ArticlePDF Available

Why The Big Picture Matters: Political and Media Populism in Western Europe since the 1970s.


Abstract and Figures

While a vast theoretical literature argues that we live in populist times, the success of populism beyond its electoral dimension is rarely investigated empirically. This paper analyses the development of populism in five Western European countries (Austria, Germany, Netherlands, Switzerland, and the United Kingdom) from the 1970s to the 2010s. First, it measures whether political parties articulate populist discourses more and more often in their election manifestos. Second, the paper tests whether the presence of populism has increased over time in newspaper articles. The results show that populism is not a new phenomenon and that there is no linear increase over the decades. Moreover, while election manifestos are significantly more populist in the 2010s than in previous decades, populism in newspaper articles remains rather stable at a low level, suggesting that the media curb rather than foster populist discourses.
Content may be subject to copyright.
Why The Big Picture Matters: Political and Media
Populism in Western Europe since the 1970s.
Department of Political Science (IPZ), NCCR Democracy, University of Zurich,
Affolternstrasse 56, 8050, Zurich, Switzerland
Department of Political Science (IPZ), NCCR Democracy, University of Zurich,
Affolternstrasse 56, 8050, Zurich, Switzerland
Abstract: While a vast theoretical literature argues that we live in populist times, the success of
populism beyond its electoral dimension is rarely investigated empirically. This paper analyses the
development of populism in five Western European countries (Austria, Germany, Netherlands,
Switzerland, and the United Kingdom) from the 1970s to the 2010s. First, it measures whether
political parties articulate populist discourses more and more often in their election manifestos.
Second, the paper tests whether the presence of populism has increased over time in newspaper
articles. The results show that populism is not a new phenomenon and that there is no linear
increase over the decades. Moreover, while election manifestos are significantly more populist in
the 2010s than in previous decades, populism in newspaper articles remains rather stable at a low
level, suggesting that the media curb rather than foster populist discourses.
KEYWORDS: Populism, Comparative Politics, Media, Western Europe, Discourse, Content
Populism is a disrupting force in Western Europe. In the last two decades, new right-wing
and left-wing parties which are labelled as ‘populist’ have been increasingly successful, and
in several cases have participated in cabinets and government coalitions (Mudde 2013).
Even in Germany, a country traditionally immune to the sirens of populism, a party such
as Alternative for Germany scores increasingly well in regional elections (Arzheimer 2015).
But is populism a new phenomenon or, rather, a companion of democracy? And what is
the role of the media vis-
a-vis its diffusion? This paper tries to answer these questions
through a content analysis measuring the presence of populist discourses in five Western
European countries (Austria, Germany, Netherlands, Switzerland, and United Kingdom)
from the 1970s to the 2010s.
We would like to thank the Swiss National Science Foundation for the support; the team from the NCCR
Democracy project “Challenges to Democracy in the 21st century”: Martin Wettstein (creator of the software for
the content analysis and key figure for the success of the project), Nicole Ernst, Anne Schulz, and Dominique
Wirz for the impressive work on the codebook; all the excellent coders that made all of this possible; and the
anonymous reviewers for their very valuable comments.
Swiss Political Science Review doi:10.1111/spsr.12267
©2017 Swiss Political Science Association
By analysing the relevant literature about the diffusion of populist discourses, it is clear
that from the 1990s several mutually reinforcing factors seem to have fostered the presence
of populism in Western Europe: the impact of globalization and supranational integration,
the role of the media in granting visibility to populist actors, and the reaction of
mainstream parties which in turn increasingly rely on populist discourses. So far, however,
only a limited number of empirical studies analysing the diffusion of populist discourses
have been conducted. It is precisely this gap that we aim to fill. Moreover, we expand the
scope of the analysis compared to other studies by observing the political and mediatized
diffusion of populist discourses over the last four decades.
The article is structured as follows. The first section provides a definition of populism as
a worldview which becomes measurable as soon as it is articulated discursively. Then, the
relevant literature and our hypotheses about the temporal evolution of populist discourses
are presented. The paper subsequently illustrates the details of the research design and the
operationalization of populism. Finally, the results are presented and discussed, and
further areas of research are suggested.
Defining Populism
In line with the ideational approach that has been gaining traction among scholars, we
understand populism as a set of ideas.
In particular, we define populism as a thin
ideology which considers in a Manichean outlook society to be ultimately separated
into two homogenous and antagonistic groups, the positively characterized ‘virtuous
people’ versus the negatively connoted ‘corrupt elite’, and which postulates unrestricted
sovereignty of the people.
On the basis of this definition, three pillars of the populist ideology can therefore be
identified: people-centrism, anti-elitism, and claim for popular sovereignty (March 2011).
First, the people is seen as a positively evaluated, unified entity, with common feelings and
opinions. Second, the elites are considered as out of touch with the people, and are
accused of following particularistic interests conflicting with the common will of the
people. Finally, as a consequence of the first two aspects, populism advocates more
popular sovereignty.
The lines of conflict between the people and the elites change connotation according to
the thick or full ideology associated with populism. In other words, the boundaries of
‘people’ and ‘elite’ vary according to the ideology which is articulated in combination with
populism (socialism, nationalism, or liberalism, to mention the most common ones).
Economic, cultural, or political interpretations of the conflict between the elites and the
people originate from different configurations of social cleavages (Canovan 1981).
The ideational approach, compared to the other existing approaches,
populism as a mental framework or a worldview and therefore it makes it possible to
measure its discursive articulation (Hawkins 2009;2010). As soon as the populist
worldview is present in a discourse oral or written it becomes measurable. In other
Among others: Canovan 2004; Mudde 2004; Abts and Rummens 2007; Jagers and Walgrave 2007; Rovira
Kaltwasser 2011; March 2011; Rooduijn et al. 2014; Kriesi 2014.
Based on the definition given by Mudde 2004; Albertazzi and McDonnell 2008; Mudde and Rovira Kaltwaser
Populism, for example, has been understood in stylistic terms (Kazin 1995), as a type of organization (Weyland
2001), or as mobilization (Skocpol and Williamson 2012).
2 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
words, by adopting the ideational approach we are able to trace variations in levels of
political and mediatized populism over time and across countries, not only between actors.
Indeed, we do not aim at a binary classification of populist and non-populist actors, but at
measuring the levels of populism (Rooduijn et al. 2014).
Between the 1980s and the 1990s, Western Europe underwent profound transformations
that have been linked to the presence and success of populism. First, counter-mobilization
of the extreme populist right against the universalistic values endorsed by the The New
Left deeply transformed the political space in Western Europe (Kriesi et al. 2006). Second,
transformations in the media market shaped political communication around sensational
news and personalities rather than on policies (Blumler and Kavanagh 1999; Ellinas 2010),
thus granting growing space for new political challengers. In turn, given the growing
success of populist actors, mainstream parties also started articulating populist discourses
increasingly often (Mudde 2004). In this section we present our hypotheses built on the
transformations described in different strands of literature on populism, which all identify
the 1990s as a turning point for the presence of populist discourses.
Globalization affects virtually every aspect of contemporary societies, including the
structure of political conflict. In particular, the cultural dimension of competition has
become more relevant and has narrowed into a battle around multiculturalism. As a result
of these transformations and the strategic repositioning of parties, a tripolar configuration
comprising the left, the moderate right, and the new populist right emerged in Western
Europe. Another aspect of globalization supranational integration triggered a political
competition between nation-states and supranational actors, which in turn has been
exploited by the anti-elitist rhetoric typical of populism (Kriesi et al. 2008).
According to this strand of literature, the process of globalization has contributed to a
restructuring of the two axes of political competition in a way that allowed populism to
mobilize the electorate along new lines of conflict (Kupchan 2012; Kriesi et al. 2006). In
particular, the so-called ‘losers’ of the modernization processes have been effectively
mobilized, thus granting unprecedented electoral success to populist actors. This was
possible because the ‘losers’ of globalization felt that higher unemployment, growing
inequalities, and decreased social services threatened their style of life and social status
(Cox and Sinclair 1999; Zimmerling 2005). Moreover, the perceived threat was initiated
also by the weakening protection of the traditional national boundaries, thus originating a
reaction invoking protectionist measures and national independence (Cerny 1999).The
process of supranational integration and the loss of autonomy of nation-states vis-
European institutions did not produce a new structural cleavage but reshaped the existing
ones, making the opposition between integration and state autonomy more crucial than
ever (Bornschier 2010).
If, on the one hand, the cultural cleavage became increasingly salient in explaining the
success of right-wing populist parties, this does not mean that, on the other hand,
economic issues lost all their salience. In fact, following the Great Recession that
characterized the period 20082012, left-wing populist discourses have resonated across
Europe with increasing frequency. The socio-economic and political effects of the
economic crisis could easily explain also the success of left-wing populist parties such as
Syriza in Greece and Podemos in Spain. The present paper does not make a distinction
between right-wing and left-wing populist discourses, but rather focuses on the overall
Political and Media Populism since the 1970s 3
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
levels of populism in manifestos and newspaper articles. However, because we do not have
a Southern European country in our study, populism in the five considered countries is
dominated by right-wing parties emerging from the new cultural divide and not by left-
wing populist parties emerging from the Great Recession.
Given the emergence of new populist parties politicizing on the new lines of conflict in
response to the transformations of globalization and modernization, we hypothesize that
party manifestos have become more populist since the 1990s. Compared to other types of
party documents, election manifestos have some general advantages as a source for
measuring parties’ ideology over time. First, they are standardized over time and across
parties. During the research period of over four decades, almost every party produced
similar documents with a clear overview of their general political goals (Rooduijn et al.
2014: 566ff.). Moreover, election manifestos are almost the only documents in regionally
fragmented party organizations that offer a univocal position. Finally, election manifestos
are more independent from the media compared to other party documents such as press
releases. We therefore test the following hypothesis:
Hypothesis 1: The amount of populist discourses in party manifestos has increased since the
As mentioned above, the transformation of the political space and cleavages fostered the
success of populism, which in turn had a strong impact on the electoral competition. In
particular, mainstream parties became less appealing to the electorate and reacted by
becoming increasingly populist themselves. This phenomenon can be compared to the
strategic adaptation of parties in terms of issues positioning (see van Spanje 2010). It has
been argued that change in party positions can come from external stimuli (Harmel and
Janda 1994), and in particular that parties tend to shift their policy positions according to
changes in their opponents’ position at the previous election (Adams and Somer-Topcu
2009). As Mudde has argued, mainstream political parties started incorporating elements
of the populist message in their own discourses “at least since the early 1990s”, and
consequently “populism has become a regular feature of politics in western democracies”
(2004: 551). The process of ‘populistization’ of mainstream parties constitutes the core of
what he defined as “populist Zeitgeist” (2004: 542). Other authors have spoken of
populism in terms of a contagious phenomenon (Bale et al. 2010; M
eny and Surel 2002).
Subsequent empirical studies did not confirm the theoretical expectations. In fact, they
found that neither mainstream nor populist parties have become more populist (in their
election manifestos) over time (Rooduijn et al. 2014). However, given the extremely low
number of empirical studies on the topic and their limited scope, and considering that the
content analysis presented in this study is based on a broader time span and a different
case selection compared to previous ones, it seems advisable to further investigate whether
or not the “populist Zeitgeist” is a real phenomenon or just a suggestive impression.
Therefore, we hypothesize that:
Hypothesis 1A: The electoral programmes of mainstream political parties became increasingly
populist since the 1990s.
Populist discourses are supposed to have become increasingly present not only in party
documents such as election manifestos, but also in the media. Once again, this
transformation took place at the end of the last century. The idea that news media are
responsible for the diffusion and success of populist discourses has been proposed by
several authors in recent years, and it has been linked to the process of mediatization of
4 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
politics (Mazzoleni 2008; Ellinas 2010; Esser and Str
ack 2014). According to this
strand of literature, newspapers and television channels follow a commercial logic which
is supposed to resonate with the populist style of communication (Mudde 2007;
Akkerman 2011). Populist discourses are indeed considered controversial and
newsworthy, thus attracting a broad audience which allows commercial media outlets to
increase their revenues. According to this media-selection logic, statements containing
populist messages acquire greater visibility compared to non-populist statements.
Following the taste of the public, political debates are therefore increasingly framed as a
competition opposing prominent personalities. Populist actors, with their flamboyant
charisma, are intended to become media darlings and attract the public’s attention
(Mazzoleni 2003). As M
eny and Surel (2000) noted, populist actors rely on provocative
and fiery statements as well as on vicious attacks on their opponents, and these
characteristics instead of being rejected by the media are becoming the state of the
art of media debates.
Empirical research has indicated that the coverage of right-wing populist actors is
related to their popular support (Vliegenthart et al. 2012), while the salience of ‘populist
topics’ and anti-politics sentiments in media coverage can contribute to their electoral
success (Boomgaarden and Vliegenthart 2007; Hellstr
om et al. 2012). Moreover, Rooduijn
(2014) showed empirically that the political debates in the newspapers of five Western
European countries became increasingly populist over the 15 years between around 1990
and 2005.
Considering both theoretical and empirical studies on this topic, it seems plausible to
argue that news media have devoted increasing space to populist discourses over time,
especially since the 1990s. Therefore, we hypothesize that:
Hypothesis 2: The amount of populist discourses in news media has increased since the 1990s.
In a second step, it can be useful to disentangle politically generated populist discourses
from media generated ones. By considering the different sources of populist discourses in
the media in isolation from one another, it is possible to better understand the
mediatized diffusion of populist discourses. An increase in the levels of populism in the
media (hypothesized above) can reflect three different but potentially complementary
First, political parties might have become increasingly populist when releasing interviews
and staging media-oriented events such as press conferences and party conventions. This
would also be in line with the expected ‘populistization’ of the political arena (H1 and
H1A). Second, even in the case where political parties did not become increasingly
populist over time, it is possible that their populist statements are reported more
frequently in the newspapers because of their controversial and newsworthy content
according to the media-selection logic mentioned earlier. Hence, we hypothesize that:
Hypothesis 2A: The amount of populist discourses articulated by political actors (politically
generated populism) reported in the newspapers has increased since the 1990s.
Third, it could be the case that journalists increasingly articulate populist discourses on
their own (Kr
amer 2014; Mazzoleni 2014). Journalists not only report and frame
discourses from political actors, they might also play the role of defenders of the ‘man in
the street’ against the corrupt elites, or advocate unrestricted popular sovereignty. Given
the changing role of newspapers in the context of the process of mediatization of politics
described above, we hypothesize that:
Political and Media Populism since the 1970s 5
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
Hypothesis 2B: The amount of populist discourses articulated by journalists (media generated
populism) in the newspapers has increased since the 1990s.
Moreover, among all media outlets, commercial television channels and tabloid
newspapers are considered the main allies of populist politicians. Their relationship can be
described as one of “complicity”, while quality newspapers are considered as “paladins” of
the status quo (Mazzoleni 2003: 8). The need to ‘sell’ political issues by making the
political process understandable and interesting for the audience is considered to resonate
with the style of populist actors. Although some recent empirical research has contradicted
this assumption (Akkerman 2011; Rooduijn 2014), further comparative studies are needed.
In particular, the database employed in this article covers a broader time span and might
offer a different perspective. Moreover, the results might also differ from previous studies
because of different case selections (Switzerland and Austria are included in the following
analysis but not in the two studies above, whereas Rooduijn (2014) considered France and
Italy, which in turn are lacking in our study). Hence, we hypothesize that:
Hypothesis 2C: Tabloids articulate or give space to populist discourses more often than quality
All the hypotheses are summarized in Table 1 in order to make it easier to visualize the
different expectations and strands of literature examined.
We selected five Western European countries: Austria, Germany, Netherlands,
Switzerland, and the United Kingdom. In these countries, the success of populist parties
has been heterogeneous: from countries that have had successful radical right populist
parties since the 1990s (Austria and Switzerland) to countries in which such parties were
unsuccessful until recently (Germany and the UK), with the Netherlands constituting an
intermediate case. As Rooduijn (2014) has shown that the diffusion of populist discourses
in the public debate varies according to the electoral success of populist parties, the
variance in the selected countries is important. Moreover, these countries display different
political cultures, with Switzerland and Germany as two clearly consensus democratic
systems, the UK as a clear majoritarian democracy (Lijphart 1999), and the other two
Table 1: Summary of Hypotheses
Hypotheses Theory Object Expectations
H1 GLOBALIZATION Levels of populism in election
Grow from the 1990s
Levels of populism in mainstream
election manifestos
Grow from the 1990s
H2 MEDIATIZATION Levels of populism in the news Grow from the 1990s
H2A MEDIATIZATION Levels of politically generated
populism in the news
Grow from the 1990s
H2B MEDIATIZATION Levels of media generated populism
in the news
Grow from the 1990s
H2C MEDIATIZATION Levels of populism in tabloids Higher than in quality
6 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
cases between them (see e.g. Papadopoulos 2002: 53 for a discussion about the impact of
this dimension on the degree of populism in a country). However, the above countries are
all from central or northern Europe, which means that the conclusions from the following
analysis are restricted to these areas and may not hold true for Southern Europe.
The longitudinal dimension of this study considers one election per decade from the
1970s to the 2010s, thus including five elections per country. In each decade, the elections
have been selected in order to be as close in time as possible in each country, thus
increasing the comparability of the results (see Table 2).
To test the presence of populism in party manifestos (H1 and H1A) we selected for
each country and each time point those parties which gained at least 3% of the vote
share in the respective election.
Only the Netherlands constitute an exception. Due to
their exceptionally proportional electoral system and the high number of small parties, the
threshold for the Netherlands has been raised to 5%. The total amount of election
manifestos included in the analysis is 111, containing 9176 coded statements (Table 2).
For the measurement of populism in print news, for each country we selected one
tabloid and one quality newspaper (cf. H2C). Although other news media exist
television, magazines, and websites, among others newspapers still represent a central
arena for public opinion (Rooduijn 2014, Roggeband and Vliegenthart 2007).
Moreover, we decided to focus on the four weeks of electoral campaigning immediately
preceding election day because the media generally pay more attention to politics during
the salient phases of election campaigns (Koopmans 2004). Then, every article containing
the name (or acronym) of at least one of the parties which eventually received more than
3% at the elections was selected.
Afterwards, we removed every article without a
reference to the politics of the respective country.
Table 2: Number of Election Manifestos, Newspaper Articles, and Coded Political Statements
Statements in
in Articles
Germany S
72, 83, 94, 02, 13 23 1739 485 2986
Netherlands NRC
72, 82, 94, 02, 12 21 2405 783 5388
UK The Times/The
74, 83, 92, 01, 10 16 1916 640 4745
Austria Die Presse/Kronen
75, 83, 94, 02, 13 22 1771 909 5705
Switzerland Neue Z
75, 83, 95, 03, 11 29 1345 726 5147
SUM 111 9176 3543 23971
The party manifestos have been found mainly on the website of the party manifesto project
( see Lehmann et al. (2016). Other sources are the websites of the respective parties,
national libraries, and archives.
For the Netherlands the threshold was again raised to 5% (see above).
Political and Media Populism since the 1970s 7
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
Finally, the articles were randomly sampled in order to reduce the total amount of
articles to within the limits of the available budget.
On average, 110 articles per country
and per election-year were coded between the 1970s and the 2000s. For the more recent
elections (20102014), on average 273 articles were selected. The total amount of articles is
3543, with 23971 coded statements on societal actors and political issues (Table 2).
Operationalization of Populism
Our data were obtained through a quantitative content analysis of electoral manifestos
and newspaper articles. The manifestos and newspaper articles were analysed by
extensively trained coders, who used a comprehensive codebook. Political statements were
used as a unit of measurement. They contain either an evaluation of a societal actor or a
position regarding a political issue. Societal actors can be political actors (politicians,
parties, parliaments, governments), economic actors, national and supranational
organizations, the people, judiciary actors, security actors (police, army), religious actors,
or the media. For the political issues, a set of 12 issue-categories containing all relevant
topics (as welfare, budget, army, immigration, Europe, security, etc.) was used.
Statements are coded as populist according to the above-outlined conceptualization of
the populist worldview with its three aspects: “people-centrism”, “anti-elitism”, and
“popular sovereignty”. A statement is coded as people-centrist if the speaker either claims
to be close to the people, speaks of the people as a monolithic actor with a common will,
stresses the virtues of the people, or praises the positive achievements of the people. A
statement is anti-elitist if it discusses any kind of cleavage between the elites and the
people. Finally, a statement contains a claim for popular sovereignty if it demands more
power for the people in general or regarding a specific political issue, such as EU
membership. As soon as one of the thus operationalized three pillars of populism
(summarized in Table 3) is present, a statement is coded as populist.
Three important caveats concerning the operationalization of populism have to be
introduced. First, it is important to stress that in order to code the presence of anti-elitism,
the target of the critique must be an elite in general. Therefore, criticizing part of an elite
like a specific party or politician is not considered as a populist statement. However,
which type of elite is criticized might vary according to the broader ideology of the
speaker. The criticized target may be political (‘established parties’, ‘the government’),
financial (‘the capitalists’), economic (‘the banks’) or cultural (‘the academics’) elites, the
media, or just unspecified elites (‘the mighty ones’). Furthermore, even criticizing the elites
in general is only populist if the critique is made with a reference to the people.
Second, the concept of ‘people’ can also be defined in different ways. The codebook
includes also the expressions for people as an ethnos (‘the Dutch’), people as a function
(‘the voters’), a hypothetical prototype of the people (‘the common man’), or any other
term which can stand for ‘the majority in the society’.
Finally, and contrary to other studies (Rooduijn et al. 2014: 567, Rooduijn 2014: 734),
we do not look for the co-occurrence of the dimensions of populism. We agree that a
For the elections of the 1970s to those of the 2000s a systematic sample of days has been drawn from every
election period. In a first round, one day for each week was randomly selected. The same procedure was repeated
until a total of at least 40 articles was reached for every newspaper for one decade. For the elections in the 2010s,
the articles from all days were selected and afterwards a random sample of them was used for the content
8 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
critique of the elites is not populist per se, but only in the case that it has any reference to
the people. Hence, we incorporate this requirement in the operationalization of anti-elitism
(see Table 3). By contrast, we do not believe that the opposite is true as well. For
example, political statements like “We are the only party in this country that cares about
the people” (Closeness to the people, see Table 3), “The people simply does not want
more immigration” (Stating a monolithic people), “The people is totally capable of taking
Table 3: Operationalization of Populism
Dimensions of
Populism Aspects Questions in Codebook
Closeness to the
Does the speaker claim to belong
/ be close to / know / speak for
/ care for / agree with / perform
everyday actions like / represent
/embody the people?
0.871 0.890
Stating a
Does the speaker describe the
people as homogeneous, sharing
common feelings, desires, or
0.263 0.073
Stressing the
virtues of the
Does the speaker describe the
people in a positive way (moral,
credible, competent, no lack of
understanding, etc.)?
0.935 0.959
Praising the
Does the speaker stress positive
actions and positive past and
future impacts of the people
(responsible for a positive
development / situation, not
being responsible for a mistake,
0.895 0.931
Anti-elitism Exclusion of the
elite from the
Does the speaker describe the
elites as not belonging to / not
being close to / not knowing
the needs of / not caring about
/ not speaking on behalf of /
not empowering / deceiving the
0.602 0.526
Claiming power
for the people
Does the speaker argue that the
people should have / gain / not
lose power?
Does the speaker give the people
the competence to act or decide
on a specific political issue?
Does the speaker demand
institutional reforms for more
participation of the people in
0.966 0.972
Political and Media Populism since the 1970s 9
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
the important decision in this country by themselves” (Stressing virtues of the people),
“The people was always right” (Praising the people’s achievements), or “The people should
regain the power in this country!” (Claiming power for the people) can certainly be
labelled as populist even without an explicit critique of the elites.
Not counting such statements as populist seems to be problematic both theoretically and
empirically. Since around 90% of the elite-critique of the parties is directed towards
political elites, such as the government, the parliamentary majority, the established parties,
etc., its use is almost exclusively limited to oppositional and non-mainstream parties.
Therefore, if one of the above-mentioned statements needs to be accompanied by a
critique of the elites, governmental parties and mainstream parties would almost never
turn out to be populist by following such an operationalization of populism.
However, as
the results below will show, new non-mainstream parties are clearly more populist than
mainstream parties even with our operationalization.
A large team of coders was trained for one week before coding. For the assessment of
the codebook reliability and the coder validity, two kinds of measurement were
implemented. First, we conducted a test of expert validity where we checked the agreement
of the coders with a ‘gold standard’ solution. The solution was prepared in collaboration
with the project leaders, and the agreement with it serves as an indicator of coder validity.
In addition to this test, a hidden reliability test consisting of 29 texts in German and
English was conducted during the content analysis. Coders were assigned single texts from
this test among their usual work without notification. The results of these two tests for
every aspect of populism presented in Table 3 are generally acceptable, with average-levels
of Cohen’s Kappa of 0.73 for the ‘gold standard’ test and 0.76 for the hidden reliability
test respectively.
The calculation of the populism scores as the dependent variable of all hypotheses, then,
is straightforward. The units of analysis are the election manifestos (H1 and H1A), and
the newspaper articles (H2H2C). The level of populism
is simply the percentage of
populist statements compared to the total amount of coded statements (target evaluations
and issue positions) in the respective manifesto or article. The dependent variable thus has
a theoretical range from 0 to 100. In the election manifestos, 537 or 5.9% of the political
statements are populist according to the operationalization outlined above. The populism
scores of the manifestos range from 0% (which is the case for 20 of the 111 manifestos) to
30% (Green Party of Germany in 2013) of the statements containing populism. In the
newspaper articles, a total of 312 (1.3%) of the coded statements contain populism.
In order to better interpret what 1% of populist statements means when compared to
5% or 10%, one can think about the difference between apple cider, beer, and wine: the
presence of populist statements can be compared to the percentage of alcohol in the three
drinks. For example, the manifesto of a highly populist party like the SPD (Germany) in
1983 or the FP
O (Austria) in 2013 with 2030% of populist statements, would be a
Martini cocktail, while the manifesto of a moderately populist party 5% populist
statements, as in the case of the new Austrian party NEOS in 2013 would be a pilsner
This becomes apparent in the study by Rooduijn et al. (2014: 567), which defines a political statement about the
people as populist only if there is an explicit elite-critique in the same paragraph of the election manifesto. This is
why nearly all mainstream parties have a very low populism score of between 0 and 1 percent (paragraphs
containing a populist statement).
‘Populist statements’ and ‘messages’ are used as synonyms throughout the paper, while ‘populist levels’ or
‘degree of populism’ refer to the percentage of populist statements.
10 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
beer. To provide another point of reference which can help in understanding the
proportions, we calculated the presence of three recurrent issues in the same samples of
manifestos and articles we used to measure populism. In manifestos, 5.9% of the
statements contain populism, compared to 4% of statements about immigration politics,
7% about European integration and EU-politics, and 16% about the welfare state. In
newspaper articles, only 1.3% of coded statements contain populism, compared to 3.1%
of statements about immigration and Europe, and 5.1% about welfare.
For hypothesis 1A, all parties were categorized as mainstream or non-mainstream.
order to test hypotheses 2A and 2B, the coders had to define the speaker for every
statement in the newspaper articles. Therewith we were able to differentiate between
populist statements articulated by a political actor, by the media outlet itself, or by any
other type of actor. For the measurement of the total amount of populism in the media
(H2 and H2C), political statements of all possible speakers were considered, be they
political actors, journalists, or any other person or organization. For testing H2C we used
a dummy variable which differentiated between the two types of newspapers, with 0
referring to the higher-quality and 1 to tabloid newspapers. Another dummy variable
differentiated between two types of newspaper articles that were coded: normal news
articles and editorial articles. Finally, we controlled for the length of the election
manifestos and newspaper articles by their number of words (standardized by their
respective group-means).
Since both units of analysis election manifestos for H1 and H1A and newspaper articles
for H2 are not independent from each other, we conducted multi-level regression
analysis for the formal testing of the hypothesis. The 3543 newspaper articles are nested in
time as well as in journals which are in turn nested in countries. Given that time is the
main independent variable of our hypotheses and the small number of countries
considered, we conducted a two-level model in which the articles are nested in the ten
newspapers. The five decades and countries will be accounted for by dummy variables.
The election manifestos, on the other hand, are nested in time and in parties which are
themselves nested in the five countries. Again we designed a two-level model, in this case
with the 111 manifestos nested in 39 parties. We then introduced dummy variables for the
countries and for the decades. The estimation method is always restricted maximum
likelihood. For the analysis of the degree of populism in the election manifestos of all
parties (H1, Models 1 and 2), of mainstream parties (H1A, Models 3 and 4), and of
newspaper articles (H2, Models 5 and 6) we always proceeded in two steps: first, with the
1990s as the reference category for the time-dummies because our hypotheses state an
increases in the level of populism since this decade; second, we want to analyse whether
It is important to note that a statement can address a certain issue and at the same time contain populism.
Mainstream parties belong to the following four party families: conservatives, liberals, social democratic, and
Christian democratic, as defined by the Party Manifesto Project Dataset (Lehmann et al. 2016). All the parties
belonging to other party families have been coded as non-mainstream apart from the Swiss People’s Party, which
was until the beginning of the 1990s an agrarian party and coded as mainstream before it transformed to a
classical radical right party, thus becoming non-mainstream from then on. The same is true for the Freedom
Party of Austria, which we also regard as mainstream in the 1970s and 1980s and non-mainstream thereafter. A
third exception is the Ring of Independents (LdU), a small party in Switzerland which is labeled “social
democratic” in the manifesto project, but we considered it as non-mainstream.
Political and Media Populism since the 1970s 11
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
the degree of populism in manifestos and newspapers is significantly higher in the most
recent decade compared to the past and therefore use the 2010s as reference category. For
the country-dummies (not shown in the result table), we always selected the UK as the
reference category for the manifesto models (Models 14) because it has the median value
of the dependent variable. For the same reason, we selected Germany as the reference
category for the newspaper models (Models 5 and 6).
Figure 1 shows the longitudinal development of populism in election manifestos and daily
newspapers in the five countries separately, as well as the average values of all of them.
The values indicate the average degree of populism in the election manifestos of all
considered parties (solid lines) and the average degree of populism in the two selected
newspapers per country (dotted lines). The most apparent finding is that election
manifestos generally contain more populist statements than articles. This is the case for all
countries and nearly all decades. On a closer look, another pattern appears. In the last
decade, election manifestos contain an above-average degree of populism in each country
apart from the Netherlands. In Germany, Austria, and the UK populism even reaches its
highest level of the whole research period.
However, there are also several time trends which are much less in line with the
expectation of the hypotheses which state that populism is mainly a new phenomenon
which only began its rise since the 1990s. In the election manifestos, the UK in the 1970s
has a relatively high value of populism which dropped in the two successive decades.
Germany and Austria both have a peak in the 1980s, while Switzerland has (a more
modest) one in the 1990s. The Netherlands, by contrast, has a rather stable level of
1970 1980 1990 2000 2010
0 5 10 15
% populist statements
1970 1980 1990 2000 2010
0 5 10 15
% populist statements
1970 1980 1990 2000 2010
0 5 10 15
% populist statements
1970 1980 1990 2000 2010
0 5 10 15
% populist statements
1970 1980 1990 2000 2010
0 5 10 15
% populist statements
1970 1980 1990 2000 2010
0 5 10 15
All 5 countries
% populist statements
Figure 1: Populism in Election Manifestos and Newspaper Articles
Note: The last subfigure contains the average percentages of populist statements in all five countries.
12 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
populism during the research period which runs also counter to the expectation of the
hypotheses. These results seem to indicate that populism is a cyclical occurrence in
Western European countries rather than a new phenomenon.
In contrast to the high variation in the degree of populism in the election manifestos,
newspaper articles remain more or less stable on a low level of populism without clear
time patterns. In Austria, the Netherlands, and the UK the populism level has remained
rather stable in the last three decades. In Germany the level of populism in the last two
decades is even lower than in previous decades. Over all countries, the high average level
of populism in the manifestos of the 2010s is reflected in only a narrow increase of
populism in the newspapers. This increase is substantial in absolute values only in the case
of Switzerland (from 0.82.5% populist statements on average in the articles of the two
selected newspapers).
One last interesting finding concerning the variation between the countries relates to
Germany: in the last decade it has by far the greatest gap between a high degree of
populism in the manifestos and a low degree in the newspapers. Although in the 2013
elections German parties were on average more populist in their election manifestos than
the parties of any other country in any other decade of the study (16% of political
statements containing populism), the respective degree of populism in the newspapers
during the month preceding the election was lower than in any of the other four countries
Table 4: Populism in Election Manifestos.
Dependent Variable: Percentage of populist statements in Manifestos
All Parties Mainstream Parties
Model 1 Model 2 Model 3 Model 4
Intercept 5.76*** 8.92*** 6.22** 6.58***
(1.74) (1.63) (1.92) (1.89)
1970s 2.26 5.42*** 2.89 3.24
(1.77) (1.65) (1.74) (1.67)
1980s 0.23 2.93 0.15 0.21
(1.80) (1.69) (1.79) (1.74)
1990s 3.16* 0.36
(1.55) (1.71)
2000s 1.09 4.25** 1.16 1.51
(1.69) (1.56) (1.79) (1.73)
2010s 3.16* 0.36
(1.55) (1.71)
Length (cent) 0.50 0.50 0.95 0.95
(1.40) (1.40) (1.39) (1.39)
AIC 683.42 683.42 473.89 473.89
BIC 715.93 715.93 502.47 502.47
Log Likelihood 329.71 329.71 224.94 224.94
Num. Obs. 111 111 80 80
Num. Groups: Parties 39 39 23 23
Var: Parties (Intercept) 0.02 0.02 3.21 3.21
Var: Residual 30.13 30.13 22.86 22.86
Note: ***p <0.001, **p <0.01, *p <0.05. Results of two-level regression models with election
manifestos nested in parties. All models contain country-dummies (not shown).
Political and Media Populism since the 1970s 13
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
in the same decade. There seems to be a huge gap between the parties and the (two
selected) newspapers in Germany regarding the frequency of populist statements in favour
of the people or which discuss the gap between them and their elites.
The last part of Figure 1 shows the average values of populism for all countries. What
is worth mentioning is that the above-stated difference between the level of populism in
election manifestos compared to the one in newspaper articles was not only high across
the whole research period, but also increased over time: in the 1970s the amount of
populism in manifestos was double compared to the amount in newspapers, while in the
2010s the percentage of populist statements in manifestos was four times higher than in
newspaper articles.
The formal test for hypotheses 1 and 1A concerning the increase in the degree of
populism in the election manifestos is presented in Table 4. The dependent variable in
Models 1 and 2 is the degree of populism in the 111 election manifestos analysed.
Country-dummies (not shown in the table) are also included in all models. The decade
coefficients show the same non-linear time-trend as the aggregated degree of populism in
the manifestos. With the 1990s as the reference decade (Model 1), only the difference
compared to the 2010s is significant on the 5% level. If we change the reference category
to the 2010s (Model 2) we can state that in this decade the manifestos were significantly
more populist than in the decades before (except for the 1980s). The differences are also
substantive, ranging from 2.9 to 5.4 percentage points of populism.
The results thus show that unprecedented levels of populism in the party manifestos
have characterized the last elections. This would confirm hypothesis 1. However, it is
important to note that the increase is not linear: in the 1970s and 2000s the levels of
populism in party manifestos were very similar and rather low compared to other decades,
while in the 1980s and 1990s the levels of populism were average. For this reason, it is
difficult to consider the process of globalization and modernization as the only cause of
the increase of populism: the 1990s do not seem to constitute a watershed in this regard,
and it is possible to hypothesize that the presence of populism in party manifestos is rather
a cyclical phenomenon.
Models 3 and 4 are identical to the first two models except that only the 80 manifestos
of mainstream parties are included to test whether also the established parties became
more populist over time in order to adapt to a “populist Zeitgeist” (H1A). The regressions
result for the degree of populism for mainstream parties varies much less over time than
the respective values for all parties. The 1970s and 2000s again have somewhat lower
values, while the highest value is reached in the 2010s. The differences, however, are
generally much lower, and none of the decade differences is significant on the 5% level
whether we use the 1990s (Model 3), the 2010s (Model 4), or any other of the decade-
dummies as the baseline category.
For these reasons, hypothesis 1A cannot be
confirmed. As we will see below, the increase in populism in party manifestos is mainly
linked to non-mainstream parties.
Figure 2 presents the presence of populism in mainstream and non-mainstream parties’
manifestos in all countries over time: most notable is the rise of the average degree of
populism of non-mainstream parties, which has tripled (from 3.4% to 11.9%) in the
period between the 2000s and the 2010s. Sticking to the metaphor introduced earlier, it is
like passing from a light beer to a wine. It should be noted that non-mainstream parties
barely managed to gain more than 3% of the votes in most of the countries until the
Although the difference between the 1970s and the 2010s is very close to reaching the 5% level.
14 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
1990s (see Table 5),
but they largely contribute to raising the levels of populism in party
manifestos during the last elections.
Here it is important to keep in mind that a critique of (political) elites without reference
to the people is not counted as populist while statements in favour of the people do not
need to be combined with an explicit elite-critique in order to be populist (see
operationalization, above). This explains why mainstream parties can be populist to a
similar degree to non-mainstream parties. It presumably also explains why in the study of
Rooduijn et al. (2014) the differences between mainstream and non-mainstream parties are
much larger, since they do not count as populist statements about a common will of the
people, their virtues, or claims for more sovereignty of the people if they are not combined
with critique of the elites.
Since our conceptualization of populism is thus less restrictive than other studies in
populism (e.g. Rooduijn et al. 2014), we checked whether this difference affects the above-
reported results of Models 14. For this, for every election manifesto we counted the
percentage of statements which contain critique of elites (with and without reference to the
people) and the percentage of statements with positive evaluations of the people
We then took the lower of the two values as the dependent variable. With
1970 1980 1990 2000 2010
0 5 10 15
% populist statements
Mainstream Parties
Non−Mainstream Parties
Figure 2: Populism in Election Manifestos of Mainstream and Non-Mainstream Parties
Note: The percentages of populist statements are mean values of all five countries.
Table 5: Number of Mainstream and Non-Mainstream Parties Considered in the Analysis
1970s 1980s 1990s 2000s 2010s
Number of Mainstream Parties 17 15 15 16 17
Number of Non-Mainstream Parties 1 2 7 8 13
In the UK, with its majoritarian electoral system, non-mainstream parties obtained more than 3% of the votes
only in the 2010s elections.
Statements with reference to the people are all those called “people-centrist” and “popular sovereignty” in
Table 3. Statements with critique of elites are those called “anti-elitism” in the same table, plus statements
containing elite-critique without reference to the people (not present in Table 3, since these statements are not
considered as populist in our conceptualization of populism).
Political and Media Populism since the 1970s 15
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
this more restrictive operationalization of populism, the dependent variable has generally
lower values, but the main cross-time results do not change substantially when we run the
four regressions.
Besides the diffusion of populism in party manifestos, we hypothesized that newspapers
might have contributed to the creation of a mediatized “populist Zeitgeist” by publishing
an increasing amount of populist messages over time, in particular since the 1990s (H2).
However, the results of the regressions in Table 6 reveal that the amount of populism in
the 1990s was not significantly different from any other decade (Model 5). Only if we take
the 2010s as the reference category do we see a significant difference compared to the level
in the 2000s (Model 6). By focusing on this increase during the last decade, one might
conclude that the media spread populist discourses and amplify the politically driven
“populist Zeitgeist”. However, in the 2000s when the effects of the process of
mediatization of politics should have been striking the recorded levels are the lowest
over a time span of five decades.
Table 6: Populism in Newspaper Articles
Dependent Variable: Percentage of populist
statements in Manifestos
Model 5 Model 6
Intercept 0.82 1.29*
(0.57) (0.54)
1970s 0.04 0.51
(0.35) (0.30)
1980s 0.01 0.48
(0.36) (0.31)
1990s 0.47
2000s 0.25 0.72**
(0.33) (0.28)
2010s 0.47
Tabloid Newspapers 0.26 0.26
(0.41) (0.41)
Length (cent) 0.75*** 0.75***
(0.22) (0.22)
Editorial articles 1.30*** 1.30***
(0.27) (0.27)
AIC 22370.18 22370.18
BIC 22456.60 22456.60
Log Likelihood 11171.09 11171.09
Num. Obs. 3543 3543
Num. Groups: Newspapers 10 10
Var: Newspapers (Intercept) 0.32 0.32
Var: Residual 32.02 32.02
Note: ***p <0.001, **p <0.01, *p <0.05. Results of two-level regression models with election
manifestos nested in parties. All models contain country-dummies (not shown).
16 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
Moreover, between the 1970s and the 2010s the difference is rather narrow and it
cannot be attributable to a change of media-logic in the 1990s. It is true that “The public
debates in Western European countries have become more populist in the last two
decades” (Rooduijn 2014: 740), or at least during the last decade, but it is equally true
that the levels of populism have been decreasing during the two previous decades (from
the 1980s to the 2000s) and furthermore are generally on a very low level compared to the
election manifestos. For these reasons H2 cannot be confirmed.
To better understand the role of newspapers in spreading populist messages, we decided
to investigate whether the amount of populism present in newspaper articles reflects the
(increasingly populist) discourses of political actors (H2A), and whether journalists play an
active role in creating populism content on their own (H2B). The findings presented in
Figure 3 suggest that both hypotheses should be discarded. The u-shaped curve describing
the levels of populism articulated by journalists suggests once again the importance of
keeping in mind the big picture, and of also observing the pre-1990s era: by cutting the
figure in half, one would conclude that H2B could eventually be confirmed, yet this is not
the case. The increase since the 1990s is clear (from 0.1% to 0.8%) but it was preceded by
an almost symmetrical decline during the 1970s1990s (from 0.6% to 0.1%). Therefore,
H2B cannot be confirmed. Concerning the levels of populism articulated by political
actors, the data show that there has been no increase since the 1990s, while in the 1980s
the levels were much higher. Tellingly, the levels recorded in the 1970s and in the 2010s
also do not differ a lot (0.3% and 0.5%). Therefore, H2A cannot be confirmed either.
Finally, hypothesis 2C differentiated between tabloids and quality newspapers. The
coefficients for the tabloid dummy variable in Models 5 and 6 indicate that tabloids are
more populist on average than quality newspapers, as stated in the hypothesis. However,
the difference is clearly not significant on any common level. This finding remains true
after controlling for the length of the articles.
Furthermore, looking at the country level (Figure 4), we see that only in Switzerland is
the tabloid newspaper clearly more populist than the quality newspaper: Blick (the tabloid)
presents a much higher level of populism than NZZ (the quality newspaper). Indeed, Blick
is the newspaper with the highest level of populism of all ten newspapers over the whole
1970 1980 1990 2000 2010
0.0 0.5 1.0 1.5 2.0
% populist statements
Party Actors
Figure 3: Populism in Newspapers by Type of Speaker
Note: The percentages of populist statements are mean values of all five countries.
Political and Media Populism since the 1970s 17
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
research period (1.9%). In Austria (+0.1 percentage points) and the Netherlands (+0.3)
tabloids also display slightly higher levels of populism than quality newspapers, but in the
UK and Germany it is the other way around (0.8 percentage points for Germany and
0.9 percentage points for the UK). Consequentially, quality newspapers would be slightly
more populist than tabloids if we excluded Switzerland from the analysis.
If we look at the difference between the two types of newspapers over time (Figure 5), it
seems to be the case that quality newspapers were marginally more populist from the
1970s until the last decade, when tabloids became more populist than quality newspapers
for the first time. Also, this time-trend mainly reflects the comparably huge increase of
populist statements in the Swiss tabloid Blick. Moreover, the difference between quality
newspapers and tabloids would be reversed if we took the absolute number of populist
Quality Newspapers
Tabloid Newspapers
% populist statements
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Figure 4: Populism in Tabloids and Quality Newspapers Across Countries.
Note: The percentages of populist statements are mean values of the five decades between the 1970s and 2010s.
1970 1980 1990 2000 2010
% populist statements
Tabloid newspapers
Quality newspapers
Figure 5: Populism in Tabloids and Quality Newspapers over Time
Note: The percentages of populist statements are mean values of all five countries.
18 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
sentences per article and not their percentage. With 0.10 populist statements per article the
number is slightly higher for quality newspapers than for tabloids (0.08).
Taking all this into account, it is possible to conclude that articles in quality newspapers
and tabloids contain roughly the same amount of populism and therefore hypothesis 2C
cannot be confirmed. Future research might focus on the reasons behind these findings,
which contradict the main theories on the topic but are in line with other empirical studies
(Akkerman 2011; Rooduijn 2014). Finally, the two control variables of Models 5 and 6
are clearly significant: longer articles (as measured by word count) seem to have higher
percentages of populist statements (which was not the case for the manifestos), and
editorial articles are clearly more populist on average than normal newspaper articles. The
latter finding might explain why Roodujin (2014) found on average a higher degree of
populism in newspapers, since he mainly analysed editorial and other types of opinion
While much has been written about the electoral success of populist parties during recent
decades, the discursive dimension of populism has been largely neglected.
This paper
shows that the two dimensions do not necessarily follow the same trajectory. The period
between the 1980s and the 1990s, which is often identified as a watershed for the
‘populistization’ of Western European politics and the growing electoral success of
populist parties, does not seem to represent a critical juncture for the diffusion of populist
discourses. In fact, it is only after 20 years and the occasion of the last elections that
populism became particularly widespread in party manifestos. Moreover, it is important to
acknowledge that, in considering the bigger picture, populism appears as a cyclical
phenomenon that periodically takes the political stage but never disappears. Indeed, it is
possible to assume that populism is a constant companion of democracy (Canovan 1999).
What is yet to be fully understood is which combination of socio-political and cultural
elements triggers outbursts of populism like the one we have witnessed in recent years. In
other words: what can explain the shift from a pilsner beer (4.1% of populist sentences in
manifestos) to a Tripel Trappist beer (8.8%)?
Confirming the impression of populism as a ‘natural’ component of the public discourse,
it is worth noting that populist messages do not seem to have provoked an
‘anthropological mutation’ in Western European politics. So far, mainstream parties have
not become particularly populist, although this might change in the near future. What is
(relatively) new is the combination of parties articulating very populist discourses and the
fact that they manage to gain significant popular support.
Whether or not these parties
will remain a stable presence in the Western European political arena and whether they
will remain consistently populist remains to be seen.
Also, the alarm concerning a mediatized “populist Zeitgeist” and the idea that news
media increasingly spread populist messages could not find empirical support in this study.
One could also consider how many articles about politics there are in each newspaper’s edition. In total, it is
very likely that quality newspapers generally display more articles about politics than tabloids per day. However,
we do not include this aspect in our research.
With some commendable exceptions: among others, Hawkins (2009); Akkerman (2011); Rooudijn (2014),
Rooudijn et al. (2014).
It would be interesting to go further back in time and measure the levels of populism in the 1920s and 1930s.
Political and Media Populism since the 1970s 19
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
On the contrary, newspapers seem to curb rather than foster the diffusion of populist
discourses since the almost unchanged levels of populism in newspapers do not seem to
reflect the increasing amount of populist messages found in party manifestos. The reasons
for this phenomenon should be investigated in future research, especially since the media
are often themselves criticized in populist statements for speaking more in the name of the
elites than the people. Furthermore, future research should also include other news media
such as TV and radio shows, in order to gain a more complete picture, and also social
media such as Facebook and Twitter, to capture the essence of a fast-changing media
Several other questions remain open for future investigation. Concerning the political
dimension, it is important to observe the evolution of populist discourses in the coming
years while keeping in mind the big picture. To establish whether or not mainstream
parties are actually becoming more populist, researchers should measure populist
discourses during and outside election times. Concerning the role of the media, it is
important to explore whether news media portray populist actors positively or negatively.
All in all, the study of populist discourses in party manifestos and news media is just
beginning, and it should become increasingly comparative, such as including other regions
like Southern and Eastern Europe, Asia, and Latin America, as well as going as far back
as possible in the past to find patterns in the quantity and quality of populist discourses,
thereby identifying different waves of populist discourse across time and space.
Abts, K. and S. Rummens (2007). Populism Versus Democracy. Political Studies 55(2): 405424.
Adams, J. and Z. Somer-Topcu (2009). Policy Adjustment by Parties in Response to Rival Parties’
Policy Shifts: Spatial Theory and the Dynamics of Party Competition in Twenty-Five Post-War
Democracies. British Journal of Political Science 39(4): 825846.
Akkerman, T. (2011). Friend or Foe? Right-wing Populism and the Popular Press in Britain and the
Netherlands. Journalism 12(8): 931945.
Albertazzi, D. and D. McDonnell (2008) (eds.). Twenty-first Century Populism: The Spectre of
Western European Democracy. Basingstoke: Palgrave.
Arzheimer, K. (2015). The AfD: Finally a Successful Right-Wing Populist Eurosceptic Party for
Germany? West European Politics 38(3): 535556.
Bale, T., C. Green-Pedersen, A. Krouwel, K. Luther and N. Sitter (2010). If You Can’t Beat Them,
Join Them? Explaining Social Democratic Responses to the Challenge from the Populist Radical
Right in Western Europe. Political Studies 58(3): 410426.
Blumler, J.G. and D. Kanavagh (1999). The Third Age of Political Communication: Influences and
Features. Political Communication 16(3): 209230.
Boomgaarden, H. and R. Vliegenthart (2007). Explaining the Rise of Anti-immigrant Parties: The
Role of News Media Content. Electoral Studies 26(2): 404417.
Bornschier, S. (2010). The New Cultural Divide and the Two-Dimensional Political Space in Western
Europe. West European Politics 33(3): 419444.
Canovan, M. (1981). Populism. New York: Harcourt Brace Jovanovich.
(1999). Trust the People!Populism and the Two Faces of Democracy. Political Studies 47(1):
(2004). Populism for Political Theorists? Journal of Political Ideologies 9(3): 241252.
Cerny, P.G. (1999). Globalization and the Erosion of Democracy. European Journal of Political
Research 36(1): 126.
20 Luca Manucci and Edward Weber
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
Cox, R.W. and T.J. Sinclair (1999). Approaches to World Order. Cambridge: Cambridge University
Ellinas, A. (2010). The Media and the Far Right in Western Europe: Playing the Nationalist Card.
New York: Cambridge University Press.
Esser, F. and J. Str
ack (eds.). (2014). Mediatization of Politics: Understanding the Transformation
of Western Democracies. Basingstoke: Palgrave Macmillan.
Harmel, R. and K. Janda (1994). An Integrated Theory of Party Goals and Party Change. Journal
of Theoretical Politics 6(3): 259287.
Hawkins, K. (2009). Is Ch
avez Populist? Measuring Populist Discourse in Comparative Perspective.
Comparative Political Studies 42(8): 10401067.
(2010). Venezuela’s Chavismo and Populism in Comparative Perspective. Cambridge: Cambridge
University Press.
om, A., T. Nilsson and P. Stoltz (2012). Nationalism vs. Nationalism: The Challenge of the
Sweden Democrats in the Swedish Public Debate. Government and Opposition 47(2): 186205.
Jagers, J. and S. Walgrave (2007). Populism as Political Communication Style: An Empirical Study
of Political Parties’ Discourse in Belgium. European Journal of Political Research 46(3): 319345.
Kazin, M. (1995). The Populist Persuasion: An American History. New York: Basic Books.
Koopmans, R. (2004). Movements and Media: Selection Processes and Evolutionary Dynamics in
the Public Sphere. Theory and Society 33(3): 367391.
Kriesi, H., E. Grande, R. Lachat, M. Dolezal, S. Bornschier and T. Frey (2006). Globalization and
the Transformation of the National Political Space: Six European Countries Compared. European
Journal of Political Research 45(6): 921956.
(2008). West European Politics in the Age of Globalization. Cambridge: Cambridge University
Kriesi, H. (2014). The Populist Challenge. West European Politics 37(2): 361378.
amer, B. (2014). Media Populism: A Conceptual Clarification and Some Theses on its Effects.
Communication Theory 24(1): 4260.
Kupchan, C.A. (2012). The Democratic Malaise. Globalization and the Threat to the West. Foreign
Affairs 91(1): 6267.
Lehmann, P., T. Matthieß, N. Merz, S. Regel and A. Werner (2016). Manifesto Corpus. Version:
2016-6. Berlin: WZB Berlin Social Science Center.
Lijphart, A. (1999). Patterns of Democracy: Government Forms and Performance in Thirty-six
Countries. Yale: University Press.
March, L. (2011). Radical Left Parties in Europe. Abingdon: Routledge.
Mazzoleni, G. (2003). The Media and the Growth of Neo-Populism in Contemporary Democracies.
In Mazzoleni, G., J. Stewart and B. Horsfield (eds.), The Media and Neo-populism. A
Contemporary Comparative Analysis. Westport: Praeger (120).
(2008). Populism and the Media. In Albertazzi, D. and D. McDonnell (eds.), Twenty-First
Century Populism: The Spectre of Western European Democracy. Basingstoke and New York:
Palgrave Macmillan (4964).
(2014). Mediatization and Political Populism. In Esser, F. and J. Str
ack (eds.),
Mediatization of Politics: Understanding the Transformation of Western Democracies. Basingstoke:
Palgrave Macmillan (4256).
eny, Y. and Y. Surel (2000). Par le Peuple, Pour le Peuple. Le Populisme et les D
emocraties. Paris:
eny, Y. and Y. Surel (2002) (eds.). Democracies and the Populist Challenge. Basingstoke: Palgrave.
Mudde, C. (2004). The Populist Zeitgeist. Government and Opposition 39(4): 542563.
(2007). Populist Radical Right Parties in Europe. Cambridge: Cambridge University Press.
Political and Media Populism since the 1970s 21
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
(2013). Three Decades of Populist Radical Right Parties in Western Europe: So What?
European Journal of Political Research 52(1): 119.
Mudde, C. and C. Rovira Kaltwasser (2012) (eds.). Populism in Europe and the Americas: Threat or
Corrective for Democracy?. Cambridge: Cambridge University Press.
Papadopoulos, Y. (2002). Populism, the Democratic Question, and Contemporary Governance. In
eny, Y. and Y. Surel, Democracies and the Populist Challenge. New York: Palgrave (4561).
Roggeband, C. and R. Vliegenthart (2007). Divergent Framing: The Public Debate on Migration in
the Dutch Parliament and Media, 19952004. West European Politics 30(3): 524548.
Rooduijn, M. (2014). The Mesmerising Message: The Diffusion of Populism in Public Debates in
Western European Media. Political Studies 62(4): 726744.
Rooduijn, M., S. de Lange and W. Van der Brug (2014). A Populist Zeitgeist? Programmatic
Contagion by Populist Parties in Western Europe. Party Politics 20(4): 563575.
Rovira Kaltwasser, C. (2011). The Ambivalence of Populism: Threat and Corrective to Democracy.
Democratization 19(2): 184208.
Skocpol, T. and V. Williamson (2012). The Tea Party and the Remaking of Republican Conservatism.
Oxford: Oxford University Press.
van Spanje, J. (2010). Contagious Parties: Anti-immigration Parties and their Impact on Other
Parties’ Immigration Stances in Contemporary Western Europe. Party Politics 16(5): 563586.
Vliegenthart, R., H. Boomgaarden and J. van Spanje (2012). Anti-immigrant Party port and Media
Visibility: A Cross-party, Over-time Perspective. Journal of Elections, Public Opinion and Parties
22(3): 315358.
Weyland, K. (2001). Clarifying a Contested Concept Populism in the Study of Latin American
Politics. Comparative Politics 34(1): 122.
Wirth, W., F. Esser, M. Wettstein, S. Engesser, D. Wirz, A. Schulz, ..., C. Schemer (2016). The
Appeal of Populist Ideas, Strategies and Styles: A Theoretical Model and Research Design for
Analyzing Populist Political Communication. NCCR Democracy Working Paper No. 2016/88.
Zimmerling, R. (2005). Globalization and Democracy. Tampere: Tampere University Press.
Supporting Information
Additional Supporting Information may be found in the online version of this article:
Appendix S1: Alternative Operationalization.
Luca Manucci is a junior researcher at the NCCR Democracy and PhD candidate at the University of Zurich.
His research focuses on populism in Western Europe, political parties, and mediatization of politics. Address for
correspondence: Affolternstrasse 56, 8050 Zurich, Switzerland. Email:
Edward Weber is a junior researcher at the NCCR Democracy and PhD candidate at the University of Zurich.
His research focuses on populism in Western Europe and political parties in Switzerland. Address for
correspondence: Affolternstrasse 56, 8050 Zurich, Switzerland. Email:
©2017 Swiss Political Science Association Swiss Political Science Review (2017)
22 Luca Manucci and Edward Weber
... Na druhou stranu je politický program základní prostředek komunikace politických stran a strany v něm stanovují a rámují zásadní témata (navzdory novým formám komunikace a politického marketingu vyplývajícím například z možností sociálních sítí), analýzy stranických programů (i ve vztahu k populismu) mají rozsáhlou tradici [např. Rooduijn, Pauwels 2011;Manucci, Weber 2017] a i v současnosti představují jeden z nejčastějších typů politologických analýz. Zároveň jde v českém kontextu o jediný typ dat, který (vzhledem k dlouhému časovému období a dřívější neexistenci možností vyplývajících z existence webu 2.0) lze získat pro všechny sledované aktéry napříč celým postkomunistickým vývojem. ...
Full-text available
The article examines how use of the topic of corruption in the election manifestos of Czech political parties has transformed over the course of the entire post-communist period. The theoretical framework used to grasp the anti-corruption positions expressed in the manifestos is populism. The study builds on the discussions about mainstream political actors (not) adopting populist principles and on the methodological debate around the possibilities offered by populist communication research. Using a content analysis and the principle of triplets, the article analyses the election manifestos of all the parties that were successful in the parliamentary elections between 1990 and 2017 in order to determine whether there was a stronger presence of pop-ulist principles in communication on the topic of corruption. Among other things, the study shows that: (1) statements on corruption that are consistent with a populist communication strategy can be found in parties that are not commonly understood as populist throughout the entire researched period; (2) mainstream parties did not respond to the success of populists in the 2010, 2013, and 2017 elections by adopting populist principles; (3) the populists who won a share of power strengthened the elements of populist communication in the next election manifesto.
... The literature agrees that RWPPs tend to express nativist and conservative values that bring forth an exclusionary vision of the nation (among others, see Betz, 1994;Norris & Inglehart, 2019;Skenderovic, 2007;, conceptualized as an "imagined community" (Anderson, 2006). In a Manichean outlook, RWPPs represent the people as a homogeneous group based on a nativist conception of the nation (Manucci & Weber, 2017;. By creating a division between an idealized us and the dangerous others (Taggart, 2000), RWPPs manage to create a romanticized construction of the past through historical, traditional, religious and cultural values. ...
Full-text available
Right-wing populist parties (RWPPs) view immigration as a threat to national identity and cultural and political cohesion. This paper explores the discourse on immigration of two such parties during a systemic crisis (the COVID-19 pandemic) in a borderland area that entails a specific “symbolization” of the border as a result of geographic proximity. What kind of bordering narratives occur in this context, and how do they evolve? To answer these questions, the major RWPPs operating in the Geneva region were studied: the Genevan Swiss People's Party and the Geneva Citizens' Movement. Through a critical discourse analysis, a total of 181 documents published between 1 January and 31 August 2020 were analyzed, including the parties' official Facebook posts, press releases, and newspaper articles. Results show that the context of the pandemic favored the emergence of a strong re-bordering narrative.
... Para contrastar las hipótesis de la investigación se ha llevado a cabo un análisis de contenido de los programas electorales presentados a nivel nacional desde las elecciones europeas de 2014 hasta las últimas elecciones europeas de 2019. En primer lugar, como muchos estudios en la materia ponen de manifiesto, los programas electorales son documentos oficiales, escritos, accesibles y comparables entre sí, que ofrecen una descripción completa y estructurada de las posiciones de los partidos políticos en diferentes asuntos, así como de las grandes líneas ideológicas que siguen estas organizaciones (March, 2017;Manucci y Weber, 2017;Rooduijn y Akkerman, 2015;Rooduijn et al., 2014;Fernández-García y Luengo, 2018a). En segundo lugar, esta investigación busca analizar la evolución que ha seguido los principales elementos del populismo en los programas electorales de la formación. ...
Full-text available
Este artículo de investigación tiene como objetivo analizar la evolución que ha seguido el mensaje populista de Podemos. Para ello, se ha realizado un análisis de contenido de los programas electorales presentados por el partido político desde su fundación en 2014 hasta las elecciones europeas de 2019. Los resultados sugieren que la presencia de elementos populistas en los programas electorales de la formación no sigue una tendencia claramente descendente como consecuencia de la mayor presencia del partido en las instituciones legislativas y ejecutivas del país. No obstante, se ha identificado una evolución del contenido de los mensajes populistas por la cual, a) estos siguen un encuadre más económico que político, y b) el enemigo queda retratado en términos ideológicos, rompiendo así con la estrategia de «desdiferenciación» con la que surgió la formación en sus inicios: «las tres derechas» españolas, el «eje reaccionario» europeo y el neoliberalismo dominante en Europa son identificados ahora como la gran amenaza para los intereses del pueblo, mientras que el PSOE queda configurado como una formación ideológicamente diferenciada. Estos resultados sugieren, por tanto, que el eje nueva/vieja política que irrumpió con fuerza entre 2014 y 2016 podría considerarse eclipsado en la actualidad por los ejes tradicionales de la política española izquierda/derecha y centro/periferia.
... As Manucci et al. (2017) explain, a key aspect of populism is that it is the kind of politics best adapted to the media: it has a pure commercial logic in its electoral view, it adjusts itself and its message to the market, and its simple and personalistic messages work well on existing media channels. In other words, populism thrives by being the centre of attention. ...
Full-text available
Populism has become a major factor in understanding politics in Europe and the rest of the world, giving birth to several interpretations and explanations of the phenomenon. In particular, Spain’s political system has been radically changed by it, while in France its importance seemed to increase with each election between 2016 and 2020. Rather than focusing on economic, purely discoursive or strategic understandings of populism, this thesis focuses on the ideational approach, treating populism as a separate, albeit thin-centred ideology (one which does not offer a complete worldview, and needs to attach additional ideological tenets, from socialism to nationalism). From this, it analyses both the supply of populism – which parties can be considered populist in Spain and France; and the demand for it – who are the voters of these parties, how varied are they and how different are they from mainstream party supporters? It also searches for a match between the supply and demand of populism, and looks for a definition of the populist voter. Both dictionary-based computerised content analysis and discourse analysis is used to identify the parties, based on both written and spoken material. For demand, electoral surveys of the relevant periods are examined by means of both frequency analysis and logarithmic regressions. The findings reveal four parties of interest for this research: two social populist (Podemos and La France Insoumise) and two populist radical right (Vox and Rassemblement National). Voter analysis shows a common core of dissatisfaction with democracy, political extremism and concern for social issues. Beyond this, voters for populist radical right parties also share a hostility to immigration and a display of ethno-nationalism. However, in the Spanish case voters appear aligned mostly along the left-right cleavage, with Podemos and Vox occupying both extremes of the divide. In the French case, while La France Insoumise attracts mostly traditional left-wing voters, Rassemblement National has instead a variety of sources of electoral support, also from the centre and the left, an eclectic electorate that is hostile to immigration but in favour of state control of the economy and redistribution. The thesis finds a clear match between supply and demand in the Spanish case, as well as in the case of Rassemblement National. The case of La France Insoumise is more complex, with its voters not following the party’s more populist positions, such as its Euroscepticism. The found common core, and the specifics of Rassemblement National, lead to the conclusion that the populist voter as such can indeed exist, not only as a subordination to the ideological vote, but in and of itself, provided there is sufficient disintegration of the traditional left-right party system and the domination of the global v. national cleavage, something that is not (yet) seen in Spain, but that is present in France
Populism has become a widely used concept in both academia and the media. The term’s popularity has encouraged scholars to question how it is applied and to theorize on the consequences of its use. However, there is little empirical research on the temporal and cross-country changes in the use of the term in the public sphere. This article analyses the significations given to the terms ‘populism’ and ‘populist’ in six countries’ daily newspapers over a period of nearly two decades. It presents the results of a quantitative content analysis of texts ( N = 3252) published in legacy daily papers in Finland, Sweden, the United States, the United Kingdom, the Netherlands and Turkey in the years 2000–2018. The article shows how the salience, meanings and perceived repercussions of ‘populism’ change over time and vary between the countries. The study reveals how, towards the end of the 2010s, the term is increasingly used in the context of right-wing populism and as a reference to political ideas that are detrimental to democracy. The results are examined in the context of developing academic discussions regarding the effects of ‘populism’ becoming a ubiquitous signifier in the media.
This paper assesses how parties strategically vary their populist positions in party competition. The useful conceptualization of populism as a matter of degree has been established by previous studies. However, we know little about the conditions that drive parties to alter their degree of populism and, importantly, whether this varies between mainstream and populist parties. This paper analyzes the impact of vote and office loss on parties’ populist position-taking. It draws upon German and Austrian parliamentary debate speeches from the 1990s until 2018. Populism is measured using a dictionary method, an automated text analysis approach. The results show that parties react to contextual incentives by altering their degree of populism. Mainstream and populist parties are more populist when in opposition. Electoral losses affect mainstream and populist parties less clearly. The contribution of this paper is to integrate research on populism with a theoretical framework on party competition.
In this article, we explore the consequences of the increasing presence of both left- and right-wing populist parties in government, critically reflecting on the recent scholarship on the topic, underlining promising venues for future research and outlining a conceptual framework which constitutes the background of this special issue entitled ‘Populism in Power and its Consequences’. Our main contribution is empirical, since – by reflecting on the various articles hosted in the special issue – we assess the impact of populist parties in government on politics, polities and various policy domains. We also provide an account of potential moderating factors of the influence of populists in government, focus on different ideological underpinnings of types of populisms (left-wing and right-wing) and discuss their relevance. We conclude by identifying four possible scenarios for European populist parties in governments: radicalization, compromise and moderation, splintering, or loss.
Full-text available
A great deal of scholarly attention has been devoted to the success of populist actors around the world. However, to a lesser extent, the focus has been directed towards the way in which populist political communication is used by other parties and how populist political communication has developed over time. Hence, this article takes a longitudinal approach to examine how Finnish party leaders have used populist communication during four parliamentary elections and if populist communication is contagious. This study argues that populism is best understood as a discursive frame, combined with the necessary element of people centrism with antielitism and/or exclusion of ‘dangerous others’. Manual content analysis has been used to measure the usage of populism in Finnish party leader interviews during the parliamentary elections between 2007 and 2019. The analysis reveals that several party leaders used populist political communication and that an increase in its overall usage occurred during the 2011 parliamentary election. Most commonly, populism was used by the Finns Party's leaders and by leaders of other niche parties in the opposition. However, in the 2011 parliamentary election, certain mainstream parties also used populist political communication. The article critically examines the idea of a populist Zeitgeist and the contagious effect of populist political communication as an explanation for the significant increase in populist political communication. The article suggests that populist communication spreads to certain parties, but to speak of a full‐blown populist Zeitgeist seems exaggerated in a consensus‐seeking multiparty democracy.
Full-text available
This working paper approaches the study of populism from an explicit political communication perspective and thereby adds a new and original dimension to the existing literature. It first reviews the existing approaches to populism and then combines them within an integrative framework for populism research. This framework understands populism as ideology, strategy, or style and arranges them in a way that permits the empirical analysis of populism in political communication, media and the attitudes of the people. The framework serves as a guide for developing straightforward instruments to identify and quantitatively measure populism in the self-presentation of actors, in news media and other forms of communication, as well as in public opinion. By virtue of the broadness of the framework, it is applicable to longitudinal and cross-cultural analyses of populism, as well as the investigation of the flow of populist ideology among political actors, the media, and the public. It takes a non-normative and primarily analytical perspective. The working paper has two major aims. First, it is intended to provide a common ground for populism research within Research Module II of the NCCR Democracy at the University of Zurich. Second, it is intended to distribute our framework for investigating the relationship between populism and political communication to other scholars and thereby facilitate cumulative research.
Full-text available
The European political landscape of the last decade has been home to numerous political figures that have stood out by virtue of their personality and their voicing of popular discontent. These include the likes of Jean-Marie Le Pen, Jörg Haider, Christoph Blocher, Pim Fortuyn and Silvio Berlusconi, all of whom are among the more recent manifestations of the populist political climate affecting much of contemporary Europe, as discussed in the introduction to this volume.
There is now a substantial body of research on populism which focuses, in particular, on what motivates people to support successful national-populist parties, such as ‘Verdrossenheit’ (frustration) with established parties (Schedler 1996), reactions against the success of ‘left-libertarian’ movements (Kitschelt and McGann 1995), or the concern of the losers in the modernisation process triggered by internationalisation (Kriesi 1999), and the fears of those negatively affected by post-industrial forms of social organisation (Ignazi 1999). In contrast to these important empirical findings, this chapter is guided by more theoretical and conjectural considerations. As it is reductionist to consider populism as either a pathology of, or a threat to, democracy, I shall focus on the ambivalent link between populism and democracy, which in turn requires a certain familiarity with major debates taking place in democratic theory. In this analysis I will attempt to formulate new hypotheses as to the origins of populist successes, and possible links with the transformations taking place in democratic decision-making, which in turn requires an understanding of the lessons of policy analysis.
What has happened to the European radical left after the collapse of the USSR? How has it reacted, reformed, even revived? This new volume is one of the first to provide an overview of the main developments in contemporary European radical left parties (those defining themselves as to the left of, and not merely on the left of social democracy), which are now an increasingly visible phenomenon in European party politics. Unlike many of the existing studies it focuses on communist and non-communist parties, addresses their non-parliamentary and international activity, and takes a pan-European perspective, focusing on both Eastern and Western Europe.
Populism is best understood as a Manichaean worldview linked to a characteristic language or discourse. Chavismo, the movement that sustains Hugo Chávez in Venezuela, is a paradigmatic instance of populism. Using a novel, cross-country dataset on populist discourse, combined with extensive data from within Venezuela and across other countries, this book demonstrates that populist movements can be understood as responses to widespread corruption and economic crisis. The book analyzes the Bolivarian Circles and government missions in Venezuela, revealing how populist ideas influence political organization and policy. The analysis provides important insight into the nature of populism, including its causes and consequences, and addresses broader questions about the role of ideas in politics.