Electoral campaign effects: An aggregate analysis of electoral issue competition

Abstract

Elections are not only verdicts on incumbents but also contests over the issues that parties emphasize. This paper asks whether parties’ issue-specific campaign strategies influence their electoral success, and how parties decide on those strategies. We leverage a new dataset on parties’ electoral campaign efforts in 15 European countries to examine three dimensions of campaign strategy: the salience of issues, the position they take on those issues, and the extremity or moderation of those positions. As for the parties’ strategic choices, the results confirm that European parties organize their issue-specific campaigns largely in line with their ideological positions and their status as challengers or mainstream parties. In response to the electoral effect of their campaign strategies, the results confirm the received wisdom that the parties’ issue-specific campaigns have only a limited effect on the electoral outcome. However, this general result has to be nuanced by party family: challengers and mainstream party families have benefited to varying degrees from their respective strategies. The effects of campaigns on electoral success are the largest for the radical right, which has benefited from putting the emphasis on cultural issues and taking clear-cut or even extreme positions on economic issues. Our findings shed new light on party responsiveness and the limits of political persuasion, showing when campaign appeals can sway voters and when structural factors trump campaign effects.

Introduction

Democratic elections are the primary mechanism through which citizens hold leaders accountable and ensure their preferences are represented in government. At the heart of this process is party competition: political parties vie not only for votes but also to shape the agenda of public debate. As Schattschneider (1975) famously observed, “the definition of the alternatives is the supreme instrument of power”, in other words, those who can set which issues elections are fought over wield great influence over outcomes. Whether voters choose to replace incumbents depends on the issues that dominate the campaign. Campaigns therefore play a crucial role in linking citizens’ concerns to electoral choices, providing an arena in which parties highlight certain problems and priorities and cue voters to what is at stake. One of the central functions of parties in democracies is indeed to serve as intermediaries between citizens and decision-makers (Klüver and Spoon, 2016), translating societal demands into policy agendas during elections. By selectively emphasizing favorable issues parties seek to persuade voters that these are the questions that matter. The way parties compete over issues in a campaign can directly impact democratic accountability and representation by shaping the alternatives voters perceive and the basis on which governments are elected. In short, issue competition in campaigns is not just about policy differences; it fundamentally influences who gains power and how responsive that power will be to public preferences.

In multi-party democracies, parties do not only compete by taking adversarial positions on ideological dimensions; they also compete by elevating issues that play to their strengths while downplaying those that do not (Budge and Farlie, 1983; Petrocik, 1996). Recent research formalizes this behavior as the logic of “issue yield” (De Sio and Weber, 2020), positing that parties tailor their campaign agendas to focus on issues that offer the greatest electoral return given the party’s profile. In practice, parties strategically avoid engaging with issues that are not electorally “profitable,” and instead concentrate on topics where they enjoy a credibility or salience advantage. When established parties neglect issues that large segments of the public care deeply about, challenger parties often seize the opportunity to politicize those neglected issues. Indeed, the politicization of previously sidelined concerns tends to work in favor of challenger and niche parties (De Vries and Hobolt, 2020). Recent decades have borne this out, with new parties gaining traction by campaigning on immigration, European integration, or environmental issues that mainstream parties were slow to address. Established parties, for their part, face a strategic dilemma: if they shift their stance to recapture voters on an emerging issue, they risk losing credibility or alienating their base. Empirical evidence shows that such repositioning often fails to yield the intended gains. For example, social democratic parties moving toward more restrictive immigration positions have not stemmed their vote losses (Abou-Chadi and Wagner, 2020). Thus, which issues parties emphasize, and how they position themselves on those issues, can have profound electoral consequences.

If governing parties fail to respond to voters’ foremost concerns, they are likely to be punished at the polls or eclipsed by new competitors (Klüver and Spoon, 2016). Conversely, when parties do engage with the public’s issue priorities, they fulfill their representative function and may blunt the appeal of anti-establishment challengers. Programmatic, issue-focused campaigning is normatively seen as a sign of healthy democracy, in contrast to non-programmatic tactics like clientelism, because it gives voters a real policy choice. As Stokes et al. (2013) argue, when citizens shift from trading votes for patronage to voting based on policy appeals, it marks a move “from a less to a more democratic polity”. In essence, understanding how parties compete over issues during campaigns, and with what effects, is fundamental to the study of democratic representation and accountability.

For all the theoretical importance of campaign issue competition, empirical research has not fully answered when and how campaigns influence electoral outcomes. In an article reviewing media and campaign effects on vote choice in Europe, Boomgaarden and Schmitt-Beck et al. (2016) argue that there is little convergence regarding approaches to this question and related research findings. Based on the literature reviewed, the authors conclude that campaigns matter for various voting qualities. They add, however, that we learn very little, if anything at all, about what aspects of campaigns would drive the effects (p. 139). Moreover, the dimension of parties’ electioneering that is least thoroughly studied concerning its relevance for electoral choices is the content of their campaigns (p. 142). This assessment of the literature may seem surprising. However, it makes a lot of sense if we consider that the large literature, which is mainly based on party manifesto data or expert panels, has focused on parties’ long-term issue-specific strategies rather than on short-term aggregate campaign effects. Thus, party manifestos are a poor predictor of issue salience during electoral campaigns (Helbling and Tresch, 2011: 181) and the relevant literature based on them does not adequately address how aggregate issue emphasis and positioning affect their electoral performances. In other words, existing literature has had difficulty pinning down the effects of campaign content, as opposed to long-standing party positions, on election outcomes. Consequently, we still know little about how campaign content affects a party’s performance at the polls. Addressing this open question is crucial for understanding whether short-term campaign strategies can sway voters, especially in multiparty systems.

The American literature on campaigns has, above all, been preoccupied with the minimal effect of the parties’ communication and mobilization strategies on vote choice in U.S. Presidential elections (e.g. Brady et al., 2006; Broockman and Kalla, 2023; Gelman and King, 1993; Kalla and Broockman, 2018; Kinder, 1998, 2003; Nickerson and Rogers, 2020). On the part of the voters, partisan resistance, indifference, and neutralizing effects of competing campaigns have been invoked to explain this minimal effect. Additional supply-side determinants have also been adduced. Thus, Lichtman (2020) argues that the outcome of a presidential election depends almost entirely on the performance of the incumbent government party. In other words, governing is much more important than campaigning, and the voters’ verdict rests on performance evaluations. Similarly, Zaller (1998) used the example of Clinton’s approval ratings despite the media hype around the Lewinsky scandal to show that political substance, i.e., the state of the economy, the absence of war, and political moderation, matter for the public’s assessment of the incumbents. In the same vein, Bremer (2024) has shown that policy feedback has influenced the vote for social democrats in Europe in the aftermath of the recent crises: they have been punished for their fiscal consolidation policies. There is no doubt that incumbency effects constrain the effect of the campaign: the incumbency advantage is well known (e.g., Gelman and King, 1990) – incumbents have a high awareness and name recall, they have a fundraising advantage, and they have the opportunity to use the government’s agenda for their electoral purposes (Perron, 2024). Contrastingly, incumbents also suffer from the “costs of ruling” (Nannestad and Paldam, 2002). In short, structural factors and voter predispositions place bounds on campaign influence, and this has led some observers to conclude that campaigns only marginally alter election outcomes. While much of this evidence comes from two-party presidential races, similar structural constraints have been documented in European parliamentary systems (Adams et al., 2011).

Against this backdrop, we focus on the content of the campaign by studying parties’ issue-specific strategies on broad aggregate issue categories. We try to answer two questions: whether European parties’ issue-specific campaign strategies affect their electoral success and how the parties choose their issue-specific campaign strategies in the first place. By addressing both the consequences and the determinants of campaign strategies, our study builds on and goes beyond existing campaign-effects literature, which has tended to examine either voter responses or party behavior in isolation. We argue that issue-specific campaign strategies are affected and decided by a host of factors and contextual constraints, such as (a) the heritage of the country-specific electoral history, (b) the contemporary electoral configuration (incumbency effects and entry effects of new challengers), and (c) the vicissitudes of problem pressures and public events. The ups and downs of problem pressure are independent of the electoral cycle, and public attention reacts to it. These exogenous shocks constrain election campaigns, and parties must deal with them (Dennison and Kriesi, 2023; Michel, 2026). Within the constraints imposed by these three factors, the parties develop their electoral campaigns. Grossman and Guinaudeau (2021) have argued that they are like “snakes in a tunnel.” They must develop their strategies within the context conditions from which they cannot choose. In terms of strategic choices, they have different options: they can emphasize either actors or issues or both. In our paper, we focus on the effects of their issue-specific campaigns. We shall study three issue-specific parameters of partisan campaign strategies: salience, position, and moderation. Salience refers to the attention a party pays to a given issue. Position refers to the direction of the parties’ positioning and corresponds to a strategy that takes a clear issue-specific position. Moderation, as opposed to extreme positioning, is a strategy that seeks an issue-specific middle ground when addressing salient issues. By analyzing these dimensions of campaign content, we capture both what parties talk about and how they talk about it.

We study issue-specific electoral campaigns and their effects in Europe at the aggregate level. To do so, we rely on a newly available data set on European electoral campaigns, which we will describe in more detail in the design section. Our data covers 214 parties in 110 elections from 15 European countries during the period 1988 to 2023. The countries covered include seven northwestern European (Austria, France, Germany, Ireland, the Netherlands, Switzerland, and the UK), four southern European (Greece, Italy, Portugal, and Spain), and four eastern European countries (Hungary, Latvia, Poland, and Romania). This large-N, cross-national scope provides an unusually broad empirical basis to generalize about campaign dynamics. By examining campaign content and its impact across many different party systems and historical contexts, we are able to move beyond the limitations of single-country or single-election studies.

Our analysis makes three contributions to understanding democratic competition. First, we demonstrate that European parties’ campaign strategies remain strongly structured by ideological positions and party family, suggesting that democratic competition maintains predictable patterns despite systemic changes. Parties largely campaign on issues they “own” and take positions consistent with their historical identities, confirming that voters face meaningful choices between distinct alternatives. Second, we provide nuanced evidence about campaign effects on electoral outcomes. While overall effects are limited, consistent with theories emphasizing structural constraints on party agency, we find important variation by party family. Challenger parties, particularly the radical right, can still shape their electoral fortunes through strategic choices, while mainstream parties face stronger constraints. This shows that competitive dynamics can operate differently for established versus insurgent forces. Third, we advance methodological understanding by distinguishing between long-term positioning and short-term campaign adaptation. This temporal distinction reveals that parties face different constraints when trying to maintain consistent positions versus responding to immediate contexts. Short-term adaptability appears limited, but long-term strategic positioning retains electoral relevance for some party families. These findings speak to broader debates about party competition, democratic responsiveness, and representation quality in contemporary Europe. They suggest that while structural forces increasingly constrain electoral outcomes, strategic agency remains possible within specific parameters. Understanding when and how campaign strategies matter is essential not just for electoral analysis but for evaluating the health of democratic competition itself.

We begin by introducing our theoretical framework, followed by the presentation of the design of our study. After the design section, we present our results and a brief final summary discussion.

Theoretical framework

Three structuring parameters

As a result of the country-specific electoral history, the party configuration varies from country to country: the various party families are of very different strength and composition. Bartolini (2000) has shown this for the European left, the size and configuration of which depend on the country’s historical cleavage structures. This is an obvious point, but it bears repeating. In our study, we shall distinguish between six party families: the radical left, Greens, the mainstream left (social democrats), two families of the mainstream right – liberals (social and conservative liberals) and conservatives (conservatives and Christian democrats), and the radical right. We define challenger parties by their party family. We consider three families as challenger parties: the radical left, the Greens, and the radical right. Even if some of these parties have already governed ¹ , overall, they are still more marginal than the three mainstream party families. Most party systems covered in this study are dominated by the mainstream right, whose composition varies greatly from country to country. Overall, the mainstream right gets about half of the vote in the countries covered (see Table A1 in the Appendix), the social democrats about one-fourth, and the rest goes to the three types of challenger parties, of which the Greens are, overall, the weakest. The configuration of the challenger parties also varies by country: there is no radical left worth speaking of in Austria, Switzerland, the UK, Hungary, Latvia, and Poland, no Green party in Latvia, Poland, Romania, Italy, and Spain, and no radical right party in Ireland and (until recently) Portugal. The radical left is particularly strong in southern Europe, while the Greens and the radical right are strongest in northwestern Europe. This historical legacy and spatial distribution have created some expectations for parties that constrain their positioning on a variety of issues, as they cannot credibly oscillate on issues they are traditionally associated with or risk their electoral base abandoning them.

In addition to the historical configuration of party families, the contemporary electoral space in a given country is also characterized by the incumbent-opposition configuration and the entry of new challengers. In the European context, the incumbent-opposition configuration is complicated because most governments are coalition governments, with a dominant party that supplies the Prime Minister (PM) and coalition partners who take a minority position. In such a situation, the incumbent advantage and the costs of ruling may be asymmetrically distributed. The PM enjoys the greatest recognition advantage and, most likely, financial advantages. At the same time, the minority partners are likely to pay higher costs of ruling because they are often forced to comply with the majority partner’s position at the expense of their constituency. We need to control these configurations, acknowledging that the issue space is, to a large degree, shaped by the policies and preferences of the incumbent parties.

The issue-specific electoral campaign of a given party takes place against the background of these country-specific configurations. In addition, the issue-specific electoral campaign is also constrained by the vicissitudes of problem pressures or public events. To some extent, the electoral agenda is imposed on the parties by the public agenda and the campaign strategies of the other parties – i.e., by the “tunnel” (Grossman and Guinaudeau, 2021), constraining their strategic choices. On top of that, political heritage also reinforces the tunnel’s walls, as parties cannot credibly shift positions erratically each election cycle towards more popular ones without a significant cost to their credibility. To gain electoral advantage, parties need to “ride the waves,” i.e., emphasize issues that are salient in the media and the public (Damore, 2004, 2005; Green and Hobolt, 2008; Sides, 2006; Wagner and Meyer, 2014). In other words, in terms of salience, they must follow the public (Klüver and Sagarzazu, 2016; Klüver and Spoon, 2016; Stimson, 2004). The public issue salience, in turn, is, at least partly, determined by exogenous events and trends independent of political and media cues (Klüver and Sagarzazu, 2016; McAllister and bin Oslan, 2021; Singer, 2011; Dennison, 2020; Gilardi et al., 2022; Seeberg, 2017, 2021). Overall, agenda-setting has been argued to be largely exogenous to both parties and media (Gilardi et al., 2022: 190; Dennison and Kriesi, 2023; Dennison, 2019).

Parties’ issue-specific electoral strategies

Parties have two basic issue-specific parameters they can strategically modify – the salience of issues in their electoral campaigning and their positioning on these issues. They combine these two parameters differently, depending on the issue in question. We shall consider three issue-specific parameters among which parties can choose – issue-specific salience, and two aspects of positioning – issue-specific position (or direction) and moderation (or extreme positioning).

First, parties can pay attention to or try to avoid an issue. Within the constraints set by the public agenda, they attempt to choose issues that they “own,” i.e. for which they have a reputation for competence, with which they are habitually associated and which provide them with a natural advantage over their competitors (Petrocik, 1996; Brazeal and Benoit, 2008; Budge and Farlie, 1983; for an overview: Walgrave et al., 2015). The opposite side of the coin is that they try to avoid issues they do not “own,” such as issues where their preferred position (Bräuninger and Giger, 2018; Campbell, 1983; Milita et al., 2014) is far removed from their constituency or where their constituency’s preferences are heterogeneous (Rovny, 2013).

There are, of course, many issues that can be potentially put into evidence or avoided in an electoral campaign. To simplify and focus on the essential aspect of electoral contests, we shall distinguish between two types of issues in this study: economic and “new cultural issues.” Economic issues broadly fall into two categories: social policy issues (health care, pensions, housing, education, unemployment insurance, etc.) and economic policy issues (labor market, fiscal policy, monetary policy, privatization, competition policy, agriculture, bailouts, etc.). New cultural issues include four broad categories: immigration (and nationalism), European integration, environmental issues, and cultural liberalism (gender, lgbtq rights, minority rights, international solidarity, etc.). Together, these two types of issues cover roughly two-thirds of all issues ² . We focus on these two sets because they are generally relevant in all countries and typically the most contested, and because they define the two dimensions of the political space in European party systems (Hutter and Kriesi, 2019). Economic issues are “owned” by the mainstream parties of the left and right, as well as by the radical left among the challenger parties; new cultural issues, by contrast, are “owned” by the Greens and the radical right (see, e.g., Budge, 2015: 769; Abou-Chadi, 2016: 421). Issue ownership has been shown to be rather stable over time (Seeberg, 2017). In other words, in terms of salience attributed to the various issues, we expect the following:

H1

In electoral campaigns, mainstream and radical left parties pay more attention to economic issues, while the Greens and radical right parties emphasize new cultural issues.

Position and moderation (extreme position) are the two aspects of positioning that we address in this study. Position refers to the extent to which parties take distinct issue-specific positions: parties not only emphasize issues but also politicize them in line with their overall ideological orientation, which is similar across countries for parties from a given party family. In the two-dimensional space of the European party systems (see Borbáth, 2026; Hutter and Kriesi, 2019, 2026, Kriesi et al., 2006, 2008), left-of-center parties tend to assume pro-state positions on the economic dimension and cosmopolitan and culturally liberal positions on the new cultural dimension (from now on just «cultural dimension»), while the center-right parties tend to assume pro-market positions on the economic dimension and nationalist and traditional positions on the cultural dimension. However, the two challenger parties on the left are distinguished because the radical left is expected to be positioned more to the left on the economic dimension. In contrast, the Greens are expected to be positioned more to the left on the cultural dimension. Depending on the country, the parties’ positioning on the two dimensions will likely be more or less aligned (Hutter and Kriesi, 2019).

In addition, we expect mainstream parties from both sides of the political spectrum to adopt more moderate positions than challenger parties. Under Downsian assumptions, mainstream parties are incentivized to moderate their positions and converge to the median voter (Downs, 1957; Grofman, 2004). Kirchheimer (1966) famously put into evidence the “catch-all” strategy of the so-called “people’s parties,” which attempt to appeal to very diverse constituencies by moderating their positions. This does not apply to the same extent to challenger parties, which appeal to more circumscribed constituencies (Abou-Chadi and Orlowski, 2016; Spoon, 2009). This leads us to the following summary hypothesis:

H2a

In electoral campaigns, parties take more extreme positions on the issues they own.

H2b

Mainstream parties take more moderate positions than challenger parties.

Alternatively, it is possible that mainstream parties attempt to undermine the success of their challengers by adopting an accommodating strategy, i.e., by moving their position towards the position of the challenger party on the same side of the political spectrum or by relying on adversarial strategies, i.e., by deliberately choosing a position opposite to that of the challenger party on the opposite side of the political spectrum (Meguid, 2005). In line with this argument, several studies confirm that challenger parties induce mainstream parties to reposition themselves accordingly (Abou-Chadi, 2016; Alonso and Fonseca, 2012; Dalton, 2009; Green-Pedersen and Krogstrup, 2008; Meguid, 2005; Pardos-Prado, 2015; Pardos-Prado et al., 2014; Van Spanje, 2010). Moving their position in the direction of the challenger party on their side of the political spectrum or away from the challenger party on the other side of the spectrum means that mainstream parties are radicalizing their position, even if it is likely that they remain more moderate than the challenger party, given that they have to appeal to broader constituencies.

H3

When mainstream parties are facing a successful challenger party in electoral campaigns, they adopt more extreme positions on the dimension privileged by the challenger party.

Finally, parties’ strategies may change over time due to their changing character, changing constituencies, intensified competition in the party system, or specific events. For example, Kitschelt and McGann (1995) formulated a “winning formula” for the radical right, which ought to combine neo-liberal economic positions (to attract the old middle class) with nationalist cultural positions (to attract the working class). Arguably, the neo-liberal element in this combination has become less important for the radical right, which has been increasingly focusing on nationalist positions on the cultural dimension and adopting welfare chauvinist or workfare-ist positions in socio-economic terms (Careja and Harris, 2022). As for highly publicized events, we may think of events such as economic crises (Marsh and Mikhaylov, 2012; Sanders, 1999) or other crises, such as the Covid-19 pandemic (Fernandez-Navia et al., 2021) or the refugee crisis (Kriesi et al., 2024), scandals (von Sikorski et al., 2020) or terrorist attacks (Getmansky and Zeitzoff, 2014; Montalvo, 2012). In order to take the changing character of issue-specific strategies and their differential effect into account, we shall attempt to distinguish between long-term and short-term strategies in our empirical design.

The impact of parties’ issue-specific strategies on their electoral success

The same factors that impose constraints on the parties’ electoral campaigns also constrain the impact on the electoral outcome. On the one hand, country-specific historical legacies and long-term structural changes in society circumscribe the scope of electoral outcomes. For example, deindustrialization, technological change, and educational expansion reduced the size of the working class, the traditional constituency of the social-democratic parties, and facilitated a process of electoral dealignment. As a result of these long-term trends, across Europe, for example, social democratic parties experienced declining vote shares and a changing character of their constituencies (e.g., Bremer and Rennwald, 2022; Gingrich and Häusermann, 2015). On the other hand, the impact of electoral campaigns on electoral outcomes is also constrained by short-term problem pressure of contemporary crises and public events, which set the public issue agenda independently of the parties’ strategic choices. The parties’ attempts to adapt to these short-term constraints may not be as successful as anticipated.

Given that the ideological distance between voters and their preferred party on specific issues is a strong determinant of electoral behavior (e.g., Ansolabehere et al., 2008; Bélanger and Meguid, 2008; van der Brug, 2004), we may expect that the parties’ issue-specific electoral campaigns have an effect on the voters’ electoral choices. However, in line with American no-results, comparative analyses find that the effects of parties’ electoral campaigns in Europe are also generally weak, inconsistent, and depend on various context conditions. Thus, Adams et al. (2011) found little to no reaction from voters to shifts in party policy statements as reported in party manifestos. De Vries and Hobolt (2020: 197) observe that issue entrepreneurship, i.e., essentially the emphasis put on issues parties own, enhances the electoral success of challenger parties. However, the effect is weak and barely significant. There are signs that accommodation pays. Thus, Spoon and Klüver (2020) report that accommodating shifts in an anti-immigration direction provide electoral benefits for mainstream left parties. Abou-Chadi and Wagner (2020) show that mainstream left parties also can increase their vote share by shifting towards more investment-oriented policies – an accommodating policy towards the new left, but only when they also take liberal positions on second-dimension issues and when unions are limited in their capacity to mobilize against those shifts. As for moderation, Polk and Karreth (2024) find that moderation in the previous electoral cycle on economic issues decreases the propensity of social democratic voters to stay with the party in subsequent elections, while moderation on the cultural issues also drives voters away to minor left parties but seems to penalize social democrats less down the road. Kitschelt and Rehm (2024) conclude that the contest in the center of the space remains critical for the social democrats. They find, among other things, that social democratic parties do best if they take a position slightly left of the center on economic and cultural issues and a dead-center position on immigration. They also show that the size of the social democratic party and the strategies of competing parties play a role in this respect.

Kitschelt and Rehm (2024) caution that their findings rest on a still precarious empirical foundation. They stress that the small number of cases makes it difficult to explore the many contingencies upon which the electoral payoff of party strategies may depend. This is another way of saying that it is very hard to generalize in this domain. Bremer (2024: 362) also cautions that his results about the electoral consequences of centrist policies are driven by patterns that existed when party systems were more stable and less volatile than they are today. However, as we shorten the time frame to more recent periods, we reach the limits of quantitative analysis.

Against the background of the empirical record, we do not expect issue-specific electoral campaigns to have a great impact on the parties’ electoral success. But, to the extent that they do have an effect, we assume that it is the parties’ preferred strategies that are most likely to be electorally successful:

H4

The parties’ electoral success is increased by:

a: the salience parties attribute to issues they own.

b: the distinct positioning in line with the parties’ ideological orientation

c: the moderation of positions on issues in general for mainstream parties and the extreme positioning on issues for challenger parties.

Design

For this study of the determinants of the parties’ issue-specific strategies and their electoral success as a function of these strategies, we mainly use a combination of two datasets: the CSA data set for the parties’ strategies, which currently covers 110 elections in 15 countries ³ (see Wang et al., 2026), and the Parlgov datasets for the parties’ electoral success, incumbency, and new entry. The countries covered are the fifteen countries already mentioned. For the joint analysis of the two data sets, we needed to match the information from the two for each election. The CSA data goes back to the 1970s, but only for six countries, so we only include elections from the 1990s onward. Moreover, we only include parties (or electoral coalitions) that have obtained at least one percent of the votes in one election ⁴ . This leaves us with 214 parties from 110 parliamentary elections in 15 countries. Our first election is the French election in 1988, and the last one was the Greek election in the summer of 2023. Table A2 in the Appendix provides the list of parties included in the analysis, ordered by country and party family.

Our analysis has two types of dependent variables – the party strategies and their electoral success. For electoral success, we use the change in a party’s vote share in a given election. For the calculation of the vote share, we classify parties into six party families, based on the Parlgov database (Döring and Manow, 2024): ⁵ radical left, Greens, social-democrats, liberals, conservatives/Christian-democrats (from now on called “conservatives”), radical right. As already pointed out, three of these party families – radical left, Greens, and radical right – we consider as challenger parties, while the remaining three – social-democrats, liberals and conservatives/Christian democrats – are considered as mainstream parties. Note that the composition of European party systems varies considerably across countries. Thus, the only party family that participates in all elections we use for this study is the social-democratic family. The conservatives are missing in only one election, but the other families are absent in more elections, varying from 30 (liberals) to 58 (Greens). Our indicator for the change in vote share is simply the difference between the party’s current vote share and its vote share in the previous election ⁶ .

The key variables consist of the issue-specific electoral campaign characteristics, which we treat as both dependent (in the first step) and independent (for determining vote shares) variables. We retrieve them from the CSA dataset (Kriesi et al., 2020). Researchers usually rely on Comparative Manifesto Data or Chapel Hill expert data to characterize the parties’ issue-specific electoral supply. These data, however, have the drawback that they measure the parties' long-term supply and do not consider the short-term content of the election campaigns. As we have argued in the theoretical section, during election campaigns, the salience of a particular issue for a given party is likely to be influenced by short-term context factors, including the agenda-setting strategies of other parties, the dynamics of the public debates or unexpected events (e.g., terrorist attacks, natural catastrophes, political scandals). Thus, Helbling and Tresch (2011) show that the salience measures of expert and CMP data hardly correlate with the salience data collected during election campaigns at all. Gessler and Hutter (2026) confirm these differences. This is why we operationalize the campaign characteristics using the CSA dataset, which is based on a relational content analysis of articles from two newspapers per country during election campaigns (Kriesi et al., 2020). Nowadays, newspaper-mediated communication is not the only way to study political communication during electoral campaigns. However, we still consider it a kind of ‘master arena’ to observe party statements in the public sphere. Because journalists record what voters can actually read on the front pages each day, the dataset captures the publicly visible arena in which parties respond to one another and try to shape the agenda.

The dataset consists of a sample of articles that report on the campaign and national party politics in general during the 2 months preceding Election Day. They are coded using core sentence analysis (CSA). This method reduces each grammatical sentence to its most basic ‘core sentence(s)’ structure, which contains only the subject (actor), the object (actor or issue), and the direction of the relationship between the two. For the following analysis, we rely on all coded relations between party-affiliated actors as subjects and any political issue or actor as objects. The analysis is based on around 166′000 actor-issue statements.

At the same time, relying on print news introduces three potential biases. First, visibility bias: larger parties tend to receive more column inches than smaller ones, which could inflate their measured salience scores. As shown below, we mitigate this by using relative salience, i.e. each party’s share of attention across issues, rather than raw counts. Second, editorial slant: newspapers paraphrase, frame, and occasionally criticise party statements, so reported positions may differ slightly from how parties describe themselves. To reduce partisan skew we code the two most widely circulated quality and tabloid newspapers published per country. Including a tabloid offers a broader perspective on how parties were portrayed during the campaign period. Third, platform bias: parties increasingly campaign on social media, which allows for broader discussion and direct non-mediatized messaging, raising the question of whether print alone overlooks important messages. However, our focus on the 8-week official campaign when parties aggressively court mass-media attention limits the risk that major themes unfold entirely outside newspapers. Taken together, print data are not flawless, but they provide the only medium-term, cross-national window on campaign content that is (a) comparable across systems, (b) granular at the sentence level, and (c) anchored in what a broad electorate is exposed to.

For the subsequent analysis, as already pointed out, we classify these sentences into two types – economic issues (composed of social and economic policy issues ⁷ ) and new cultural issues (composed of immigration, European integration ⁸ , environment, and cultural liberalism). For each issue category within the types, we measure the individual parties’ issue salience and issue-specific positioning. Issue salience for individual parties is operationalized by the relative frequency with which the party addresses a given issue during its electoral campaign compared to all other issues. For the parties’ positioning, we use an equally straightforward indicator: the average direction of the issue-specific campaign of a given party, with pro-issue statements coded as +1 and anti-issue statements as −1.

Based on these two indicators, we construct indicators for position and moderation (extreme position). The indicator for (normalized) position equals the deviation of party i’s position on issue j from the mean issue-specific position of all the parties in a given campaign:

$\text{Position}\hspace{0.17em}{\left(\text{normalized}\right)}_{\text{ij}}={\text{position}}_{\text{ij}}-\text{mean}\_{\text{position}}_{\mathrm{j}}$

Positive (negative) deviations from the mean indicate pro-state (pro-market) positions on the economic dimension and universalistic (nationalistic-particularistic) positions on the cultural dimension. To operationalize moderation of individual parties, we rely on its inverse – extreme positioning, and use the squared distance of their issue-specific position from the election mean:

$\text{Extreme}\hspace{0.17em}{\text{position}}_{\text{ij}}={\left({\text{position}}_{\text{ij}}-\text{mean}\_{\text{position}}_{\mathrm{j}}\right)}^2$

The resulting indicator is no longer directed but varies from 0 (=greatest moderation) to a maximum value (=greatest extremism) ⁹ .

For the analysis of the party strategies, our key independent variable is the party family. In addition, we include three control variables – whether or not the parties were cabinet members or provided the prime minister during the previous legislature (incumbent-opposition dynamics), and a linear time trend, starting in 1988, the year of the first election in our dataset. We limit the number of controls to a minimum, given the constraints of the limited number of cases in the party families. Table A3 in the Appendix provides the descriptives of the key variables by party family.

For the analysis of the parties’ electoral success, our dependent variable is the change in their vote share since the last elections. In addition to the already mentioned controls, we also introduce lagged vote shares, i.e., the vote shares obtained by the parties in the previous elections. To some extent, these lagged vote shares take care of the country- and party-specific context conditions (such as the general configuration of the party system and the relative size of the parties in particular), which are not explicitly taken into account by the limited number of controls. In order to distinguish between the long-term and short-term impact of issue-specific campaign strategies on the changing electoral success, we introduce both the lagged issue-specific strategy (its long-term component) as well as the change in the issue-specific strategy (its short-term component) as independent variables.

First, we analyze the determinants of the parties’ strategies, i.e., issue salience, position and moderation (extreme position). In a second step, we account for the effects of these strategic parameters on the changing vote shares of the parties. We use multi-level linear regression for our data analysis, with two additional levels: the units of analysis, party-election pairs, are nested in parties that are nested in countries. In addition, we use robust standard errors ¹⁰ .

Determinants of issue-specific party strategies

Table 1 presents the determinants of issue-specific party strategies for economic and new cultural issues. Before turning to the individual parties’ strategies, let us briefly comment on the effects of the controls. First, with one exception, we do not observe any long-term trends concerning the three electoral strategies. The exception concerns the generally increasing extremism on cultural issues. Second, cabinet members have a somewhat more progressive cultural position than other parties, an unanticipated result.

Table 1.

Determinants of issue-specific party strategies for economic and cultural issues, mixed regression: regression coefficients, t-values, and significance levels (Bold coefficients indicate p < 0.05).

	Economic			Cultural
	Salience	Position	Extreme position	Salience	Position -	Extreme position
	b/t	b/t	b/t	b/t	b/t	b/t
trend	0.001	0.000	0.001	0.001	−0.000	0.001**
	(0.645)	(0.526)	(1.494)	(1.075)	(−0.328)	(2.402)
cabinet	0.010	−0.002	−0.012	−0.002	0.010**	−0.007
	(0.616)	(−0.332)	(−1.285)	(−0.106)	(2.339)	(−1.458)
pm	0.005	−0.005	0.000	0.008	−0.002	−0.003
	(0.348)	(−0.646)	(0.017)	(0.744)	(−0.417)	(−0.476)
rad left	0.007	0.078***	0.054***	−0.027	−0.011	−0.001
	(0.296)	(5.050)	(4.431)	(−1.139)	(−0.767)	(−0.041)
greens	−0.128***	0.005	−0.002	0.114***	0.049***	0.024
	(−5.044)	(0.454)	(−0.215)	(3.312)	(3.091)	(1.251)
social-democrat, ref	0.000	0.000	0.000	0.000	0.000	0.000
liberal	−0.009	−0.057***	0.019**	−0.029	−0.011	−0.023*
	(−0.322)	(−4.817)	(2.164)	(−1.359)	(−1.593)	(−1.695)
ch-dem/cons	−0.033*	−0.047***	0.015**	0.016	−0.062***	0.008
	(−1.811)	(−5.432)	(2.243)	(0.673)	(−4.760)	(0.469)
rad right	−0.123***	−0.034*	0.019*	0.090***	−0.134***	0.099***
	(−4.418)	(−1.668)	(1.870)	(3.455)	(−6.420)	(2.880)
other	0.023	0.061*	0.060	−0.003	−0.000	0.004
	(0.373)	(1.781)	(1.550)	(−0.095)	(−0.017)	(0.193)
Constant	0.281***	0.360***	0.023*	0.126***	0.559***	0.027
	(6.147)	(43.873)	(1.767)	(3.726)	(78.107)	(1.513)
Observations	693	693	693	693	693	693
aic	−839.25	−1824.03	−1641.11	−1050.68	−1853.65	−1448.12
bic	−780.22	−1765.00	−1582.07	−991.65	−1794.62	−1389.09

Turning to the test of our hypotheses concerning the parties’ issue-specific strategies, Figure 1 presents the effects for the various party families. Note that we use the social democrats, i.e., the party family present in all countries, as a reference category. This means that the effects show how the party families deviate from the social democrats. Starting with salience, we observe that the two challenger parties opposed on the cultural dimension, the Greens and the radical right parties, pay less attention to economic and more attention to cultural issues (which they own). Conversely, we find no effect for the radical left and the liberals, which is in line with expectations since they are expected to behave like the social democrats, the reference category. Only a slightly negative effect for the conservatives on economic issues contradicts the expectation that these parties ought to pay similar attention to such issues as the social democrats. In other words, H1 is generally confirmed.

Figure 1.

The determinants of issue-specific strategies by party family: AME’s for economic and cultural issues.

In line with H2a/b, the challenger parties polarize on the issues they own, with the radical left taking clear-cut and even extreme left-wing positions on economic issues and the radical right taking clear-cut and even extreme nationalist and traditional positions on cultural issues. The Greens take clear-cut but not extreme positions on cultural issues. As for the mainstream parties, the liberals and conservatives take clearly more pro-market positions than the social democrats, and their corresponding positions are even slightly extreme compared to those of the social democrats. Conservatives also take more nationalist and traditionalist positions on cultural issues than the social democrats. This is in line with H2a. However, mainstream parties do not opt for accommodating strategies when faced with radical right success. As shown in Figure 2, which is based on Table A4 in the Appendix, a more formal test of H3 for cultural issues shows no evidence for accommodating strategies of centre-right parties. It is the radical left and the social democrats who re-position themselves to some extent in reaction to the support of the radical right. However, they do not accommodate the radical right but rather take more pronounced adversarial positions. Interestingly, the radical right itself reacts to its own success by radicalizing its position, whereas successful green parties rather tone down their message.

Figure 2.

Accommodation of position on cultural dimension to success of Greens and radical right: AMEs.

The issue-specific strategies we have identified are rather long-term strategies that parties do not change from one election to another. This is shown by an analysis of change in strategies (see Table A5 in the Appendix): the effects we found in our previous analysis all disappear except for one – the conservatives also re-position themselves more to the right on economic issues in the short term. Given the importance of contextual constraints, this is not so surprising after all. Depending on the specific circumstances, parties are likely to make short-term strategic adjustments with different emphasis and in different directions. This means that our analysis of the choice of issue-specific strategies also primarily captures the more continuous elements of the parties’ strategies. Overall, the issue-specific strategies the European parties adopt in their election campaigns largely correspond to the long-term expectations: parties largely adopt issue-specific strategies in line with their ideology on the issues they own and their status in the party system, either as a mainstream or challenger party.

Determinants of the parties’ changing electoral success

We test the impact of salience, position, and moderation/extreme position on the parties’ changing electoral success. We are analyzing the effect of these three parameters of the campaign strategy on the changing vote share of the six party families one by one, in an attempt to limit the number of independent variables. This way of proceeding does not take into account possible interaction effects between the three parameters. In an attempt to keep the analysis as straight-forward as possible, given the limits of our data, we do not control either for the competitors’ campaign strategies in a given election, although they may have an important effect on the success of the strategies of each party (Kitschelt and Rehm, 2024). This means that our results provide only a first approximation of the possible campaign effects on the electoral success of the various parties. We control for past success (lagged vote shares) and party-family-specific time trends, as well as for whether the party has been in government and has provided the prime minister during the legislative period prior to the elections. Figure 3, which is based on Table A5 in the Appendix, presents the issue-specific determinants of changing vote shares for the six party families.

Figure 3.

The effect of issue-specific campaign strategies on changing vote shares: AMEs.

First, we estimated a model with only the controls. It confirms what we knew already: the three mainstream party families have been losing voters since the 1990s, the social democrats and the conservatives more so than the liberals. In addition, parties that have been part of the government during the previous legislative cycle are losing vote shares to the tune of around 5%, except if they held the office of the prime minister. The costs of ruling seem to be entirely falling on the minority partners in government. Moving on to the effects of the campaign strategies, we find that these effects are generally weak, in line with the notion that country-specific historical legacies, long-term structural change, and short-term problem pressure and its repercussions on the public agenda constrain the maneuvering space of electoral campaigning. The standard errors are generally large, suggesting that the campaign effects of parties from a given family vary strongly by country and election.

There are some noticeable effects, however. With respect to salience, we find that the long-term emphasis on economic issues tends to enhance the vote share of mainstream parties on the right (the effects are significant at the 10%-level for the liberals and at the 1%-level for the conservatives). By contrast, putting the long-term emphasis on cultural issues tends to enhance the vote share of the Greens and to hurt the radical left. Short-term emphasis on economic issues again benefits the conservatives, but also the social democrats, while the radical right benefits from emphasizing cultural issues in the short term. Taken together, these results provide some weak support for H4a, which suggests that the emphasis on issues they own reinforces the parties’ electoral success and, by implication, the emphasis on issues they do not own, like in the case of cultural issues and the radical left, may hurt their electoral success.

With respect to position, we find three instances where the long-term distinct positioning enhances the parties’ electoral success (H4b): this is the case for the liberals and the radical right taking rightist positions on economic issues and for the Greens taking leftist positions on cultural issues. In all three cases, these parties even benefit from extreme positions (see Figure A2 in the Appendix). In the short term, social democrats benefit from moving to the left on economic issues, and the radical right from moving to the right on such issues, even moving to extreme positions (see Figure A2). All the other effects are insignificant. Except for the Greens benefiting from moving to extreme positions on cultural issues, H4c, which refers to extreme positions, hardly finds any support at all. Thus, unexpectedly, extreme positioning pays off for the liberals, a mainstream party. Only the fact that the radical right benefits from extreme positioning on economic issues is in line with H4c. There are no indications of moderation effects.

Both long-term and short-term effects confirm Kitschelt and McGann’s (1995) “winning formula” for the radical right. The long-term effects also confirm the crucial importance of economic liberalism for the liberals and of cultural liberalism/cosmopolitanism for the Greens. The short-term effects suggest that the social democrats benefit from a return to what Häusermann and Kitschelt (2024) called “old left” strategies. They had identified programmatic strategies that combine progressive left-wing positions on sociocultural issues (“new left”) with left-wing or moderate positions on economic issues (“old left”) as most promising for social democrats. Our results confirm the promise of a reinforcement of the “old left” component.

Conclusion

Based on the literature, we started out by casting doubt on the effectiveness of issue-specific electoral campaigns. Our results confirm these doubts to a large extent, although not entirely. While European parties organize their issue-specific campaigns largely in line with their long-term ideological positions and their status as challengers or mainstream parties, overall, we could find only limited effects of their issue-specific campaigns on the electoral outcome. In line with expectations, during their electoral campaigns, parties generally emphasize the issues they own, and they take clear-cut or even extreme positions on these issues. However, the effects on electoral outcomes are generally weak and/or come with large standard errors, which indicate that there are large country- and election-specific differences for each party family. Having said this, we still find some expected effects. We should also stress that in increasingly volatile campaigns, fluid electorates, and small margins of difference between parties, even small effects might have important implications.

The general result of limited effects has to be nuanced by party family: challengers and mainstream party families have benefited to varying degrees from their respective strategies. On the one end of the spectrum, we could not find any positive effects for the issue-specific electoral campaigns of the radical left. The only issue-specific campaign effect for them is that they are punished for putting the emphasis on cultural issues. On the other hand, the effects of campaigns are the largest for the radical right, which has benefited from increasingly putting the emphasis on cultural issues, from taking clear-cut or even extreme positions on economic issues, and from moving to the right on these issues. As for the mainstream parties, it should be noted that we did not find any effects for moderation or for accommodation. In fact, the social democrats react to the success of the radical right with adversarial positioning on the cultural dimension, and so do the Greens. The centre-right parties neither moderate nor accommodate at all on the cultural dimension. In other words, the success of the radical right does not have an effect on the emphasis they put on cultural issues or on their positioning on these issues. On their economic home turf, the centre-right parties take slightly extreme positions, and the liberals, but not the conservatives, benefit from their emphasis on economic issues and from taking explicit, or even extreme right-wing positions on such issues.

We have made an attempt to differentiate between long-term and short-term issue-specific electoral campaign strategies and their effects on the electoral outcome. We come to the conclusion that the various party families pursue long-term issue-specific strategies in line with their ideological orientations and issue ownership. In this paper, it proved to be impossible to identify the factors that are associated with short-term strategic choices of the different party families, most likely because these short-term choices react to highly specific events and context characteristics. Future research should more closely investigate the factors associated with short-term strategic choices. In terms of the effects of these strategies on the electoral success of the various party families, we believe that the distinction between short-term and long-term effects is promising, even if only a few components of the issue-specific strategies have any effect at all. The distinction opens the possibility for investigating how the parties’ responses to short-term changes in their environment provide opportunities and threats for their electoral success. Originally, we had approached the distinction between long-term and short-term strategies by distinguishing between different periods (pre-crisis vs crisis period) within the time span covered by our data. However, this strategy met with a double constraint: on the one hand, introducing period-specific effects rapidly increases the number of parameters in the models. On the other hand, the number of cases decreases if we at the same time also introduce lagged effects. For the time being, the combination of the two types of constraints has led us to give up on the analysis of specific sub-periods, although we think this is a promising path to pursue for future analyses.

Supplemental material

Supplemental Material - Electoral campaign effects: An aggregate analysis of electoral issue competition

Supplemental Material for Electoral campaign effects: An aggregate analysis of electoral issue competition by Hanspeter Kriesi, Chendi Wang and Argyrios Altiparmakis in Party Politics.