parliament in 2010: the radical right Jobbik—Movement for a Better Hungary and the green LMP—Politics Can Be Different.Footnote 1 We examine the period after the emergence of the two parties up to the most recent parliamentary elections (2002–2022).

The two parties are not new to the comparative politics literature (e.g., Kovarek and Littvay 2019; Mikecz 2017), with Jobbik being particularly often-cited as one of the textbook examples of far-right movement parties (e.g., Caiani and Císař 2019; Castelli Gattinara et al. 2022; Mikecz 2014; Pirro 2019; Pirro et al. 2021; Pirro and Gattinara 2018). We build on this literature on Hungarian party competition, but our approach differs in three respects. First, unlike existing studies that focus on the protest involvement of Jobbik (e.g., Caiani and Císař 2019; Castelli Gattinara et al. 2022; Mikecz 2014; Pirro 2019; Pirro et al. 2021; Pirro and Gattinara 2018) or Fidesz (Greskovits 2020; Susánszky et al. 2016), we broaden the perspective to include the second most successful party that entered after 2010, LMP. Based on a comparative assessment of the trajectories of the two new parties, Jobbik and LMP, we identify mobilization in protest politics as a strategy of party system transformation from below. Second, unlike previous studies, our analysis considers the overall mobilization in the Hungarian protest landscape (for a similar approach, see Greskovits and Wittenberg 2016). This allows us to embed the efforts of both parties in the broader context of protest politics. Third, we focus on the temporal dynamics of party involvement in protest politics, comparing protest mobilization before and after each party entered parliament. In doing so, we contribute to the debate on the transitional character of movement parties by using a research design that zooms in on a well-identified national context and holds constant changes over time in key background conditions.

Our results show that the two new parties heavily rely on protest mobilization. Both are more likely to protest than mainstream left or right formations. Yet, the protest involvement of the two is driven by different mechanisms. Particularly before entering parliament, Jobbik relied on protest to increase the association between the party and the radical right issue it owns while also using protest mobilization to enter alliances with radical right movements. In contrast, LMP mainly organizes protests on topics other than the environment, mostly in cooperation with nongreen actors. Over time, the protest presence of both parties is dominated by events outside of Budapest.

In what follows, we first provide an overview of the literature on protest mobilization with a focus on the role of parties in organizing events. Then, we discuss the Hungarian case and the context of mobilization by Jobbik and LMP. In the empirical part, we present the dynamic of protest mobilization by political parties in Hungary and examine the protest network of the two parties. The concluding section discusses the results and places them in the context of the broader scholarly literature.

Parties and the protest arena: an uneasy relationship

According to Kitschelt’s conceptualization (2006), ‘movement parties’ are social movements competing in elections. Being characterized by low organizational and ideological investment levels, movement parties are not expected to survive the test of time. They emerge because of a mismatch between the arena where political entrepreneurs mobilize, their organizational resources, and their claim’s narrow focus. The tension is expected to become especially strong after movement parties enter parliament and face the pressure of taking a position on the tradeoffs involved in policy making.

Kitschelt’s influential conceptualization of movement parties as a transitional phenomenon is challenged by Anria (2018) with an account of the Bolivian MAS party. The case of MAS shows that parties that originate from social movements, competing in the context of “strong, dense, and autonomous civil society” (2018, p. 41), are likely to maintain an intra-party balance of power between leadership and associated civil society organizations. This intra-party balance of power introduces decision-making mechanisms that hold the leadership accountable, even in the case of charismatic leaders such as Evo Morales (for the role of intra-party divisions in the European context, see Weisskircher et al. 2022). The emerging equilibrium allows parties like MAS to maintain their movement character beyond a transitional phase.

The diverging expectations are partly explained by the scope conditions of the two perspectives. Kitschelt’s account has been developed to explain the dynamic in advanced democracies, where, over time, the protest and electoral arenas became well differentiated (Borbáth and Hutter 2021). Anria’s account focuses on Bolivia, a country democratized in 1982 and situated in a region, where the electoral and protest arenas are highly intertwined (e.g., Almeida 2010). Given the background conditions, the CEE region resembles Anria’s account of enduring party involvement in protests for two reasons.

First, the relatively young age of CEE democracies leads to comparatively low differentiation between the protest and electoral arenas (Borbáth and Hutter 2021).Footnote 2 This is, on the one hand, due to the long time horizon needed for learning processes to unfold. Kitschelt (2003, 2006) conceptualize long-term learning as the driver of the differentiation of electoral and movement politics. On the other hand, low differentiation is reinforced by historical legacies of strong interactions between parties and movements during the transition (Glenn 2003). Second, the lack of organizational autonomy and weak civil society participation leads to a central role played by electoral politics. Despite the debate on the strength of civil society (e.g., Foa and Ekiert 2017; Howard 2003), there is a virtual consensus on low participation in nonelectoral forms (e.g., Borbáth and Gessler 2020). As the accounts emphasizing the strength of civil society highlight, movements' influence on policy often originates from close ties to political parties (Petrova and Tarrow 2007). The low participation in civil society creates an environment, where parties can instrumentally rely on protest mobilization for electoral purposes. Accordingly, a broad range of parties in CEE regularly get involved in protest mobilization (Borbáth and Hutter 2021) and can likely maintain their movement character even in the long term.

Protesting parties: mainstream, niche, left, and right

Existing studies show that some parties are more likely to protest than others. In a long-term comparison of six Western European countries, Hutter (2014b, pp. 123–126) finds a strong presence in the protest arena in the case of the radical left and, to a lower extent, in the case of extreme right parties. The over-representation of the extreme left and right parties is in line with the findings of the literature on the far-right as a social movement (Pirro and Gattinara 2018). It also aligns with the original formulation of the movement party theory, which expected green and radical right formations to represent the closest approximation of the ideal type (Kitschelt 2006). There are two explanations for the strong presence of these types of parties in protests.

The first explanation starts by comparing the agenda of electoral and protest politics (e.g., Císař and Vráblíková, 2019; Nonnemacher 2023). Císař & Vráblíková (2019) show that protest is more likely on issues that are less salient and lie outside of the primary dimension of electoral competition. According to their findings, in Hungary before 2006, when the party system was characterized by high cultural and low economic polarization, there was a high share of economic issues in the protest arena. The pattern changed after 2006, when the share of economic and cultural protest balanced out due to the substantial increase in protest events with culturally conservative claims. They theorize that niche parties, primarily the green and the radical right, compete in interaction with protest mobilization by increasing the salience of second-dimension issues and decreasing the salience of first-dimension issues. However, they do not examine the mechanisms behind the issue linkage between the two arenas. Although their theory implies that niche parties mobilize in protest as one element in their strategy of shifting issue salience, the assumption remains untested (also see Nonnemacher 2023). From this perspective, the critical distinction in party-protest involvement is captured by the mainstream-niche dichotomy rather than ideological differences.

The second explanation emphasizes ideological proximity with other organizations mobilizing in the protest arena as a driving force of parties’ protest involvement. In Western Europe, the protest arena is predominantly characterized by left-wing mobilization. Accordingly, due to their proximate programmatic goals, it is easier for left-wing parties to ally with existing protest movements. There is empirical evidence of the reverse relationship in CEE: while protest participation is much lower in the region, the political right is more likely to protest (Borbáth and Hutter 2021; Kostelka and Rovny 2019), and party identification increases the ideological gap, suggesting mobilization by ideologically right-wing parties (Borbáth and Gessler 2020). Accordingly, ideological congruence between protest participants and mobilizing parties implies a high presence of left-wing and green parties in Western Europe and right or radical right parties in Eastern Europe.

Mechanisms of protest mobilization: issue ownership, alliance building, and local presence

Although from different starting points, the previous literature highlights niche parties and parties' ideological proximity to protest politics as the main drivers of protest involvement. We aim to reconcile the two perspectives and provide an account of three different mechanisms that explain why both parties may share a high level of mobilization in the protest arena. In this regard, we distinguish between issue ownership, alliance building, and local presence as mechanisms driving protest involvement.

The perspective emphasized by Císař and Vráblíková (2019) expects niche parties to shift the issue agenda away from the primary dimension of party competition. In this account, parties compete by emphasizing the issues they “own” and de-emphasizing the ones they are seen as less competent to tackle. The literature has shown that issue ownership, conceptualized as selective emphasis, drives competition in multiparty systems (e.g., Hobolt and de Vries 2015). However, issue ownership has exclusively been studied in the context of electoral politics.

As the social movement literature argued, protest is a signaling device able to influence the media, set the agenda of party competition, and reach the voters (e.g., Vliegenthart et al. 2016). Therefore, mobilizing in protest potentially allows niche parties to selectively emphasize issues and strengthen voters’ association between the party’s problem-solving capacity and the respective issue (Walgrave et al. 2012). Parties organizing in protests can be understood as a strategy to increase the salience of issues and signal issue ownership. We call this the mechanism of issue ownership.

The second perspective emphasizes ideological congruence between social movement organizations mobilizing in protest and parties aiming to get involved (Borbáth and Hutter 2021; Tarrow 2021). Typically, organizations that mobilize in protest are able to bring large groups of people to the ‘streets’ without being directly involved in electoral politics. The absence of other actors competing for votes allows political parties to engage in protest politics to find new allies for their programmatic and organizational goals.

As the social movement literature highlights (e.g., Della Porta and Diani 2010), networks and alliances are essential to understanding the dynamics in the protest arena. With nonelectoral forms of mobilization becoming normalized and a more diverse set of organizations mobilizing in protest (Meyer and Tarrow 1998), there are increasing opportunities for parties to broker alliances in the protest arena (McKeever 2021; Pirro et al. 2021). We call this the mechanism of alliance building.

As both perspectives assume, mobilization in protest allows parties to (re-)invigorate their ties to their social base. As Tavits (2013) has shown, especially in the case of new parties that did not build/ inherit a sizeable organizational network, mobilizing outside the political center represents a considerable organizational challenge. However, protest as an episodic form of mobilization requires less resources than investment in an enduring organizational infrastructure. Organizing or sponsoring a protest event allows the party to gain name recognition and recruit new supporters outside the electoral strongholds of the party. A party-sponsored protest event can be a first step in establishing a presence in places where ensuring a permanent organizational infrastructure is too resource-intensive and logistically challenging.

While analytically distinct, the three mechanisms are not necessarily empirically differentiated. They might take place simultaneously to the extent to which parties consider their presence in the protest arena advantageous for their issue profile, gaining new allies, or strengthening their presence at the local level. In this regard, the different mechanisms might reinforce each other and increase the presence of niche parties in protest. Before formulating our expectations on the three mechanisms, we present a qualitative account of the Hungarian case and the trajectory of the two parties we consider, Jobbik and LMP.

Protesting parties in Hungary

Bipolar competition and established parties

Shortly after the 1994 election, a bipolar competition emerged in the Hungarian party system, with the Fidesz—Hungarian Civic Alliance as the dominant right-wing party and the MSZP—Hungarian Socialist Party as the dominant left-wing party. From the two, Fidesz has maintained a closer relationship with civil society (similarly to the anti-communist parties in other countries, see Glenn 2003), creating the party-aligned Civic Circles movement from the opposition (Greskovits 2020) and occasionally mobilizing its supporters in government-sponsored marches after 2010 (Susánszky et al. 2016). Having the advantage of successor parties (Tavits 2013), MSZP inherited a widespread organizational network, allowing the party to have members and organizations in all major localities. Due to the ambivalent programmatic profile, the burden of the historical heritage of pre-1989 state-sponsored mobilizations, and the competitive advantage of its organizational network (Tavits 2013), MSZP avoided organizing in the protest arena. This led to the relative majority of right-wing actors in protest by the second half of the post-communist period (Greskovits and Wittenberg 2016). From a comparative perspective, Hungary became one of the most institutionalized party systems in CEE, with MSZP and Fidesz collectively receiving over 80% of the votes in 2002 and 2006.

However, shortly after the 2006 election, a political crisis emerged (Gessler and Kyriazi 2019), and a series of events disrupted the previous pattern of party competition. After winning the 2006 election in a head-to-head contest against Fidesz, in a speech addressed to his inner circle and subsequently leaked to the press, MSZP Prime Minister Ferenc Gyurcsány revealed the actual state of the economy and announced the need to introduce spending cuts. This revelation starkly contrasted with the promises made during the campaign to maintain and increase welfare and social benefits (Kim and Borbáth 2022). The fallout resulted in demonstrations and violent confrontations with the police. Fidesz led the charge against the MSZP government in parliament and on the streets. The protest movement resulted in a critical juncture and led to the collapse of MSZP and the emergence of Fidesz as a dominant party in government since 2010.

Tripolar competition with Jobbik and LMP

Fidesz has been opposed by a fragmented set of nominally left-wing parties and a strong radical right formation in the new tripolar equilibrium. In 2010, two new parties entered parliament: the radical right Jobbik and the green LMP. Both new parties followed a similar political/ discursive strategy of establishing a new competitive pole against Fidesz and MSZP. From the two, LMP has been ideologically closer to the mainstream opposition (including MSZP); nevertheless, both Jobbik and LMP refused to enter into alliances with MSZP against Fidesz. Instead, they aimed to selectively emphasize their ‘own’ issues: radical right and, respectively, environmental concerns. At the same time, both, but especially LMP struggled to establish organizational structures outside of Budapest, the country's political center.Footnote 3 Both parties originated from nonelectoral, civil society organizations. The network of these organizations spread outside the capital and allowed both parties to recruit elites at the local level.

For the establishment of Jobbik, the protest wave in 2006 was a pivotal moment of mobilization (Mikecz 2014; Pirro 2019). By this point, Jobbik already had a history of involvement in protest politics. As one of the textbook examples of a movement party (Pirro et al. 2021; Pirro and Gattinara 2018), Jobbik grew out of a protest movement that emerged after the 2002 victory of the MSZP against Fidesz. After the 2002 elections, right-wing protesters called for a recount, blocked the Erzsébet bridge in Budapest, and violently confronted the police. When Fidesz withdrew its support from the protest movement, one of the organizers, a right-wing youth initiative, decided to establish Jobbik. Although the 2002 demonstrations were much smaller compared to the events in 2006, their significance lies in leading to the establishment of Jobbik and providing its leaders with experience in organizing protests. Jobbik contested the 2006 election in coalition with another, previously influential radical right party, the Hungarian Truth and Justice (MIÉP), but gained only 2.2% of the vote. With the 2006 post-election protests, Jobbik substantially increased its popularity (Pirro 2019).

Around the time of the 2006 crisis, Jobbik also benefitted from the media coverage of a series of high-profile violent crimes with perpetrators of Roma origin. This increased the salience of its core issue, fighting against what the party labels ‘Roma-criminality.’ During this time, in 2007, Jobbik leader Gábor Vona registered the Hungarian Guard as the movement wing of the party. Taking up some elements of the Civic Circle model, Jobbik relied on the Hungarian Guard to develop its party organization (Krekó and Juhász 2017), particularly in areas outside of Budapest. The Guard was a paramilitary organization with the self-declared purpose of maintaining law and order. Although outlawed in 2009, several successor organizations were established in their place, many of which supported Jobbik. The movement and media hinterland (Jeskó et al. 2012) facilitated the entry of Jobbik to the European Parliament in 2009 with almost 15% of the vote share, a level of support similar to the results achieved in the 2010 (16.67%), 2014 (20.3%), and 2018 (19%) national elections.

LMP was the second new party formed in the same period which entered the Hungarian Parliament. Established in 2009, LMP grew out of the environmental civil society organization Protect the Future (Védegylet). The environmental movement has been one of the mobilizations with the most extended history and among the few civil society associations allowed to function during the state socialist regime (Císař 2022; Mikecz 2017). Protect the Future, in particular, has mobilized around opposition to large-scale infrastructural projects like the Gabčíkovo–Nagymaros Dams in the 1990s (peaking in 1998) or the NATO radar station on the Zengő mountain (peaking in 2004).

LMP was not the first attempt of the environmental movement to get involved in electoral politics. Due to the political constellation at the time, in 2005, Protect the Future-sponsored László Sólyom became the President of Hungary. However, having a president with progressive views on environmental issues did not lead to more permanent interactions between the environmental movement and the mainstream political parties. As a political party, LMP split its issue profile between focusing on the environmental issue and providing an anti-establishment critique of the post-1989 party system/elite. The party entered parliament in 2010 with a 7.5% vote share that stayed relatively stable in 2014 (5.34%) and in 2018 (7%), making LMP the most successful progressive green party in the CEE region (Kovarek and Littvay 2019).

The return of bipolar competition with Fidesz and its opposition

Under the pressure of a highly majoritarian electoral system, Jobbik and LMP faced increasing demands to ally with opposition forces—including MSZP—to oppose the governing Fidesz party. Jobbik resisted this pressure and began a strategy to moderate its stance by cutting ties to grassroots radical right groups (Pirro et al. 2021) and selective messaging in various media outlets (Borbáth and Gessler 2023). As a result, the party maneuvered itself into a more centrist position (Pirro et al. 2021, p. 24) but did not come close to Fidesz’ popularity. The strategy stopped after the 2018 elections, the last time the party ran independently. The party’s leader, Gábor Vona, resigned, and an influential group of politicians split to establish the more radical Our Homeland Movement. By 2022, what remained of Jobbik became part of the broad-tent alliance of a diverse set of parties opposing Fidesz, while the Our Homeland Movement entered parliament alone, with 5.7% of the vote (Caiani et al. 2024). Since the 2022 elections, support for Jobbik has gradually declined, with the party staying below or around the 5% threshold in the polls.

The post-2010 trajectory of LMP followed a highly similar pattern. Similarly to Jobbik, LMP resisted the pressure to ally with MSZP or other opposition parties against Fidesz. However, the question divided the party and led to the split of Dialogue—The Greens' Party (PM—A Zöldek Pártja) in 2013. Similar to Jobbik’s case, the party leader, András Schiffer, resigned from the party presidency (2016) and eventually left the party (2018). By the 2022 election, the independence strategy stopped, and LMP joined the broad-tent alliance of opposition parties against Fidesz. Since 2019, electoral support for the party has consistently stayed below the 5% threshold in public opinion polls.

Despite differences in their ideological background, Jobbik and LMP followed a remarkably similar electoral trajectory, characterized by strong alliances with nonelectoral civil society actors, initial electoral successes, followed by ideological dilution, electoral decline, organizational splits, and eventual cooperation with other opposition parties. Given their similarities, the previously discussed regional specificities, and the movement origins of both parties, we expect them to maintain their movement character over time. Since both parties aimed to introduce a new programmatic pole in party competition, we expect Jobbik and LMP to combine the mechanism of alliance building and issue ownership. Similarly, both parties faced coordination difficulties in organizing in the countryside, albeit LMP to a greater extent. We expect them to rely on protests to establish a presence in areas outside of Budapest.

Data and methods

To examine the involvement of political parties in the protest arena, we rely on protest event analysis (PEA). PEA is a widely used method in social movement research that allows for comparisons over time about the level and the composition of protest (Hutter 2014a). We combine two PEA datasets. First, we build on the work of Ondřej Císař and colleagues (Císař and Vráblíková, 2019), updated with the support of the ERC funded Political Conflict in Europe in the Shadow of the Great Recession research project at the European University Institute to cover the period up to 2015. The data are based on the coverage of the national news agency, MTI (Magyar Távirati Iroda). Following Císař, we define protest events as a gathering of at least three people who convene in a public space to assert claims that have a bearing on the interests of an institution/ collective actor. To compile the dataset, the electronic archive of MTI was searched for a series of 22 keywords referring to protest events.Footnote 4The news items covering a protest event corresponding to our definition were manually selected and hand-coded for the variables of interest. For the current analysis, we only focus on public demonstrations and exclude all other forms of protests.

Second, we complement these data with a second, original PEA dataset (Hungarian Protest Event Database—HPED) covering the 2017–2022 period. HPED relies on the official records of the police on registered demonstrations. The dataset has been collected by Susánszky (2024). In this dataset, protest events are defined as a “public gathering held with the participation of at least two persons for the purpose of expressing opinion in a public affair.” (Act LV of 2018 on the right of assembly, Section “Protesting parties in Hungary”). According to the law, protest events must be preregistered with the police at least 48 h before. For our analysis, we filtered out (1) the canceled demonstrations,Footnote 5(2) the incorrectly reported events (e.g., religious mass procession, private events), and finally, (3) the party campaign events (e.g., spreading party campaign materials (e.g., flyers, leaflets), table stands, and campaigning on the street). The dataset contains the issue/ claim, the organizer, and the place and time of the events.

The two datasets have different strengths and weaknesses. MTI is part of the public media, and its functioning is supervised by an Advisory Board, whose members are elected by parliament. Theoretically, MTI covers all newsworthy events, regardless of the political leaning of the organizers. Given its reach in the Hungarian press, a protest covered by any outlet—with the notable exception of social media—has likely been covered by MTI. However, the independence of MTI has been gradually compromised after 2010 (e.g., Bajomi-Lázár 2013: 83), introducing inconsistency in how the organization covers anti-government demonstrations. In contrast, police records capture all legal demonstrations but do not include nonpreregistered, illegal gatherings, even if these are newsworthy. In our view, in the context of the intense party/ state colonization of the media in Hungary, the combination of the two sources minimizes the bias that the methodological literature on PEA documents (Hutter 2014a). Notably, both datasets aim to cover the population of protest events in Hungary without sampling.

Significantly for our argument, the two datasets vary in their capacity to account for parties involvement in protest politics. With the MTI dataset, we are only able to capture protest sponsorship. Following Rucht (1998, p. 41), we broadly define sponsorship as (co-)organizing, participating in, and/or calling for participation in a protest event. With the HPED dataset, we capture event organizers as the ones who registered the event with the police. Accordingly, with the HPED, we can pin down party-organized protests, not only party-sponsored events. The advantage of identifying organizers comes with the disadvantage of not having information about potential co-organizers, as the police records a single organizing actor.Footnote 6 In contrast, the MTI data include up to five organizations, all of which have been explicitly mentioned to have organized and not only taken part in the event.

In the analysis, we focus on the 20 years spanning from the first radical right protest organized by Jobbik on the 20th of February, 2002, to the national elections on the 3rd of April, 2022. In terms of the issues being protested, we aggregate the two datasets according to the coding scheme provided by Císař and Vráblíková (2019) with the difference of including urban planning among environmental issues.Footnote 7 The combined data contain 2450 radical right, environmental, or Jobbik/ LMP organized demonstrations.

To operationalize our indicator of issue ownership, we adapt the measure proposed by Hobolt and De Vries (2015, pp. 1169–1170) to the protest arena. They suggest multiplying party-specific issue salience with the party position on the issue (centered on the party system mean). In the context of the protest arena, we adapt their measure by calculating the share of protest events sponsored on the radical right issue within all protests sponsored by Jobbik.Footnote 8 Similarly, we calculate the share of protest events with a progressive position on the environmental issue within all protests sponsored by LMP. This indicator accounts for party-specific salience (issue-specific relative share within all sponsored events) and position on the ‘owned’ issue (radical right, pro-environment). By relying on shares, this indicator allows us to compare the parties’ involvement across datasets.

To examine the extent to which parties rely on protest mobilization at the local level, we adapt our indicator of issue ownership. Namely, we calculate the share of protest events outside of Budapest within all protest events sponsored by the respective party. Although the measure equates events in all localities outside of Budapest, we are limited by our data in capturing region- or locality-specific party mobilization. Given the highly centralized character of Hungarian politics (Szabó, 1997), with approximately 60% of protests occurring in Budapest (Susánszky et al. 2019), we believe that this is an adequate measure to account for the theoretical mechanism we previously outlined.

To assess alliance formation, we follow two strategies. On the one hand, we account for the share of co-sponsored events in all protest events sponsored by the respective party. On the other hand, we rely on network analysis and examine the relative importance of the issue-specific mobilization network (Jobbik-radical right; LMP environment) and non-Budapest-based events in the network of events co-sponsored by the respective party.

In the network analysis, the nodes are the organizations Jobbik or LMP has co-organized events with, and the edges are the number of events co-organized between the party and another organization. We then calculate the share of nodes (organizations) and edges (events) that are associated with the subnetwork defined by its issue focus (radical right/environment) or location (Budapest/non-Budapest). The analysis allows us to examine how issue ownership and local presence interact with alliance building in the protest network of the respective party. Due to the previously discussed data constraints, we can only conduct the network analysis with the MTI dataset until the 2014 election.

First, we present the level of protest mobilization by political parties to assess whether the niche parties of Jobbik and LMP rely more on protest than their mainstream counterparts. Second, we present the prevalence of the three mechanisms in the protest mobilization strategy of the two parties over time. Third, we examine the interaction of the different mechanisms by analyzing the network of events co-organized by the two parties.

Empirical analysis

Protesting parties in Hungary

We start by assessing the presence of parties in the Hungarian protest arena. During our observation period, 23.7% of all protest events have been organized by one of the political parties.Footnote 9 To examine which parties mobilized in protest, we calculate the proportion of the number of events organized by a party from the total number of events organized by any party for each electoral period. Similarly to Hutter (2014b, p. 125), we compare this with the parties’ presence in the electoral arena, as captured by their vote share and the share of the media coverage their electoral campaign received. The latter is based on the Kriesi et al. dataset (2020), covering one left and one right-leaning national newspaper two months before national elections. The vote and the media share are calculated for the campaign at the end of the respective electoral period. Figure 1 presents the values of the two parties in the focus of our analysis, Jobbik and LMP, and for a comparative perspective, the values of the mainstream right (Fidesz) and left-wing (MSZP) parties.

Fig. 1
figure 1

Electoral and protest mobilization by the main parties over time (2002–2022). Note: The share of party protests is calculated from the number of all protests organized by parties. The media coverage reflects the average percentage of reporting received by the respective party’s campaign in two national newspapers based on the Kriesi et al. (2020) dataset. Vote shares of pre-electoral coalitions are indicated with dashed lines. They are Jobbik-MIÉP (2006), MSZP-Együtt-DK-PM-MLP (2014), MSZP-PM (2018), DK-Jobbik-Momentum-MSZP-LMP-PM (2022)

Full size image

The figure highlights two critical dynamics. First, both LMP and Jobbik are consistently more overrepresented in protest relative to their presence in the electoral arena, as captured by their vote share and media coverage. The difference is especially striking in the case of Jobbik, which was associated with more than 40% of party-sponsored protests between 2010 and 2014. In the case of LMP, the difference is smaller but still significant. 2010 represents an exception since LMP was only established in 2009, shortly before the election. Over time, the presence of both parties in protest declined. Still, even in the latest period, between 2018 and 2022, both continued to mobilize in protest to a greater extent than their electoral popularity would indicate.Footnote 10

Second, the opposition-government dynamic is critical to interpreting the protest presence of mainstream parties. During the first period between 2002 and 2006 under the MSZP government, both the governing MSZP and Fidesz from the opposition organized fewer protests than their presence in the electoral arena would suggest. The dynamic under the second MSZP government changed to some extent. In line with the discussion of the events following the 2006 election, we observe Fidesz with a greater protest presence than its presence in the media or its subsequent 2010 vote share leads us to expect. In contrast, the presence of MSZP in protest stays stable, with around 18% of party-sponsored protest events. Since 2010, under the Fidesz government, MSZP has been more active in protest than Fidesz that was mainly associated with a smaller number of high-profile events (Susánszky et al. 2016). Overall, Fig. 1 confirms our expectation that Jobbik and LMP are the two parties proportionally most overrepresented in protest.

Mechanisms of protest mobilization

Next, we examine the prevalence of the three mechanisms previously distinguished in the protest presence of Jobbik and LMP. Figure 2 shows the dynamic of issue ownership, alliance building, and local presence outside of Budapest. All values are shown in percentage points as shares within the overall protest network of the respective party (the underlying values are included in the appendix).

Fig. 2
figure 2

Issue ownership, alliance building, and local presence in protests by Jobbik and LMP (2002–2022). Note: The first pair of panels shows the share of protest events organized on the issues ‘owned’ by Jobbik (radical right) and LMP (environment) from all protest events sponsored by the two parties. The second pair of panels shows the share of protest events co-organized with other actors by Jobbik and, respectively, by LMP. The third pair of panels shows the share of protest events organized outside of Budapest by Jobbik and, respectively, by LMP

Full size image

In the case of Jobbik, the figure documents the transformation of the party’s protest presence over time. In the 2002–2006 period, before the party entered parliament, Jobbik was almost exclusively associated with co-organized, radical right protest events (76%). We observe the fading of mobilization on the ‘owned’ radical right issue in the protest arena, with a 3% share by the 2018–2022 period. This trend is accompanied by the decreasing share of co-organized protest events, which declined to less than 50% in the 2010–2014 period.Footnote 11 Similarly to previous research (Borbáth and Gessler 2023; Pirro et al. 2021), we interpret these findings as part of Jobbik’s moderation strategy.

At the same time, we observe the increasing importance of protest mobilization outside of Budapest. While between 2002 and 2006, only 5% of Jobbik protests took place outside of Budapest, by the 2018–2022 period, their share reached 82%. We take this as a sign that, over time, the protest fills a different purpose for Jobbik. While in the first period, it mainly serves as an additional arena to mobilize on the issues owned by the party and enter into alliances with like-minded organizations, as the party moderates its image, mobilizing in protest becomes a tool for establishing a presence outside of Budapest. This shift goes hand in hand with a declining trend of Jobbik’s involvement in the protest arena (Fig. 1).

In the case of LMP, the share of environmental events in protest mobilization stays marginal. Only in the latest period does the share of environmental events reaches 10%. Still, it never gets close to the high values observed in the case of radical right mobilization by Jobbik. Protest by LMP mainly serves an alliance-building function, particularly after the party entered parliament when the share of co-organized events reached 77% of all events associated with LMP. Similarly to Jobbik, in the latest period, we observe the over time increase in mobilization outside of Budapest, increasing from a 10% share in 2010–2014 to an 82% share in 2018–2022. As overall protest by LMP declines (Fig. 1), the party mainly relies on protest events to establish a presence at the local level outside of Budapest.

Importantly for our argument, the over time shifts we document are not sudden ruptures in the time series of Jobbik/LMP organized protest that follows the party’s entry into parliament. Instead, they reflect incremental transformations that take place over twenty years. In fact, protests by both parties peaked in 2010–2014, during their first term in parliament. Next, we zoom in on the mobilization network in the 2002–2014 period to illustrate the interaction of alliance building with issue ownership and local presence.

Networks of protests

For the network analysis, we only included co-sponsored events until the 2014 parliamentary elections, as recorded by the MTI dataset.Footnote 12 We first constructed the network of all co-sponsored events by Jobbik and, respectively, by LMP. We then zoom in on the subnetworks defined by the issue focus of the co-sponsored events (radical right/environment) or their location (Budapest/non-Budapest). Based on these, we calculate the number of organizations (vertices) associated with mobilization on the issue owned by the party or mobilization outside of Budapest. Similarly, we calculate the share of events related to issue-specific mobilization or protests outside Budapest. The calculation quantifies the interaction between alliance building and issue ownership/local presence. Table 1 displays the results.

Table 1 Co-sponsorship networks of Jobbik and LMP (2002–2014)
Full size table

As Table 1 shows, the alliance network of the two parties is similar in size, with the Jobbik network having a somewhat lower number of organizations (59) than the LMP network (62). Half of the organizers (48%) within the Jobbik network (see Table 3 in the appendix) are radical right organizations such as paramilitary groups (Hungarian Guard, Hungarian National Guard, Hungarian Self-Defense Movement), revisionist, nostalgic groups (Trianon Society, Miklós Horthy Historical Society), radical right movements (64 Counties Youth Movement), and Neo-Nazi organization that identify with and spread national-socialist ideas (Conscience’88). In contrast, the proportion of green actors within the LMP network (see Table 4 in the appendix) is considerably more sparse (12%). One-quarter of actors in this network (28%) focus on democracy and human rights issues, such as freedom of speech (One million people for the freedom of the press), women’s rights (NANE), and the rights of homeless people (The City is For All).

Both parties occasionally organize protest events with other political parties. In the case of LMP, the share of events co-organized with other parties reaches 48% of all events in LMP’s co-sponsorship network, whereas in the case of Jobbik, their share only gets 9%. In the case of Jobbik, some of these events took place outside of Budapest and on radical right issues. In contrast, in the case of LMP, almost all took place on issues other than the environment in Budapest (see Table 2 in the appendix).

The two parties strongly differ in how they use alliances to mobilize on their key issues outside of Budapest. In the case of Jobbik, the radical right subnetwork encompasses about half of the organizations the party co-organizes events with and 60% of the connections. In the case of LMP, the share of organizations involved in environmental protest mobilization is about 13%, accounting for only 8% of the edges. Similarly, in the case of Jobbik, nearly half of the organizations the party co-organizes events with are involved in the protest network outside of Budapest and account for almost two-thirds of all edges. In the case of LMP, only 8% of organizations are associated with mobilization outside of Budapest, which only accounts for 4% of all connections. The result suggests that in the case of Jobbik, co-organizing protest events are part of a broader mobilization strategy on the party’s key issue and in places outside of the capital. In the case of LMP, co-organized events tend to focus on issues other than the environment (e.g., democracy) and mainly occur in Budapest.

Conclusion

We started this paper by arguing that the blurring boundaries between the electoral and the protest arena provide an important lens to examine the dynamic of party system transformations. In post-2006 Hungary, this is most certainly the case, with protest mobilization playing a pivotal role in the development of the two most important new challenger parties, Jobbik and LMP. Our results show that LMP and Jobbik have a considerable presence in the protest arena. Following in their path, other new parties in the political center (TISZA), the left (e.g., Momentum), and the radical right (Our Homeland Movement) built their political strategy on protest mobilization and benefitted from strong ties to nonelectoral organizations (e.g., Caiani et al. 2024).

We introduced a distinction between the mechanism of issue ownership, alliance building, local presence, and their combination to explain when and why parties mobilize in protest. As our results show, in the first period, Jobbik successfully combined both mechanisms: the party benefitted from protest mobilization by allying itself with other actors who have mobilized on radical right issues. At the same time, Jobbik organized most of its protest events on its core, radical right issues. In contrast, LMP enters into alliances with actors (often parties) not associated with the environmental issue and relies on protest mobilization to organize events on democracy-related and economic issues. Over time, protest mobilization by both parties declined and transformed: it became increasingly focused on mobilizing in places other than Budapest to establish a presence at the local level.

Our results address the debate on the transitional character of movement parties. The empirical analysis demonstrates that Jobbik and LMP continued to mobilize even after they entered parliament. Given that LMP was established shortly before the 2009 election, we do not find protest mobilization by the party before it entered parliament, while both parties continued to be present in the protest arena after entering parliament. The main difference over time is the transformation of protest mobilization by both parties to encompass a broader range of localities.

Naturally, our conclusions are limited by focusing only on press and police records of protest behavior. We emphasize the agency of political parties, but differences might also arise from the diverging character of radical right and environmental movements. While the radical right movement has successfully been channeled hierarchically by Jobbik (Jeskó et al. 2012), the environmental movement continued to maintain its autonomy vis-à-vis LMP. One natural extension of our research would be to conduct interviews with the representatives of both parties and movements to understand the limits of party agency in protest mobilization.

One important question refers to how these results travel to other contexts. In the theoretical framework, we highlighted the contextual conditions that distinguish Central and Eastern Europe, specifically Hungary. In this context, party–protest interactions might be more influential in explaining electoral breakthroughs and party system transformation. Still, new parties regularly enter on the back of movement mobilization in advanced democracies (e.g., Weisskircher et al. 2022), indicating the more general importance of interactions between the electoral and protest arenas. Therefore, we hope the conceptual distinction between the three mechanisms we introduce contributes to the broader literature on party–protest interactions.