Worse than the bugs was the lack of proper scriptability. Pointing and clicking  loses its appeal when you need to run the same model on 12 different datasets, or when you are looking at three variants of the same model and 10 recodes of the same variable. Throw in the desire semi-automatically re-compile the findings from these exercises into two nice tables for inclusion in again and again after finding yet another problem with a model, and you will agree that any  piece of software that is not scriptable is pretty useless for scientists.

MLwiN’s command language was unreliable and woefully underdocumented, and everything was a pain. So I embraced xtmixed when it came along with Stata 9/10, which solved all of these problems.

runmlwin presentation (pdf)

But xtmixed is slow with large datsets/complex models. It relies on quadrature, which is exact but computationally intensive. MLwiN works with approximations of the likelihood function (quick and dirty) or MCMC (strictly speaking a Bayesian approach, but people don’t ask to many questions because it tends to be faster than quadrature). Moreover, MLwiN can run a lot of fancy models that xtmixed cannot, because it is a highly specialised program that has been around for a very long time.

Enter the good people over at the Centre for Multilevel Modelling at Bristol, who have come up with runmlwin, an ado that essentially makes the functionality of MLwiN available as a Stata command, postestimation analysis and all. Can’t wait to see if this works with Linux, wine and my ancient binaries, too.

One common problem with this type of analysis is that not all municipalities are created equal. There is a surprisingly large number of flyspeck villages with only a few dozen voters on, whereas the state’s capital boasts more than 140,000 registered voters. Most places are somewhere in between. Having many small municipalities in the regression feels wrong for at least two reasons. First, small-scale changes of political preferences in tiny electorates will result in relatively large percentage changes. Second, the behaviour of a relatively large number of voters who happen to live in a small number of relatively large municipalities will be grossly underrepresented, i.e. the countryside will drive the results.

My PhD supervisor, who did a lot of this stuff in his time, used to weigh municipalities by the size of their electorates to deal with these problems. But this would lead to pretty extreme weights in my case. Moreover, while voters bring about electoral results, I really don’t want to introduce claims about individual behaviour through the back door.

My next idea was to weigh municipalities by the square root of the size their electorates. Why? In a sense, the observed behaviour is like a sample from the underlying distribution of preferences, and the reliability of this estimate is proportional to the square root of the number of people in a given community. But even taking the square root left me with weights that were quite extreme, and the concern regarding the level of analysis still applied.

Then I realised that instead of weighing by size, I could simply include the size of the electorate as an additional independent variable to correct for potential bias. But this still left me exposed to the danger of extreme outliers (think small, poor, rural communities where the number of Green voters goes up from one to four, a whopping 300 per cent increase) playing havoc with my analysis. So I began reading up on robust regression and its various implementations in Stata.

The basic idea of robust regression is that real data are more likely than not a mixture of (at least) two mechanisms: the “true model” whose coefficients we want to estimate one the one hand, and some other process(es) that contaminate the data on the other. If these contaminating data points are far away from the multivariate mean of the x-Variables (outliers) and deviate substantially from the true regression line, they will bias the estimates.

Robust regression estimators are able to deal with a high degree of contamination, i.e. they can recover the true parameters even if there are many outliers amongst the data points. The downside is that the older generation of robust estimators also have a low efficiency (the estimates are unbiased but have a much higher variance than regular OLS-estimates).

A number of newer (post-1980) estimators, however, are less affected by this problem. One particular promising approach is the MM estimator, that has been implemented in Stata ados by Veradi/Croux (MMregress) and by Ben Jann (robreg mm). Jann’s ado seems to be faster and plays nicely with his esttab/estout package, so I went with that.

The MM estimator works basically by identifying outliers and weighing them down, so it amounts to a particularly sophisticated case of weighted least squares. Using the defaults, MM claims to have 85 per cent of the efficiency of OLS while being able to deal with up to 50 per cent contamination. As you can see in the table, the MM estimates deviate somewhat from their OLS counterparts. The difference is most pronounced for the effect of tax receipts (hekst).

robreg mm has an option to store the optimal weights. I ran OLS again using these weights (column 3), thereby recovering the MM estimates and demonstrating that MM is really just weighted least squares (standard errors (which are not very relevant here) differ, because robreg uses the robust variance estimator). This is fascinating stuff, and I’m looking forward to a forthcoming book by Jann and Veradi on robust regression in Stata (to be published by Stata Press in 2012).

                     OLS              MM            WLS  

greenpct2006        0.193***        0.329***        0.329***
                 (0.0349)        (0.0592)        (0.0278)   

hekst               0.311***        0.634***        0.634***
                 (0.0894)         (0.124)        (0.0688)   

senioren          -0.0744***       -0.100***       -0.100***
                 (0.0131)        (0.0149)       (0.00994)   

kregvoters11      -0.0125        -0.00844        -0.00844
                 (0.0146)       (0.00669)       (0.00982)   

kbevdichte         -0.433        -0.00750        -0.00750
                  (0.464)         (0.330)         (0.326)   

uni                 1.258           0.816           0.816
                  (1.695)         (0.765)         (1.137)   

lnunidist          -0.418**        -0.372**        -0.372***
                  (0.127)         (0.113)        (0.0918)   

_cons               8.232***        7.078***        7.078***
                  (0.627)         (0.663)         (0.461)

As an aside, it’s remarkable that this book was published in English. The volume as well as the conference on which it is based were sponsored by French and German institutions. A few years ago, that would have meant a bilingual conference and publication. Outside Luxembourg, what is the number of scholars working in the field who could have actively participated in the conference? And how much larger would have been the number of potential readers? Individually and collectively, French and German political science might still be too big to fail for the time being, but it’s good to see that we as a discipline chose relevance. Occasionally.

To celebrate this moment of pre-Christmas clarity, here’s the author’s version of my chapter

Electoral Sociology: Who Votes for the Extreme Right and
why – and when?

This chapter profiles the social base of electoral support for the parties of the Extreme
Right¹ in Western Europe, i.e. the question of whether some groups in society are
more susceptible to the appeal of these parties than others. This issue is
relevant for a number of reasons: First, by looking at the social composition of
European societies we might be able to better understand why parties of
the Extreme Right are more successful in some countries than in others.
Second, a careful analysis of the link between the social and the political
might help us to gauge the potential for future right-wing mobilisation in
countries which currently have no electorally successful parties of the Extreme
Right. Third, knowing who votes for a party might help us to get a clearer
understanding of the underlying motives to cast a vote for the Extreme
Right.

The final version of this paper appears in Backes/Patrick Moreau (Eds.): The Extreme Right in Europe. Current Trends and Perspectives, p. 35-50. Göttingen 2012. This review chapter is also available as a PDF.

Over the last 15 years or so, analyses of the Extreme Right’s electorate(s) have
become a minor industry within the larger context of (comparative) Political
Sociology. By necessity, this chapter aims at summarising the main findings from this
research program, but cannot strive for a comprehensive presentation of all that has
been achieved during these years. More specifically, findings from national
and small-n studies are (almost) completely ignored. Much by the same
token, I will not delve into the fascinating literature on the social bases of the
Interwar Extreme Right in Germany and in other countries (Childers, 1983;
Falter, 1991; King, Tanner and Wagner, 2008; Küchler, 1992; O’Loughlin,
2002).

Recent events in Central and Eastern Europe (Mudde, 2005) provide a fascinating
complement to this Western perspective. However, much like Central and Eastern
European parties and electorates themselves, our (comparative) knowledge of the
social base of the Extreme Right in CEE in still very much in flux. Therefore, the
chapter aims to provide a comparative perspective on developments in West
European electoral politics since the 1980s.

1 Theory

1.1 Definitions

Much of the early literature on the Extreme Right is devoted to the twin debates on
the correct label and on criteria for membership in this party family. Initially, the
newly successful parties of the “Third Wave” that began in the late 1970s were

seen as closely linked to the Extreme Right of the Interwar years (Prowe,
1994). While such connections do exist in many cases, scholars soon began
to pinpoint the differences between a) the current and the Interwar right
and b) between different members of the emerging new party family. As a
result, scholars came up with a plethora of definitions, typologies and labels,
including (but not limited to) the “New Right”, “Radical Right”, “Populist
Right” and “Extreme Right”, to mention only the most popular ones. As
recently as 2007, Cas Mudde (Mudde, 2007, pp. 18-24), one of the most prolific
scholars in this area, made an attempt to bring a semblance of order to
the field by suggesting that “nativism”, the belief that states should be
inhabited exclusively by members of the “native” group, is the largest common
denominator for the parties of the Third Wave including those in Central and
Eastern Europe. Like a Russian doll, this family contains two subgroups
which are nested into each other: Parties of the “Radical Right” combine
nativism and authoritarianism, whereas the “Populist Radical Right” add
populism as an additional ingredient to this mixture. In a departure from
his earlier work, the label “Extreme Right” is reserved for anti-democratic
(extremist) parties (Mudde, 2007, p. 24) within the all-embracing nativist
cluster.

While Mudde’s proposal is remarkably clear and was very well received in the
field,²
it matters most to students of parties. Scholars of voting behaviour, on the other
hand, tend to go with a rather pragmatic approach that was concisely summarised by
Mudde (Mudde, 1996, p. 233) a decade earlier: “We know who they are, even though
we do not know exactly what they are.” As this quote suggests, there is (definitional
questions not withstanding) actually a very broad consensus as to which
parties are normally included in analyses of the Right’s electoral base. These
include the Progress Party in Norway, the Danish People’s Party and the
Progress Party in Denmark, New Democracy and the Sweden Democrats in
Sweden, the National Front, National Democrats and British National Party
in Britain, the National Front and the National Republican Movement in
France, the German People’s Union, Republicans and National Democrats in
Germany, the Centre Parties, Lijst Pim Fortuyn and the Freedom Party in the
Netherlands, the Vlaams Blok/Belang and the National Front in Belgium, the
Freedom Party and the Alliance for the Future in Austria, the Italian Social
Movement/National Alliance, the Northern League and the Tricolour Flame in Italy,
the Falange Parties in Spain, Political Spring, the Popular Orthodox Rally
and various smaller and short-lived parties in Greece, and the “Christian
Democrats”(PDC) in Portugal. There is even a remarkable agreement on which
parties should best be seen as borderline cases: the Scandinavian Progress
Parties before they transformed themselves into anti-immigration parties
during the early 1980s, the National Alliance after Fini began to develop its
“post-fascist” profile in the mid-1990s, the Swiss People’s Party in Switzerland
before it became dominated by its “Zurich Wing” lead by Blocher and the
True Finns in Finland and the Social Democratic Centre/Popular Party in
Portugal.

Amongst scholars of voting behaviour, there is little doubt that these parties
attract similar voters and should be grouped together in a single, albeit very
heterogeneous, party family. “Extreme Right” is currently the most popular label for
this group. Its use does not (necessarily) signify the respective parties’ opposition to the principles of liberal democracy but rather adherence to a convention in the
field.

This is not to imply that differences between these parties do not exist, do not
matter for voting behaviour or should be analysed by different typologies. The
German NPD, for instance, is unapologetically neo-fascist, whereas the Norwegian
Progress Party is, at least on the surface, remarkably moderate and libertarian.
Rather, it is next to impossible to incorporate the existing differences between parties
into studies of voting behaviour, because it is very rare to concurrently observe two
or more electorally viable parties of the Third Wave competing for votes. Therefore,
party sub-type effects are inseparable from constant and time-varying country
effects.

1.2 Explanations

Over the last eight decades or so, historians, sociologists and political scientists have
developed a multitude of theoretical accounts that aim to explain the electoral
support for the Interwar and modern Extreme Right. While many of these accounts
are highly complex, they can usefully be grouped into four broad categories (Winkler,
1996).

A first group of scholars focuses on largely stable and very general attributes of the
Extreme Right’s supporters, that is, personality traits and value orientations. The
most prominent example of this line of research is without doubt the original study of
the so-called “Authoritarian Personality’s” support for the Nazi party by Adorno and
his collaborators (Adorno et al., 1950). More recent contributions include work by
Altemeyer and Lederer, who both aim at developing “modern” scales for measuring
authoritarianism.³
In a related fashion, authors like Ignazi and Kitschelt (Ignazi, 1992; Kitschelt, 1995)
have proposed a link between allegedly stable value orientations and voting for the
Extreme Right. Both authors interpret the success of the Extreme Right as part of a
authoritarian-materialistic “backlash” against the Green and Left-Libertarian
parties that emerged from the New Social Movements of the 1970s (Inglehart,
1977).

If there is a correlation between one’s social position on the one hand and one’s
personality traits and value orientation on the other, these approaches should go
some way towards identifying the electoral base of the modern Extreme Right. And
indeed, ever since the first studies on the social bases of the original Nazi movement
were published (See e.g. Parsons, 1942), social scientists have suspected that the
working class, the lower middle-classes and particularly the so-called “petty
bourgeoisie” exhibit stronger authoritarian tendencies than other social groups. This
alleged link between class (and, by implication, formal education) was made explicit
by Kitschelt (Kitschelt, 1995, pp. 4–7), who argued that the very nature of
jobs in certain segments of the private sector predisposes their occupants
towards a mixture of market-liberal and authoritarian ideas that was at one
stage promoted by the National Front in France and the Freedom Party in
Austria.

A second strand of the literature is mainly concerned with the effects
of social disintegration, i.e. a (perceived) break-down of social norms
(“anomia”) and intense feelings of anxiety, anger and isolation brought about

by social change. Allegedly, this mental state inspires a longing for strong
leadership and rigid ideologies that are provided by the Extreme Right. A
classic proponent of this approach is Parsons in his early study on the Nazi
supporters. More recently, these ideas have returned in the guise of the “losers of
modernisation” hypothesis, i.e. the idea that certain segments of Western societies
feel that their position is threatened by immigration and globalisation and
therefore turn to political parties which promise to insulate them from these
developments.⁴
Interestingly, the losers of modernisation hypothesis identifies more or less the same
social groups – (unskilled) workers, the unemployed and other persons depending on
welfare, parts of the lower middle classes – as the main target of Extreme Right
mobilisation efforts.

A third class of accounts draws heavily on theories from the field of social
psychology. In this perspective, group conflicts are the real cause of support for the
Extreme Right. Unlike the two aforementioned approaches, this strand is
relatively heterogeneous. At one end of the spectrum, it includes classic
theories of purely emotional, hardly conscious scapegoating (See e.g. Dollard
et al., 1939). In this perspective, ethnic minorities including immigrants
provide convenient targets for the free-floating aggression harboured by a
society’s underclass. These minorities are at the same time a) suitably different
from and b) even more power- and defenceless than the members of this
group.

At the other end of the spectrum, theories of Realistic Group Conflict
that can be traced to the early work of Sherif and Sherif (See e.g. Sherif
and Sherif, 1953) emphasise the role of a (bounded) rationality in ethnic
conflicts over scarce resources like jobs and benefits. This idea is especially
prominent in more recent accounts (E.g. Esses, Jackson and Armstrong,
1998).

Theories of “ethnic competition” (Bélanger and Pinard, 1991), “status politics”
(Lipset and Bendix, 1951), “subtle”, “modern”, “symbolic” or “cultural” racism
(Kinder and Sears, 1981) and social identity (Tajfel et al., 1971) cover a middle
ground between these two poles, while the notion of “relative deprivation” – the idea
that one’s own group is not getting what they are entitled to in comparison with
another social group – provides a useful conceptional umbrella for these somewhat
disparate ideas (Pettigrew, 2002).

Again, no matter what specific concept from this research tradition is applied,
again, the usual suspects emerge: those social groups who deem themselves
threatened by immigration and related processes. But not all members of these
groups vote for the Extreme Right. Rather, the Extreme Right vote shows a
considerable degree of variation both between and within countries in Western
Europe. Some of the differences between countries might be explained by differences
in the social composition of the respective societies. However, these differences cannot
explain the huge differences in Extreme Right support between otherwise
reasonably similar countries: Norway is hardly more deprived than its neighbour
Sweden. By the same token, it is difficult to imagine that the authoritarian
underclass in Austria is six or seven times larger than its counterpart in
neighbouring Germany. Moreover, personality traits, value orientations, group
membership and even social and economic position change slowly, if at all, whereas
support for the Extreme Right often exhibits a great deal of variability within
countries.

One factor that is often overlooked, perhaps because it seems too obvious, is the
core variable of the social-psychological model of voting, i.e. party identifications.
Historically, West European parties of the centre left and the centre right have been
able to absorb considerable authoritarian potentials in their respective societies, and
even today, some voters who might otherwise be lured by the Extreme Right are
simply not available for those parties because they are still firmly attached to one of
the more established parties (Arzheimer and Carter, 2009a). Similarly, ties to other
organisations, notably churches and trade unions, are likely to reduce the
probability of an Extreme Right vote. This implies that the ongoing processes of
de-alignment in West European societies (Dalton, Flanagan and Beck, 1984)
will increase the potential for right-wing mobilisation, everything else being
equal.

However, varying degrees of de-alignment are not the only differences between
West European societies that can help to explain levels of support for the
Extreme Right. Moreover, party identifications are also supposed to be stable
over time. Therefore, processes of de-alignment and re-alignment cannot
explain short-time fluctuations of Extreme Right support within the same
country.

These insights have triggered interest in a fourth, additional perspective that has
come to the fore in recent years and aims to complement the three major approaches.
In Winkler’s original survey of the literature, this emerging perspective was presented
under the label of a “political culture” that constrains the posited effects of
individual factors on the Extreme Right vote. However, since the mid-1990s, interest
in a whole host of other, more tangible contextual factors has grown tremendously,
and it is now widely believed that the interplay between group conflicts and
system-level variables can help explain the striking differences in support for the ER
over time and across countries. Building on previous work by Tarrow and Kriesi and
his associates (Kriesi et al., 1992; Tarrow, 1996), Arzheimer and Carter have argued
that these factors should be subsumed under the concept of “political opportunity
structures”, which compromise short-, medium- and long-term contextual
variables that amongst them capture the degree of openness of a given political
system for political entrepreneurs (Arzheimer and Carter, 2006, p. 422).
As it turns out, however, the concept of “opportunities” for new political
actors might be too narrow: Many context factors like unemployment or
immigration will not only provide the political elite with an incentive to mobilise,
but will also have a direct and possibly more important impact on voters’
preferences. Empirically, it is not possible to separate these two causal mechanisms
since we have no reliable information on the mental calculations made by
(would-be) politicians. Therefore, it seems reasonable to subsume the notion of
opportunity structures under the even more general concept of contextual
factors.

Over the last 15 years or so, studies have looked at a whole host of such
contextual variables, including but not limited to:

1. Opportunity structures
   1. In a strict sense: political decentralisation and electoral thresholds
      (E.g. Carter, 2005)

• In a wider sense: positions of other parties (Arzheimer, 2009;
  Arzheimer and Carter, 2006; Lubbers, Gijsberts and Scheepers,
  2002), media coverage (Boomgaarden and Vliegenthart, 2007, 2009)
  and “discursive opportunity structures” (Koopmans and Muis, 2009;
  Koopmans and Olzak, 2004; Wal, 2000; Wimmer, 1997)
• Variables related to the Extreme Right parties themselves (e. g. availability of
  “charismatic leaders”, policy positions, reliance on populism, party
  sub-type)
• Macroeconomic variables: unemployment, growth, and their trends
• Other political variables: immigration figures

All accounts of the role of contextual variables assume – sometimes explicitly but
more often implicitly – some sort of multi-level explanation in the spirit of Coleman’s
ideal type of sociological explanations (Coleman, 1994). Put simply, these
explanations assume that changes at the macro-level (a declining economy, rising
immigration figures, a new anti-immigrant party) bring about changes in individual
preferences, which lead to (aggregate) changes in individual political behaviour, i.e.
an increase in electoral support for the Extreme Right. Since different groups in
society have different prior propensities to vote for the Extreme Right, and since they
react differently to changes in the social and political environment, both micro and
macro information are required to fully model and understand the processes that
transform latent or potential support for the Extreme Right into real, manifest
votes

2 Data

All empirical analyses of the nexus between the social and the political require data,
which fall into two broad categories: aggregate (macro) data which provide
information on the behaviour and properties of collectives (electoral districts,
provinces, countries …), and micro data, which relate to individuals and are typically
based on standardised interviews. Both categories can be further subdivided by
including additional dimensions:

1. Macro data
   1. Source: census data, electoral results, macro-economic and
      government data
   2. Temporal coverage: cross-sectional vs. longitudinal data
   3. Geographical coverage: one, few or many countries
   4. Level of aggregation: wards, constituencies, subnational units or the
      whole country
2. Micro data
   1. Source: national opinion polls vs. comparative multi-national studies
   2. Temporal coverage: cross-sectional, trend and panel studies
   3. Geographical coverage: one, few or many countries
   4. Level of aggregation: individual cases vs. aggregated survey results

The analytical leverage of the data depends on these sub-dimensions as well as on
the reliability of the information and the level of detail they provide. As a result of
technological progress and huge individual and collective investments into the
infrastructure of social science research, the quality and availability of comparative
data on the electorates of the Extreme Right in Western Europe have vastly
improved over the last decade. Consequentially, scholars of the Extreme Right are
nowadays in a much better position to analyse the social base of these parties than
fifteen or even five years ago.

Nonetheless, they still face some awkward trade-offs. Generally speaking,
micro-level data is preferable to macro-level data, especially if the level of
aggregation is high. After all, aggregate measures are usually restricted to human
behaviour but provide no information on the motives behind the aggregated
actions.⁵

Moreover, aggregation discards individual information. Therefore, inferences from
correlations at the macro-level to the behaviour of individuals are plagued by the
infamous ecological fallacy (Robinson, 1950) unless the aggregates are homogeneous.
This is most easily illustrated by an example: At the level of the 96 departments
of metropolitan France, there is a sizable positive correlation between the
number of foreign-born persons and the vote for the National Front. It is,
however, highly unlikely that immigrants have an above-average propensity to
vote for the Extreme Right. Rather, the aggregate correlation reflects a
mixture of a) the below-average propensity of immigrants to vote for the Front
National⁶
and b) a hostile reaction of other voters to the presence of immigrants.
Without individual-level data, it is not possible to disentangle these two
effects.⁷

A famous historical example for the perils of aggregate correlations concerns two
time-series that moved in sync: electoral support for the NSDAP and the
unemployment rate in Weimar Germany. Their positive relationship suggests that the
unemployed turned to the Nazi party as their economical situation declined (Frey

and Weck, 1981). However, at lower levels of aggregation (Länder and Kreise), the
relationship between unemployment and the NSDAP vote was actually negative.
Presumably, the unemployed were less likely to vote for the NSDAP while those who
(yet) had a job had a higher propensity to support the Nazis that further
increased as the economy deteriorated (Falter and Zintl, 1988; Falter et al.,
1983).

So why would anyone want to base their analyses on macro data? As it turns out,
quite often there is no alternative, because (comparable) surveys were simply not
conducted at some point in time relevant to the intended analysis, at least not in all
countries that are supposed to be studied under a given design. The United Kingdom
is a point in case. Until recently, the parties of the Extreme Right in this country
were so weak that it was next to impossible to study their supporters by means of
survey data.

Moreover, survey studies suffer from a number of limitations of their own: Even
seemingly simple questions do not translate well into other languages, interviewers
are tempted to take shortcuts, respondents might not be able (or willing) to
accurately recall past behaviour and might be too embarrassed to admit to racist
feelings and (presumably) unpopular opinions, and so on. As a result, survey data are
often plagued by relatively high levels of systematic and random error. Macro data
on the other hand are usually collected by government agencies and are therefore
highly reliable. In summary, researchers are forced to choose between richness and
reliability, in-depth and “broad picture” perspectives, theoretical adequacy and data
availability.

But not all is bleak. (Relatively) recent initiatives in the collection,
dissemination and processing of survey data have gone a long way
to improve the situation of the subfield. The European Social
Survey⁸ with
its module on immigration (2002/2003) provides a pan-European, state-of-the-art perspective
on the hearts and minds of the voters of the Extreme Right. Similarly, the Mannheim
Trend File⁹
represents a major effort to harmonise and document the multitude of
Eurobarometer surveys that have been collected in the EC/EU member states since
the early 1970s. Finally, electoral support for the Extreme Right is now
often analysed by means of statistical multi-level models (Arzheimer, 2009;
Lubbers, Gijsberts and Scheepers, 2002), which allow for the joint analyses of
micro and macro data, thereby alleviating some of the problems outlined
above.

3 Findings

While men were always overrepresented amongst the French Front National’s voters,
it is well-documented that its electoral base has changed considerably over time
(Mayer, 1998; Mayer and Perrineau, 1992). Initially, the Front appealed primarily to
the petty bourgeoisie, but it quickly transformed itself into a non-traditional workers’
party. In between, it managed to attract occasional support from segments of the
middle classes. The Front has been dubbed the “master case” of a successful New
Right Party, and its strategies have been adopted by other parties of the European
Right (Rydgren, 2005). Therefore, it seems at least plausible that other

parties of the right have followed a similar trajectory of “proletarianization”
(Oesch, 2008). At any rate, it seems safe to assume that new, relatively
unknown parties rest on relatively fluid and less than well-defined social bases,
whereas older parties that have competed for votes in three or four consecutive
elections build a more consolidated electoral base, often with a distinct social
profile.

As it turns out, the electorates of most parties of the Extreme Right do indeed
consist of a clearly defined social core that is remarkably similar to the French
pattern. The most successful of these parties – the Freedom Party in Austria, the
Norwegian Progress Party and some others – have regularly managed to attract votes
from beyond this core so that their profile became less sharp, whereas those
that project the most radical political images (e.g. the German NPD or the
British BNP) were bound to frighten off the middle classes and have therefore
been unable to achieve this feat. This not withstanding, a very clear picture
emerges from three decades of national and comparative studies of the Extreme
Right.

3.1 Socio-Demographics

3.1.1 Gender

Most national studies have found huge differences in the propensity of men and
women to vote for the Extreme Right, even if other factors such as occupation,
education and age are controlled for. While findings vary across time, parties,
countries and details of operationalisation and model specification (Givens, 2004),
men seem to be roughly 40% more likely to vote for the Extreme Right than
female voters. Even amongst the voters of the Norwegian Progress Party and
the Danish People’s Party (which have been both lead by women for the
last four/fifteen years respectively), about two thirds are male (Heidar and
Pedersen, 2006). An important exception from this general observation,
however, is the Italian National Alliance, which appeals to both men and
women. This somewhat unusual finding seems to coincide with the party
leadership’s attempts to re-define the Alliance as a Christian-conservative
party that eventually paved the way for the AN’s merger with Forza Italia in
2009.

Comparative studies that rely on various data sources confirm this general
pattern (Arzheimer, 2009; Lubbers, Gijsberts and Scheepers, 2002). A whole host of
explanations for this phenomenon have been proposed in the literature, spanning a
multitude of approaches from psychoanalysis to rational choice. Common arguments
include that

• Some parties of the Extreme Right (like the Interwar Right) still project
  images of hyper-masculinity that are intrinsically off-putting for women
• Women are moving towards the left of men in most post-industrial societies(Inglehart and Norris, 2000)
• Women are inherently conservative and therefore more likely to be
  offended by the Extreme Right’s radicalism and more likely to identify
  with parties of the centre-right.

Related to the last point is a methodological argument: If effects of conformism
and social desirability are stronger in women, they might simply be less likely to
admit that they support the Extreme Right in an interview situation. However,
analyses of the “German Representative Electoral Statistics”, a special sub-sample of
ballot papers that bear marks which record the gender and age-bracket of the elector,
have shown that the gender gap is real, at least in Germany. Moreover, gender
effects do not completely disappear when attitudes are controlled for. As
Betz noted more than 15 years ago, the magnitude of the right-wing voting
gender gap is and remains “a complex and intriguing puzzle” (Betz, 1994,
p. 146).

3.1.2 Education

Like gender, education is a powerful predictor of the Extreme Right vote in Western
Europe. Virtually all national and comparative studies demonstrate that citizens
with university education are least likely to vote for the Extreme Right. Conversely,
the Extreme Right enjoys above average levels of support in lower educational
strata.

This relationship is neither perfect nor necessarily linear. Some parties
of the Extreme Right – most notably the Austrian Freedom Party – have
managed to attract considerable numbers of graduates in some elections.
Moreover, there is scattered evidence that the Extreme Right is even more
popular amongst those with middle levels of educational attainment than
in the lowest educational strata, although differences between these two
groups are rarely statistically significant. By and large, however, the statistical
association between educational attainment and right-wing voting is remarkably
strong.

There are basically three types of explanations for this relationship. A first
approach claims that citizens with higher levels of educational attainment for various
reasons tend to hold more liberal values than others (Weakliem, 2002) and are
therefore less likely to support the authoritarian policies of the Extreme
Right.

A second argument holds that supporters of the Extreme Right are primarily
motivated by ethnic competition (Bélanger and Pinard, 1991). Since immigration
into Western Europe is mostly low-skilled, it poses a threat only to those with low to
medium levels of attainment. In fact, low-skilled immigration might be seen as a
benefitting graduates, as it might bring down wages in some sectors of the service
industry (e.g. childcare or housekeeping), thereby increasing their ability to purchase
these services.

Third, graduates might be more susceptible to effects of social desirability,
which would lead them to under-report support for the Extreme Right. This
attainment-specific bias would result in overestimating the effect of education.

3.1.3 Class and Age

Social class is a notoriously complex concept, but voting studies usually rely on either
some variant of the classification developed by Erikson, Goldthorpe and
Portocarero (Erikson, Goldthorpe and Portocarero, 1979) or some simple typology
that pits the “working class” against one or more other broadly defined
occupational groups. Either way, class (in this sense) is closely related to formal
education.

As outlined above, many parties initially appealed primarily to the so-called
“pettty bourgeoisie” of artisans, shopkeepers, farmers and other self-employed
citizens. As this group has been subject to a constant and steady numerical decline in
all European societies, the Extreme Right has been forced to broaden its social base.
Nowadays, non-traditional workers, other members of the lower middle classes and
the unemployed form the most important segment of the Extreme Right’s electorate.
Conversely, managers, professionals, owners of larger businesses and members of the
middle and higher ranks of the public service are the groups least likely to vote for
the Extreme Right. This chimes with the effect of educational attainment, although
both variables are not perfectly correlated and operate independently of each
other.

Apart from the effect of class, many studies demonstrate an effect of age, with
younger (< 30) voters being more likely to vote for the Extreme Right. Presumably,
this age group is less firmly attached to the established parties, has a more intensive
sense of ethnic competition, is subject to lower levels of social control and more prone
to experiment with their vote.

3.1.4 Social ties and other socio-demographic factors

Various studies have looked at the respective effects of other socio-demographic
factors, often inspired by a varieties of disintegration, reference-group or cleavage
theories. For rather obvious reasons, trade union membership is often a strong
deterrent to right-wing voting. Slightly less self-explanatory is the negative effect of
church attendance, which contradicts earlier American findings. As Arzheimer and
Carter demonstrate, this effect is mostly due to pre-existing party loyalties that tie
religious voters to Christian/Conservative parties (Arzheimer and Carter,
2009a).

Other alleged factors include household size and marital status, which are both
interpreted as indicators of social isolation and anomia. The effects of these variables
are, however, weak and inconsistent.

3.2 Attitudes

Especially during their early years, parties of the Extreme Right
were often seen as vehicles for “pure”, allegedly non-political
protest.¹⁰
To be sure, the parties of the Extreme Right have very mixed
roots,¹¹
and attitudes such as distrust in and disaffection with existing parties and

Euro-Scepticism have strong effects on the probability of a right-wing vote. Yet, as
immigration emerged as their central issue during the 1980s, anti-immigrant
sentiment arose as the single most powerful predictor of the right-wing vote.

Anti-immigrant sentiment is a complex attitude, and there is no consensus as to
which sub-dimensions it entails and how it should be operationalised. Just as not all
parties and politicians of the Extreme Right are extremists, not all immigration
sceptics are xenophobes or racists (Rydgren, 2008). But what ever their exact
nature is, concerns about the presence of non-Western immigrants go a long
way towards understanding support for the Extreme Right. While not all
citizens who harbour such worries do in fact vote for the Extreme Right (many
support parties of the Centre Left or Centre Right), there are next to no
right-wing voters who have a positive view of immigrants and immigration. Even
if the “single-issue thesis” (Mudde, 1999) of right-wing support does not
paint an accurate picture of these parties and their voters, it is difficult to
overstate the importance of immigration for the modern (post-1980) Extreme
Right.

Finally, identifications with either a party of the Extreme Right or another party
compromise another important class of attitudes that help to understand and predict
the Extreme Right vote. As outlined above in section 1.2, party identifications are
often ignored in models of right-wing voting, presumably because their likely
effects are self-evident. This is, however, a grave mistake, as this omission can
seriously bias the estimates for other variables and ignores the fact that
many right-wing parties have consolidated their electoral base over the last
decades.

3.3 Contextual Factors

Since the mid-1990s, contextual (mostly system level) factors have attracted a great
deal of interested as they were increasingly seen as key variables for explaining the
huge variation in right-wing support. Some technical issues not withstanding, the
analysis by Jackman and Volpert (Jackman and Volpert, 1996) was groundbreaking
in many ways. In an aggregate study that spans 103 elections held in 16 countries
between 1970 and 1990, Jackman and Volpert analyse the impact of various economic
and institutional variables on the Extreme Right vote. Their main results are
that the Extreme Right benefits from high unemployment, PR voting and
multi-partyism, whereas high electoral thresholds are detrimental for the Extreme
Right.

Later studies have elaborated on these findings by dealing with some of the
technical and conceptual problems (Golder, 2003), using aggregated survey
data (Knigge, 1998), and considering mediating effect of the welfare state
(Swank and Betz, 2003). Around the turn of the century, the view that
immigration (usually operationalised by the number of refugees or asylum
seeker applying or actually taking residence in a country) has a substantial
positive effect on right-wing voting was firmly established, whereas the effects
of inflation and of (aggregate) unemployment appeared to be much less
consistent.

The useful study by Lubbers, Gijsberts and Scheepers (Lubbers, Gijsberts and
Scheepers, 2002) represents another important step forward, as these authors were

the first to model right-wing voting in a multi-level perspective that combines
individual-level and system-level predictors. From a methodological point of view,
multi-level modelling s is currently the most appropriate approach to the research
problem. The study by Lubbers et al. was also important because they complemented
their model with political factors, namely characteristics of the Extreme Right
parties.

This approach was taken one step further again by Arzheimer and Carter, who
include various measures for the ideological positions of other parties as well as
institutional characteristics, unemployment and immigration rates into a
comprehensive model of “opportunity structures” for the Extreme Right (Arzheimer
and Carter, 2006).

As it turns out, immigration and unemployment work in the expected direction,
though their effect is moderated by welfare state interventions that insulate
vulnerable social groups from their impact. Moreover, the established parties have a
substantial impact on the success of their right-wing competitors: If they publicly
address issues such as immigration, the Extreme Right benefits, presumably because
it gains some legitimacy and relevance in the eyes of the public. If, however,
they simply ignore the issues of the Extreme Right, these parties seem to
suffer(Arzheimer, 2009).

The studies discussed in this section provide a detailed and nuanced account of
the interplay between social, economic, institutional, political and individual factors
required to transform the Extreme Right’s electoral potential into actual votes. There
is, however, a rather large elephant in the room: the media. If, as Arzheimer argues,
party manifestos (that are usually of little relevance for the general public) have a
sizeable impact on the right-wing vote, it is reasonable to assume that media effects
of agenda setting and priming are even more important. Country-level studies
for the Netherlands and for Germany demonstrate that this is indeed the
case (Boomgaarden and Vliegenthart, 2007, 2009). There are, however, no
comparative studies on media effects (yet), because the necessary data are not
available.

4 Summary and Outlook

Conceptual and data problems not withstanding, Political Sociology has come up
with a clear image of the “typical” voter of the Extreme Right: male, young(ish), of
moderate educational achievement and concerned about immigrants and
immigration. While some parties of the Extreme Right have been remarkably
successful in making inroads into other strata, this group forms the core of the
right-wing electorates in Western Europe, making the Extreme Right a family of
non-traditional working class parties.

As the size of this group is largely stable and roughly similar across countries, the
interest in contextual factors that may trigger the conversion of potential into
manifest support has grown during the last decade. While immigration,
unemployment and other economic factors emerge time and again as variables that
play a central role, recent studies demonstrate that political factors, which are (up to
a degree) subject to political control and manipulation, act as important
moderators.

The most glaring omission so far is the lack of comparative studies on the impact

that media coverage of immigrants and immigration policies has on the prospects of
the Extreme Right. Another area where more research is needed concerns the
effects of smaller spatial contexts on the right-wing vote. After all, social,
political and economic conditions vary massively at the sub-national, e.g. across
provinces, districts, towns and even neighbourhoods. It stands to reason
that citizens rely on these local conditions, which have a massive impact
on their everyday lives, to evaluate politicians, parties and policies at the
national level. This approach has been fruitfully employed at the national level
(Kestilä and Söderlund, 2007a; Lubbers and Scheepers, 2002). Comparative
studies, however, have been hampered by vastly different subnational divisions
and a lack of comparable micro- and macro-data. New initiatives for the
geo-referencing of survey data and the pan-European harmonisation of small-area
government data will hopefully help us to overcome that impasse in the
future.

References

Adorno, Theodor W. et al. (1950). The Authoritarian Personality. New York:
Harper.

Altemeyer, Bob (1996). The Authoritarian Specter. Cambridge: Harvard University
Press.

Arzheimer, Kai (2009). “Contextual Factors and the Extreme Right Vote in
Western Europe, 1980–2002”. In: American Journal of Political Science 53.2,
pp. 259–275. url: http://dx.doi.org/10.1111/j.1540-5907.2009.00369.x.

Arzheimer, Kai and Elisabeth Carter (2006). “Political Opportunity Structures and
Right-Wing Extremist Party Success”. In: European Journal of Political Research
45, pp. 419–443. doi: 10.1111/j.1475-6765.2006.00304.x.

— (2009a). “Christian Religiosity and Voting for West European Radical Right
Parties”. In: West European Politics 32.5, pp. 985–1011. doi: 10.1080/014023809
03065058.

— (2009b). “How (not) to operationalise subnational political opportunity
structures: A critique of Kestilä and Söderlund’s study of regional elections”. In:
European Journal of Political Research 48.3, pp. 335–358. doi: 10.1111/j.1475-
6765.2009.00842.x.

Bélanger, Sarah and Maurice Pinard (1991). “Ethnic Movements and the
Competition Model. Some Missing Links”. In: American Sociological Review 56,
pp. 446–457.

Betz, Hans-Georg (1994). Radical Right-Wing Populism in Western Europe.
Houndmills, London: Macmillan.

Boomgaarden, Hajo G. and Rens Vliegenthart (2007). “Explaining the Rise of
Anti-Immigrant Parties: The Role of News Media Content”. In: Electoral Studies
26.2, pp. 404–417.

— (2009). “How News Content Influences Anti-Immigration Attitudes:
Germany, 1993-2005”. In: European Journal of Political Research 48.4,
pp. 516–542. doi: 10 . 1111 / j . 1475 – 6765 . 2009 . 01831 . x. url:
http://www3.interscience.wiley.com/journal/122273548/abstract.

Brug, Wouter van der and Meindert Fennema (2003). “Protest or Mainstream?
How the European Anti-Immigrant Parties Developed into two Separate Groups
by 1999”. In: European Journal of Political Research 42, pp. 55–76.

Carter, Elisabeth (2005). The Extreme Right in Western Europe. Manchester, New
York: Manchester University Press.

Childers, Thomas (1983). The Nazi Voter. The Social Foundations of Fascism in
Germany, 1919-1933. Chapel Hill: University of North Carolina Press.

Coleman, James S. (1994). Foundations of Social Theory. Cambridge, London: The
Belknap Press of Harvard University Press.

Dalton, Russell J., Scott C. Flanagan and Paul Allen Beck, eds. (1984). Electoral
Change in Advanced Industrial Democracies: Realignment or Dealignment.
Princeton: Princeton University Press.

Dollard, John et al. (1939). Frustration and Aggression. New Haven, London: Yale
University Press.

Erikson, Robert, John H. Goldthorpe and Lucienne Portocarero (1979). “Intergenerational
Class Mobility in Three Western European Societies. England, France and Sweden”.
In: British Journal of Sociology 30, pp. 415–441. url: http://links.jstor.org/sici?sici=0007-1315%28197912%2930%3A4%3C415%3AICMITW%3E2.0.CO%3B2-F.

Esses, Victoria M., Lynne M. Jackson and Tamara L. Armstrong (1998). “Intergroup
Competition and Attitudes Toward Immigrants and Immigration: An Instrumental

Model of Group Conflict”. In: Journal of Social Issues 54, pp. 699–724. url:
\url{http://www.blackwell-synergy.com/links/doi/10.1111/0022-4537.911998091/abs}.

Falter, Jürgen W. (1991). Hitlers Wähler. München: Beck.

Falter, Jürgen W. and Reinhard Zintl (1988). “The Economic Crisis of the 1930s
and the Nazi Vote”. In: Journal of Interdisciplinary History 19, pp. 55–85.

Falter, Jürgen W. et al. (1983). “Arbeitslosigkeit und Nationalsozialismus.
Eine empirische Analyse des Beitrags der Massenerwerbslosigkeit zu den
Wahlerfolgen der NSDAP 1932 und 1933”. In: Kölner Zeitschrift für Soziologie and
Sozialpsychologie 35.3, pp. 525–554.

Frey, Bruno S. and Hannelore Weck (1981). “Hat Arbeitslosigkeit den Aufstieg
des Nationalsozialismus bewirkt?” In: Jahrbuch für Nationalökonomie und Statistik
196, pp. 1–36.

Givens, Terri E. (2004). “The Radical Right Gender Gap”. In: Comparative
Political Studies 37, pp. 30–54.

Golder, Matt (2003). “Explaining Variation in the Success of Extreme Right
Parties in Western Europe”. In: Comparative Political Studies 36, pp. 432–466.
url: \url{u:\work\Texte\PDF-Dateien\Golder2002b.pdf}.

Heidar, Knut and Karina Pedersen (2006). “Party Feminism: Gender Gaps within
Nordic Political Parties”. In: Scandinavian Political Studies 29.3, pp. 192–218. doi:
10.1111/j.1467-9477.2006.00149.x.

Ignazi, Piero (1992). “The Silent Counter-Revolution. Hypotheses on the
Emergence of Extreme Right-Wing Parties in Europe”. In: European Journal of
Political Research 22, pp. 3–34.

Inglehart, Ronald (1977). The Silent Revolution. Changing Values and Political
Styles Among Western Publics. Princeton: Princeton University Press.

Inglehart, Ronald and Pippa Norris (2000). “The Developmental Theory of the
Gender Gap: Women’s and Men’s Voting Behavior in Global Perspective”. In:
International Political Science Review 21.4, pp. 441–463. doi: 10.1177/01925121
00214007. eprint: http://ips.sagepub.com/content/21/4/441.full.pdf+html.
url: http://ips.sagepub.com/content/21/4/441.abstract.

Jackman, Robert W. and Karin Volpert (1996). “Conditions Favouring Parties of
the Extreme Right in Western Europe”. In: British Journal of Political Science 26,
pp. 501–521. url: \url{C:\Documents%20and%20Settings\karzh\My%20Documents\Work\Texte\PDF-Dateien\jackman-volpert-1996.pdf}.

Kestilä, Elina and Peter
Söderlund (2007a). “Local Determinants of Radical Right-Wing Voting: The case
of the Norwegian Progress Party”. In: West European Politics 30, pp. 549–572.
url: \url{http://www.informaworld.com/10.1080/01402380701276394}.

— (2007b). “Subnational Political Opportunity Structures and the Success of the
Radical Right. Evidence from the March 2004 Regional Elections in France”. In:
European Journal of Political Research 46, pp. 773–796. url: \url{http://www.blackwell-synergy.com/doi/abs/10.1111/j.1475-6765.2007.00715.xdoi:10.1111/j.1475-6765.2007.00715.x}.

Kestilä-Kekkonen, Elina and Peter Söderlund (2009). “Rejoinder: Response
to Arzheimer and Carter”. In: European Journal of Political Research 48.3,
pp. 359–366.

Kinder, Donald R. and David O. Sears (1981). “Prejudice and Politics. Symbolic
Racism vs. Racial Threats to the Good Life”. In: Journal of Personality and Social
Psychology 40, pp. 414–431.

King, Gary, Martin Tanner and Alexander F. Wagner (2008). “Ordinary Economic
Voting Behavior in the Extraordinary Election of Adolf Hitler”. In: Journal of
Economic History 68.4, pp. 951–996.

Kitschelt, Herbert (1995). The Radical Right in Western Europe. A Comparative
Analysis. Ann Arbor: The University of Michigan Press.

Knigge, Pia (1998). “The Ecological Correlates of Right-Wing Extremism in
Western Europe”. In: European Journal of Political Research 34, pp. 249–279.

Koopmans, Ruud and Jasper Muis (2009). “The rise of right-wing populist Pim
Fortuyn in the Netherlands. A discursive opportunity approach”. In: European
Journal of Political Research 48.5, pp. 642–664.

Koopmans, Ruud and Susan Olzak (2004). “Discursive Opportunities and the
Evolution of Right-Wing Violence in Germany”. In: American Journal of Sociology
110.1, pp. 198–230. url: href=”\url{://000223272300006}” class=”url” >\url{<GotoISI>://000223272300006}.

Kriesi, Hanspeter et al. (1992). “New Social Movements and Political
Opportunities in Western Europe”. In: European Journal of Political Research 22,
pp. 219–244.

Küchler, Manfred (1992). “The NSDAP Vote in the Weimar Republic. An
Assessment of the State-of-the-Art in View of Modern Electoral Research”. In:
Historical Social Research 17, pp. 22–51.

Lederer, Gerda and Peter Schmidt, eds. (1995). Autoritarismus und Gesellschaft.
Trendanalysen und vergleichende Jugenduntersuchungen 1945–1993. Opladen:
Leske und Budrich.

Lipset, Seymour M. and Reinhard Bendix (1951). “Social Status and Social Structure.
A Re-Examination of Data and Interpretations: II”. In: British Journal of Sociology
2.3, pp. 230–254. url: \url{http://links.jstor.org/sici?sici=0007-1315%28195109%292%3A3%3C230%3ASSASSA%3E2.0.CO%3B2-E}.

Lubbers, Marcel, Mérove Gijsberts and Peer Scheepers (2002). “Extreme
Right-Wing Voting in Western Europe”. In: European Journal of Political Research
41, pp. 345–378.

Lubbers, Marcel and Peer Scheepers (2002). “French Front National voting: A
Micro and Macro Perspective”. In: Ethnic and Racial Studies 25, pp. 120–149.

Mayer, Nonna (1998). “The Front National in the Plural”. In: Patterns of Prejudice
32, pp. 4–24.

Mayer, Nonna and Pascal Perrineau (1992). “Why Do They Vote for Le Pen?” In:
European Journal of Political Research 22, pp. 123–141.

Meloen, Jos D., Gert van der Linden and Hans Witte (1996). “A Test of the
Approaches of Adorno et al., Lederer and Altemeyer of Authoritarianism in Belgian
Flanders”. In: Political Psychology 17, pp. 643–656.

Mudde, Cas
(1996). “The Paradox of the Anti-Party Party. Insights from the Extreme Right”.
In: Party Politics 2, pp. 265–276. url: href=”\url{://A1996UK61300006}” class=”url” >\url{<GotoISI>://A1996UK61300006}.

— (1999). “The Single-Issue Party Thesis: Extreme Right Parties and the
Immigration Issue”. In: West European Politics 22.3, pp. 182–197.

— ed. (2005). Racist Extremism in Central and Eastern Europe. London and New
York: Routledge.

— (2007). Populist Radical Right Parties in Europe. Cambridge: Cambridge
University Press.

Oesch, Daniel (2008). “Explaining Workers’ Support for Right-Wing Populist
Parties in Western Europe: Evidence from Austria, Belgium, France, Norway, and
Switzerland”. In: International Political Science Review 29.3, pp. 349–373.

O’Loughlin, John (2002). “The Electoral Geography of Weimar Germany:
Exploratory Spatial Data Analyses (ESDA) of Protestant Support for the Nazi
Party”. In: Political Analysis 10.3, pp. 217–243. doi: 10.1093/pan/10.3.217.
eprint: http://pan.oxfordjournals.org/content/10/3/217.full.pdf+html.
url: http://pan.oxfordjournals.org/content/10/3/217.abstract.

Parsons, Talcott (1942). “Some Sociological Aspects of the Fascist Movements”.
In: Social Forces 21, pp. 138–147.

Pettigrew, Thomas F. (2002). “Summing Up: Relative Deprivation as a Key Social
Psychological Concept”. In: Relative Deprivation. Specification, Development, and
Integration. Ed. by Iain Walker and Heather J. Smith. Cambridge: Cambridge
University Press, pp. 351–373.

Prowe, Diethelm (1994). “”Classic” Fascism and the New Radical Right in Western
Europe: Comparisons and Contrasts”. In: Contemporary European History 3,
pp. 289–313.

Robinson, William S. (1950). “Ecological Correlation and the Behavior of
Individuals”. In: American Sociological Review 15, pp. 351–357.

Rydgren, Jens (2005). “Is Extreme Right-Wing Populism Contagious? Explaining
the Emergence of a New Party Family”. In: European Journal of Political Research
44, pp. 413–437.

— (2008). “Immigration Sceptics, Xenophobes or Racists? Radical Right-Wing
Voting in Six West European Countries”. In: European Journal of Political
Research 47.6, pp. 737–765. doi: 10.1111/j.1475-6765.2008.00784.x.

Scheuch, Erwin K. and Hans-Dieter Klingemann (1967). “Theorie des
Rechtsradikalismus in westlichen Industriegesellschaften”. In: Hamburger Jahrbuch
für Wirtschafts- und Sozialpolitik 12, pp. 11–29.

Sherif, Muzafer and Carolyn W. Sherif (1953). Groups in Harmony and Tension.
An Integration of Studies on Intergroup Relation. New York: Harper and Brothers.

Swank, Duane and Hans-Georg Betz (2003). “Globalization, the Welfare State
and Right-Wing Populism in Western Europe”. In: Socio-Economic Review 1,
pp. 215–245.

Tajfel, Henri et al. (1971). “Social Categorization and Intergroup Behaviour”. In:
European Journal of Social Psychology 1, pp. 149–178.

Tarrow, Sidney (1996). Power in Movement. Social Movements, Collective Action,
And Politics. Cambridge: Cambridge University Press.

Wal, Jessika ter (2000). “The Discourse of the Extreme Right and its Ideological
Implications: The Case of the Alleanza nazionale on Immigration”. In: Patterns of
Prejudice 34.4, pp. 37–51.

Weakliem, David L. (2002). “The Effects of Education on Political Opinions: An
International Study”. In: International Journal of Public Opinion Research 14,
pp. 141–157.

Wimmer, Andreas (1997). “Explaining Xenophobia and Racism: A Critical Review
of Current Research Approaches”. In: Ethnic and Racial Studies 20, pp. 17–41.

Winkler, Jürgen (1996). “Bausteine einer allgemeinen
Theorie des Rechtsextremismus. Zur Stellung und Integration von Persönlichkeits-
und Umweltfaktoren”. In: Rechtsextremismus. Ergebnisse und Perspektiven der
Forschung (=PVS Sonderheft 27). Ed. by Jürgen W. Falter, Hans-Gerd Jaschke
and Jürgen Winkler. Opladen: Westdeutscher Verlag, pp. 25–48.

Fringe Parties

There is no universally accepted definition of what constitutes a “fringe party”. “Fringe party” is mostly used by journalists, politicians and political scientists as a pejorative term to demarcate the boundary between “reasonable politics” and the “lunatic fringe”, a label famously applied by Theodore Roosevelt in his Autobiography to describe “the foolish fanatics always to be found in such a [reform] movement and always discrediting it” (Roosevelt 1922, 206). Consequently, some political scientists have argued that the term should best be replaced by more neutral expressions, such as “marginal parties”, “non-established parties” or “non-mainstream parties”.

It is, however, possible to derive a set of common and interrelated characteristics of fringe parties from the way the phrase is used in political language: Fringe parties do usually attract only minor segments of the electorate, they are small in terms of party membership, their leadership does not (longer) belong to the established elite groups of their respective political system, and their party ideology does either violate the political consensus or is simply considered irrelevant by most voters.

Put differently, fringe parties are not part of their countries political mainstream, and they are not normally electorally relevant parties. This statement does, however, require two qualifications: First, most new parties (e.g. the Green parties) started out as fringe groups but became both electorally relevant and accepted by the more established parties and the majority of the citizenry over time. Second, some parties remain isolated and outside the political mainstream although they attract relatively large segments of the electorate (e.g. some Communist parties and some members of the Extreme Right party family).

Moreover, the ideological marginality of a party is not only conditional on time, but also more generally conditional on political context. Within the boundaries of a liberal-democratic regime, parties that promote a dictatorship of the proletariat or biological racism are clearly beyond the pale because their ideology contravenes the system’s most basic norms and values. Left- and right-wing extremist groups which aim to abolish or radically transform liberal democracy are therefore amongst the most prominent fringe parties in Western democracies. Within the context of a stable authoritarian system, however, a nascent grouping of democrats would well be considered a fringe party while the dominant non-democratic parties define the political mainstream.

Most fringe parties are, however, marginal not because they harbour extremist views but rather because they tend to campaign for a single issue which is not – at least not in itself – important enough to secure them sufficient levels of political support. Examples from Western Democracies include, but are not limited to:

Religious parties. Historically, religious conflicts have had an impact on the formation of European party systems during the 19^th century. In post-war Western Europe, the Christian-Democratic party family has been rather successful electorally, and Christian values have had an impact on the party ideologies of many other Western parties. Today there is, however, a number of tiny Christian parties that represent fundamentalist and/or evangelical views and try to distance themselves from both mainstream churches and Christian-Democratic parties. Moreover, an even smaller number of non-Christian (mostly Islamic), spiritual and New Age parties exists in Western countries. So far, they have had no electoral success whatsoever. In other countries where religious cleavages are more prominent (e.g. India or Israel), religious parties can be much more relevant and would not automatically be considered as part of the “fringe”.

Regional and ethnic parties. In many countries, ethnic and regional cleavages are simply not salient enough to sustain a single-issue party, rendering attempts to mobilize political support on the basis of some long-forgotten territorial unit futile. However, where they exist, ethnic parties are sometimes well-integrated into the political system like the Swedish People’s Party in Finland. They might even enjoy special privileges like the parties of the Danish and Sorbian minorities in Germany, which are exempted from the five percent electoral threshold. Therefore, it would be difficult to portray these parties and their constituencies as being “on the fringe” in any meaningful way.

In other countries, regionalist or separatist movements may have started out as fringe parties. But during the revival of regionalism after the Second World War, they became relevant political players that cannot be ignored by mainstream parties. This would include many of the regional parties in Spain, the Scottish National Party or the various regionalist movements and Italy that merged to form the Lega Nord. Similarly, many regional and ethnic parties in India are too relevant to be considered as genuine fringe parties.

Social groups, specific interests and frivolous parties. There is a host of rather colourful parties that claim to speak for large segments of society like women, the elderly, or families with children. Normally, the interests of these groups are fairly well represented by mainstream parties of the left and of the right who cannot afford to ignore these groups. Consequently, women’s/feminist parties, family leagues and “grey” parties usually fail to attract relevant numbers of voters.

For more specific and concentrated interests like hunting, farming or even car driving, the incentive structure is slightly different because demands from these groups are more easily ignored by the existing major parties. In most countries, however, agrarian and similar parties were either absorbed into mainstream parties or linger at (or beyond) the border of political irrelevance. The French Hunting, Fishing, Nature, Tradition Party is a case in point.

Other parties might campaign for a single political issue which is less obviously linked to a social group but nonetheless seen as marginal by most voters. An example would be the host of tiny and ineffectual eurosceptic groups in generally europhile countries like Germany. It is, however, worth pointing out again that both the Green parties and the anti-immigration parties of the Extreme Right parties began their ascendancy as marginal single-issue movements.

Finally, there is a bewildering host of frivolous parties that exist to make fun of “real” fringe or mainstream parties, either to get access to state funding or just for the fun of it. Examples include beer-lovers parties in several post-soviet states, Canada, Germany, Norway and Poland, parties that allude to grand (and often fictional) political and religious ideas (“Imperial British Conservative Party”, “Scottish Jacobite Party”, “Church of Militant Elvis Party”), parties that exist to challenge political correctness and the establishment (the “Anarchist Pogo Party of Germany” and the “PARTY”, which campaigns for rebuilding the Berlin wall), or the many British groups that play with the word party (“Mongolian Barbecue Great Place to Party”).

Kai Arzheimer, University of Mainz

References

Capoccia, Giovanni, “Anti-System Parties. A Conceptual Reassessment.” Journal of Theoretical Politics 14, no. 1 (January 2002): 9-36.

Hainsworth, Paul, The Extreme Right in Western Europe. New York, London: Routledge, 2008.

Jolly, Seth Kincaid, “The europhile fringe? Regionalist party support for European integration.” European Union Politics 8, no.1 (March 2007): 109-130.

Kitschelt, Herbert, “Left-Libertarian Parties: Explaining Innovation in Competitive Party Systems.” World Politics 40, No. 2 (January 1988): 194-234.

Norris, Pippa, “Preaching to the Converted? Pluralism, Participation and Party Websites.” Party Politics 9, no. 1 (January 2004): 21-45.

Pedersen, Mogens N., “Towards a New Typology of Party Lifespans and Minor Parties” Scandinavian Political Studies 5, no. 1 (January 1982): 1-16.

Reiter, Howard L., “Party Decline in the West. A Skeptic’s View.” Journal of Theoretical Politics 1, no. 3 (July 1989): 325-348.

Roosevelt, Theodore, An Autobiography. New York: Charles Scribner’s Sons, 1922.

Therefore, analyses by people who read and speak the relevant languages and even live in the country they are writing about fill me with vicarious pride. While I was going back and forth between Angela’s Own Country and the Disgraced Republic Formerly Known as Hellas, two fine specimen have cropped up on the internet: My old chum Ben Stanley has a journal-length piece on the Polish parliamentary elections at the monkey cage, and Jacob Christensen of trailer park political scientist fame gives an equally detailed account of the situation in Denmark.

While their work is largely qualitative, I’ve crunched the numbers and tried to shed some light on two related questions: Is the Extreme Right really subject to a process of proletarisation, and can the strategies outlined by Bale et al. work to win back working class voters.

The result is still somewhat work in progress – comments are highly appreciated.

Working Class Parties 2.0? Competition between Centre Left and Extreme Right Parties

Kai Arzheimer

1 Introduction

1.1 The Rise of the Extreme Right and the Transformation of Western European Policy Spaces

Over the last three decades, parties of the “radical”, “populist” or “extreme” right have become an almost ubiquitous feature of Western European party systems. During this “third wave” (Beyme, 1988) of radical right mobilisation, preexisting parties modified their ideological profiles (e. g. the Austrian Freedom Party, the Swiss People’s Party, the Scandinavian Progress Parties), and many more completely new parties emerged. While some of them were nothing more than a flash in the pan (e. g. New Democracy in Sweden, see Taggart 1996), others found more durable electoral support. As of today, almost all Western European political systems had to adjust (at least for a couple of years) to sustained Extreme Right mobilisation.

Initially, many observers interpreted these developments as a throwback to the Extreme Right’s inter-war onslaught on democracy (e.g. Prowe, 1994). But soon it became clear that the more successful amongst these parties departed in a crucial way from the political stances of the interwar extreme right movements and parties. Following the highly successful strategy of the French National Front (Rydgren, 2005), they abandoned biological racism, hyper-nationalism, and open hostility towards liberal democracy and instead made immigration (or more specifically the influx of non-West Europeans into Europe) their main issue. For that reason, some authors branded the emerging new party family simply as “anti-immigrant” (e.g. Fennema, 1997; Fennema and Pollmann, 1998; van der Brug, Fennema and Tillie, 2000; Bjørklund and Andersen, 2002; Gibson, 2002; Boomgaarden and Vliegenthart, 2007; Art, 2011), whereas others disputed the “single-issue thesis” (Mitra, 1988; Mudde, 1999) or argued for a more nuanced classification of subtypes (e. g. Kitschelt, 1995; Fennema, 1997; Mudde, 2007).

This manuscript as PDF

This is certainly not the right space to re-open the (largely fruitless) “war of words” (Mudde, 1996) that dominated the scholarly debate in the 1990s. Today, most scholars working in the field agree on a set of stylised facts that can be summarised as follows:

• While there are important differences amongst the “new” parties on the right in terms of their political traditions, policy positions, and general political style, these parties also display important similarities that set them apart from the Centre Right. Therefore, they should be grouped into a single (if very heterogeneous) party family.
• While some of these parties harbour extremists and many of them are highly critical of single aspects of liberal democracy (most prominently minority protection), very few of them pursue a transition to authoritarian rule.
• Therefore, “Radical” or “Extreme” (as opposed to extremist) Right are convenient shorthands for this party family.¹
• Immigration of non-western European people into Western Europe is not the only, but the single most important issue for all members of this party family. Mobilisation against immigrants and immigration is crucial for their electoral success.

Moreover, there is broad agreement that the rise of the Extreme Right presents politicians in Western Europe with a set of formidable challenges. First and foremost, their electoral success raised important questions of legitimacy. Did a vote for the Extreme Right indicate a more general lack of trust in the elites, or even a rejection of the democratic system? Was there reason to fear new “shadows over Europe” (Schain, Zolberg and Hossay, 2002), i. e. a return to the confrontational and often violent politics of the 1920s and 1930s? Should the existing parties engage in a dialog with their challengers or just ignore them?

Second, like the emergence of Green and Left-Libertarian parties, the rise of the New Right signalled a fundamental change in the patterns of party competition and co-operation in most Western European countries. For much of the postwar period, party competition in Western Europe was chiefly organised along a single left-right axis that largely reflected conflicts about economic redistribution (Fuchs and Klingemann, 1989; van der Brug, 1999). However, both issues of the “New Politics” and matters of citizenship and immigration were not primarily perceived as economic problems and were therefore not easily aligned with the old left-right-conflict. Consequently, two or three dimensions are required to reconstruct the policy spaces of most Western European democracies (Kitschelt, 1994, 1995; Warwick, 2002; Cole, 2005; Bornschier, 2010), making party competition more complex and equilibria less likely.²

Third, and perhaps closest to the hearts of politicians, the zero-sum nature of electoral competition implies that the emergence of a new party family will bring about losses for existing parties in terms of votes, seats and eventually even ministerial portfolios. But which parties would suffer most?

1.2 Competition between Centre Left and Extreme Right Parties

From the party family’s moniker, one might be tempted to assume that the Centre Right had most to lose from the emergence of the Extreme Right, at least if voters primarily care about issues: In a classical Downsian (1957) perspective, demand for right-wing policies is fixed at least in the short- and medium term, and – depending on party positions and voters’ ideal points – the entry of a new competitor would significantly reduce the vote share of the Centre Right parties. If voters behave in line with a directional model (Merrill and Grofman, 1999), the outlook for the Centre Right is even starker, as voters who disagree with their radical policies may still vote for the Extreme Right for tactical reasons.

Aggregate trends of electoral support of electorate support in 16 Western European countries from the six decades since the end of World War II seem to corroborate these arguments: While support for the right as a whole³ has been largely stable, Christian democratic parties have on average lost about five percentage points of their electorate support while the Far Right could increase their share of the vote by almost seven points (Gallagher, Laver and Mair, 2011, 301).

Accordingly, much of the political and academic debate has focused on the negative implications that the rise of these parties has had for Conservative, Christian Democratic, Liberal, and Agrarian/Centre parties (e.g. Mair, 2001, 71).⁴ But Green/New left parties are perhaps the only ones not affected by the Extreme Right’s ascendancy, as these party families appeal to very different demographics and occupy diametrically opposed positions in Western European policy spaces.⁵

Taking a more analytical approach, Kitschelt (1994, 1995) argued almost 20 years ago that a shift of the “main axis of partisan competition” was underway that would pit the New Left against the Extreme Right and present the Social Democratic/Centre Left parties with a conundrum: They would lose many of their more liberal voters to the parties of the New Left because they did not adequately represent the issues of the “New Politics” (Flanagan and Lee, 2003). At the same time, the Extreme Right would seize a sizable fraction of the working class vote, because the Centre Left had allegedly lost touch with their traditional voter base Bale (2003, 70-74).

But why would working class voters turn to the Extreme Right? Historically, support for the post-war Extreme Right had chiefly come from the “petty bourgeoisie” of artisans, small shop-keepers and farmers that made up the lower strata of the middle classes. This constituency was authoritarian and staunchly anti-communist/anti-socialist.

Working class voters, on the other hand, were often embedded in a network of trade unions and similar intermediate organisations, held strong preferences for redistribution, and were firmly attached to traditional left parties. Even if many voters (and some of the rank-and-file members) of these parties expressed a healthy degree of working-class authoritarianism (Lipset, 1959), elites and opinion leaders within the traditional working classes were firmly committed to principles of equality and international solidarity. Therefore, the idea of a large-scale swing from the Centre Left to the Extreme Right would have looked rather far-fetched three or four decades ago.

Through twin processes of de-alignment (Dalton, Flanagan and Beck, 1984) and social change (Crouch, 1999), however, swathes of the (non-traditional) working class have become available for other parties than the traditional left. Moreover, the Extreme Right has modified its programmatic appeal considerably over the six decades since the end of World War II, thereby becoming more palatable for members of the working class.

Perhaps the most radical interpretation of these programmatic changes was developed by Herbert Kitschelt in a highly influential monograph (Kitschelt, 1995). Kitschelt argued that under conditions of economic globalisation, workers outside the public sector would develop a taste for free market policies. At the same time, they would remain authoritarian with respect to their socio-cultural attitudes. According to Kitschelt, catering for these twin demands was the electoral “winning formula” that fuelled the unprecedented successes of the French National Front and the Austrian Freedom Party during the 1980s and early 1990s. A similar argument was developed by Betz in his seminal monograph (Betz, 1994). Figure 1, which slightly simplifies the presentation in Kitschelt (1995), shows the respective policy positions of Social Democratic, old style “Welfare Chauvinist” and more modern “Radical Right” parties.

In hindsight, however, the Extreme Right’s flirt with “neoliberalism” – presumably not a very serious affair in the first place – proved short-lived and inconsequential (de Lange, 2007). Within a few years after the publication of Kitschelt’s book, many Extreme Right parties had gone all the way from vocal champions of neoliberalism to globalisation critics, and the allegedly outdated “welfare chauvinist” strategy that campaigns for a strong but ethnically exclusionary welfare state had gained a lot of currency in Far Right circles. Consequentially, Betz (2003) has altoghether abandoned the idea that the Extreme Right does seriously pursue a “neo-liberal” agenda or has done so in the past, while Kitschelt has modified his original ideas considerably (McGann and Kitschelt, 2005).

Moreover, more recent research (Arzheimer, 2009b) demonstrates that there is no working class demand for “neo-liberal” policies. Where both members of the working class and the petty bourgeoisie support the Extreme Right, they tend to disagree on economic policies and cast their vote because the salience of economic issues is low (Ivarsflaten, 2005).

But even if the mid-1990s accounts by Betz and Kitschelt were wrong in their diagnoses, they clearly identified a very important symptom: Since the early 1980s, the Extreme Right has undergone a process of “proletarization and (uneven) radicalisation” (Ignazi, 2003, 216). At least for the relatively successful parties (e. g. the Austrian Freedom Party, the Norwegian Progress Party and the French National Front), there is some evidence for a trend from electorates that were heterogeneous or centred around a core of voters from the petty bourgeoisie towards more working class-dominated constituencies (Beirich and Woods 2000; Betz 2002; Bjørklund and Andersen 2002; Mayer 1998, 2002; Riedlsperger 1998; Rydgren 2003; see Oesch 2008 for a comparative cross-sectional analysis of Austria, Belgium, France, Norway, and Switzerland).

This new pattern of class-voting in Western Europe is not based on long-standing party loyalties but rather on group- and policy-related attitudes: Public opinion data consistently shows that the Extreme Right vote is driven by intense worries about immigrants and immigration⁶ that are most prevalent amongst voters with low levels of educational attainment who are either unemployed or holding blue-collar jobs.⁷

While many authors frame these worries as “resentment” and interpret the underlying policy dimension primarily in terms of “culture” and “identity”, one should not ignore the fact that concerns about immigrants and immigration have clear economic underpinnings: The vast majority of immigrants in Western Europe are unskilled or semi-skilled workers. Obviously, members of the working class are much more likely to perceive these persons as an economic threat than middle class voters, who might actually benefit from the additional supply of cheap labour.

On the whole, research since the mid-1990s suggests that patterns of party competition and class voting have indeed changed, although in a way that is quite different from Kitschelt’s original reading of the situation (see figure 2). Instead of converging on the “Radical Right” strategy, parties of the Extreme Right are looking for a (not very) “new winning formula” (de Lange, 2007) and have incorporated elements of “welfare chauvinism” into their manifestos, although to a varying degree. Social Democratic parties, on the other hand, have cautiously moved to more economically centrist (and arguably more socially liberal) positions in a bid to respond to the new challenges of the 21st century and to become more attractive for middle-class voters (see Keman 2011 for a comprehensive analysis that outlines the extent of this shift in 19 polities). This programmatic change opened up additional space for the Extreme Right and made it even easier for them to poach working class voters from the Centre Left. That raises the question whether there is anything the Centre Left can do about this development.

The remainder of this chapter is organised as follows. Section 2 gives a brief overview of the data base and the statistical models and methods used for its analysis. Section 3 presents a comparative longitudinal analysis of the “proletarisation” of the Western European Extreme Right Vote since 1980. Section 4 directly looks at the competition between Extreme Right and Centre Left parties for the working class vote. Finally, section 5 briefly summarises the findings.

2 Data, Model, Methods

The analyses presented in the following sections cover the member states of the European Union (EU) as it existed before the Eastern enlargement rounds, plus Norway. Survey data come from the Mannheim Eurobarometer Trend File (Schmitt et al., 2009a,b), a partial cumulation of the bi-annual series of Eurobarometer surveys that greatly facilitates cross-national and longitudinal analyses. The temporal coverage of these data spans the whole period of the Extreme Right’s electoral ascendancy during the 1980s and 1990s, as well as a few years of the new millenium.

There are, however, a few gaps: Data for Austria, Finland, Sweden and Norway are not available for the whole period. Moreover, surveys without any supporters of the Extreme Right had to be excluded, which removed the United Kingdom and the Republic of Ireland from the analysis.⁸ Figure 3 gives a graphical overview of the spatial and temporal coverage.

Information on social class in the Eurobarometer series is effectively restricted to present occupation. To simplify the presentation, respondents were coded as holding blue-collar jobs (“workers”), belonging to the petty bourgeoisie (“farmers and owners”), holding any other occupation (“other”), being unemployed, or being retired.⁹

In order to model contextual effects on right-wing voting, the Eurobarometer surveys were augmented with macro data. Information on unemployment rates and unemployment benefits comes from the OECD (2002; 2003; 2004), while data on new asylum applications – in the Western European context, a very useful proxy for actual immigration figures – were taken from reports compiled by the OECD and the Office of the United Nations High Commissioner for Refugees (OECD, 1992; UNHCR, 2002).

Finally, the Comparative Manifesto Project database was used to construct a series of five variables that capture the positions of mainstream parties with respect to the issues of the Extreme Right, i.e. “internationalism”, “multi-culturalism”, “national lifestyle”, and “law and order” (see Arzheimer and Carter (2006); Arzheimer (2009a) for a more detailed discussion of the rationale behind these measures). These variables pertain to the position of the respective Social Democratic party, the most extreme position taken by any other mainstream party, the salience of these issues for the Social Democrats, the salience for all other mainstream parties, and the variation in policy positions across all other mainstream parties.¹⁰

To account for the hierarchical nature of the data (respondents are nested within 336 survey waves that were conducted in 15 polities), binary logistic multi-level models are specified. Because the Extreme Right is persistently stronger in some countries (e. g. Belgium and France) than in others (say Spain and Germany), stable unit (country) effects are represented by a series of dummies.¹¹ These dummies are also required to control for changes in the national composition of the sample over time. Specifying country effects leaves just two levels of analysis: voters, and the particular contexts in which they were interviewed.

Even when controlling for unit effects and contextual information, it makes sense to assume that people who are interviewed in the same survey wave are subject to common random political shocks that affect their voting behaviour. These shocks are modelled as draws from a Gaussian distribution with standard deviation σ_u, which estimated from the data in addition to the usual parameters. As a result of these shocks, respondents in the same context will give more similar answers than one expect by chance alone. The intraclass correlation coefficient ρ which ranges from 0 to 1 is a measure for this similarity, with values closer to unity indicating greater alikeness within a context.¹²

All models were estimated using the xtlogit procedure in Stata 11.2. Checks indicate that the number of quadrature points used was sufficient to guarantee stable estimates.

3 The Proletarisation of the Western European Extreme Right Vote, 1980-2002

The idea of a “proletarisation” (Ignazi, 2003) of the Western European Extreme Right features prominently in the literature, but very little comparative cross-temporal empirical evidence for this alleged development has been presented so far. With the Eurobarometer Trend File, however, it is possible to trace the purported trajectory of the Extreme Right’s electorate.

	(1)	(2)
Worker	0.483	0.441
	(0.0277)	(0.0307)
Farmer/Owner	0.438	0.478
	(0.0347)	(0.0363)
Retired	0.0546	0.0563
	(0.0282)	(0.0318)
Unemployed	0.555	0.552
	(0.0410)	(0.0455)
Time		0.00593
		(0.000666)
Worker × Time		0.00176
		(0.000433)
Farmer/Owner × Time		-0.00207
		(0.000512)
Retired × Time		-0.0000549
		(0.000442)
Unemployed × Time		0.000120
		(0.000665)
Observations	254726	254726
σu	0.720	0.621
ρ	0.136	0.105
Groups	336	336
t statistics in parentheses
, ,

The left column (1) of table 1 shows the estimates from a simple socio-demographic multi-level model of Extreme Right voting in Western Europe. The model is based on just under 255000 interviews.

As can be seen from the coefficients, being unemployed or belonging to the working class or the petty bourgeoisie considerably increases the chances of an extreme right vote, compared to the “other” category. Either factor increases the logit of an Extreme Right vote by 0.4 to 0.5 points. Being retired, on the other hand, does not make an appreciable difference.

The exact impact of this increase depends on the fixed country effects but is roughly proportional to a 50 per cent change in the probability of the Extreme Right vote. In Austria, for instance, members of the “other” group have an estimated probability of just under 15 per cent of voting for the Freedom Party. For workers, the estimated probability is almost 22 per cent.

The term proletarisation, however, implies change over time. In the right column (2) of table 1, the membership indicator were interacted with an additional variable that represents the time (in months) at which the survey was taken. In order to minimise collinearity, the variable was centred so that it takes a value of zero for March 1991, which is the midpoint of the period under observation. Given the huge range of the time variable (see table 2), it is not surprising that the estimated coefficients are very small. Nonetheless, the picture that emerges is remarkably clear. The effect of being a pensioner is essentially stable, while the effect of being unemployed increases only very slightly over time. The effect of being a member of the working class, on the other hand, becomes considerably stronger with time, while the effect of belonging to the petty bourgeoisie becomes weaker at roughly the same rate.

Taken together, these results show that the Extreme Right electorates indeed underwent a process of proletarisation between 1980 and the early naughties. Moreover, these findings cannot be ascribed to changes in the composition of the sample (i.e. the accession of Greece, Spain and Portugal to the European Union during the 1980s and the 1995 enlargement), because fixed country effects are controlled for. Therefore, the interaction effects represent common trends across all 15 polities. This constitutes the first truly comparative and longitudinal evidence for a general proletarisation of the Extreme Right vote in Western Europe.

But how important are these trends in substantive terms (i.e. votes and seats)? Again, the exact size is context-dependent and most easily illustrated by calculating estimates for an arbitrary country. The estimated vote share of the Danish Extreme Right amongst workers in 1980, for instance, was just under two per cent, while the respective figure for members of the Danish petty bourgeoisie was about three per cent. In 2002, the estimate for the petty bourgeoisie was eight per cent, while the figure for the working class has risen to almost 13 per cent. Although the Extreme Right has made considerable inroads into both groups, the ratio of the respective propensities to vote for the Extreme Right has been reversed. Therefore, it makes indeed sense to talk about a proletarisation of the Extreme Right vote. This trend is further amplified by the fact that the petty bourgeoisie is shrinking even faster than the working class.

One should, however, not throw out the baby with the bath water: Precisely because the working class is in decline, there is a natural limit to this process. Moreover, while social class has obviously lost some of its previous importance (Clark, Lipset and Rempel, 1993; Nieuwbeerta and Graaf, 2001), its effect on the probability of voting for the traditional left has by no means disappeared completely (Evans, 2001). Thus, the next section section will look specifically at the competition between Extreme Right and Social Democratic parties over the working class vote.

---

	min	p25	mean	p75	max
XR vote	0.00	0.00	0.04	0.00	1.00
Worker	0.00	0.00	0.18	0.00	1.00
Farmer/Owner	0.00	0.00	0.10	0.00	1.00
Retired	0.00	0.00	0.22	0.00	1.00
Unemployed	0.00	0.00	0.06	0.00	1.00
Time	-131.00	-36.00	10.22	56.00	130.00
AT	0.00	0.00	0.03	0.00	1.00
BE	0.00	0.00	0.07	0.00	1.00
DE-E	0.00	0.00	0.06	0.00	1.00
DE-W	0.00	0.00	0.14	0.00	1.00
DK	0.00	0.00	0.14	0.00	1.00
ES	0.00	0.00	0.03	0.00	1.00
FI	0.00	0.00	0.03	0.00	1.00
FR	0.00	0.00	0.11	0.00	1.00
GR	0.00	0.00	0.06	0.00	1.00
IT	0.00	0.00	0.12	0.00	1.00
LU	0.00	0.00	0.01	0.00	1.00
NL	0.00	0.00	0.13	0.00	1.00
NO	0.00	0.00	0.03	0.00	1.00
PT	0.00	0.00	0.02	0.00	1.00
SE	0.00	0.00	0.02	0.00	1.00
	254726

Table 2: Sociodemographic model: summary statistics

---

4 Left or (Extreme) Right? The Western European Working Class Vote, 1980-2002

In their recent analysis of Social Democratic reactions to the rise of the Extreme Right, Bale et al. (2010) have usefully identified three elements of this challenge, and three strategies available to the Centre Left: The presence of Extreme Right parties will heighten the salience of “right” issues in general, can increase the number of potential coalition partners for the Centre Right, and may lure working class voters away from the left. Social democratic parties can respond by holding on to their traditional relatively tolerant position towards immigrants, by trying to “defuse” the immigration issue, or by shifting their position (Bale et al., 2010, 412).

As Bale et al. (2010, 413-414) point out, the effectiveness of the “defuse” strategy is very limited, making the first strategy the default, as Social Democratic party elites are normally committed to values of tolerance and international solidarity. Therefore, they will find it difficult to abandon their support for relatively liberal immigration policies to avoid political losses. Such normative convictions seriously restrain the Centre Left’s room for manoeuvre.

Nonetheless, the qualitative analysis of developments in Austria, Denmark, the Netherlands and Norway by Bale et al. shows that Social Democratic parties have sometimes modified their positions on the immigration dimension (see Bale et al., 2010, 421 for an overview). A quantitative analysis (see figure 4) of the CMP-Data provides further evidence for such programmatic shifts: Although there is considerable national variation, Social Democratic parties in many countries including Germany, Denmark, Finland, France, Italy, and the Netherlands have taken consistently tougher stands on issues of migration and national identity over the years.

---

Fixed country effects omitted

	(1)	(2)	(3)
Male	0.445	0.449	0.448
	(0.0515)	(0.0517)	(0.0517)
Time	0.00982	0.00692	0.00651
	(0.000874)	(0.00121)	(0.00127)
Toughness (max SD)		0.0327
		(0.0270)
Toughness (mean SD)			0.0296
			(0.0309)
Ideology Salience (SD)		-0.0437	-0.0383
		(0.0257)	(0.0247)
Toughness (other)		-0.00246	0.00360
		(0.0255)	(0.0242)
Ideological Variance (other)		-0.0131	-0.0137
		(0.00437)	(0.00429)
Ideology Salience (other)		0.119	0.116
		(0.0291)	(0.0288)
New Asylum Applications		0.0386	0.0326
		(0.0667)	(0.0663)
Unemployment		0.0999	0.106
		(0.0374)	(0.0388)
Replacement Rate		0.0515	0.0520
		(0.0138)	(0.0138)
Observations	19858	19663	19663
σu	0.733	0.645	0.646
ρ	0.140	0.112	0.113
Groups	336	327	327
t statistics in parentheses
, ,

But how do working class voters respond to this repositioning of the Centre Left? The left column (1) in table 4 gives the estimates for the coefficients of a very simple baseline model. The sample is restricted to working-class respondents who intend to vote either for a Social Democratic party (0) or and Extreme Right party (1). The model features a single sociodemographic control to account for the well-known gender gap, and a linear (in the logits) trend factor. Like the models in the previous section, the model also contains fixed country effects to account for stable differences between polities. Estimates for these effects (not tabulated) are very low in countries as diverse as Germany (-3.3), Spain (-6.3), Finland (-4), Luxembourg (-4.6), Portugal (-5.7), or Sweden (-5.2), which implies that in these countries, the odds of a Social Democratic vote are between 27 (exp(3.3)) and 545 (exp(6.3)) times higher than the odds of an Extreme Right vote.

There is, however, a set of countries including Austria (-1.7), Belgium (-2), Denmark (-2.2), France (-2.4), and particularly Italy (-.65), where the odds of an Extreme Right vote are much higher in comparison. While the result for Italy might be due to the fact that the AN as the largest relevant party in the country has become relatively moderate since the 1990s, the findings for the other countries are striking: Across the board, a Social Democratic vote is only between 5.5 and 11 times more likely than an Extreme Right vote in this core constituency of the Centre Left.

Moreover, the trend factor indicates that the odds of an Extreme Right vote have risen considerably over time: If one is prepared to take the model estimates at face value, the odds of a working class respondent voting for the Extreme Right increases by a factor of almost 13 (exp(0.0098 × 261)) between the first and the last survey wave. Even if one takes potential deficiencies of the data and model specification into account, this clearly demonstrates that Social Democratic parties are losing support amongst working class voters.

While this is certainly an interesting finding in itself, time is chiefly used as a control in a second series of models (columns (2) and (3)) that build on Arzheimer’s (2009a) contextual model of Extreme Right voting.¹³ This amended model allows for a direct test of the viability of two of the strategies outlined by Bale et al. as well as for an indirect test of the third.

Since some elections were contested by two or more parties that were classified as Social Democratic by the CMP, Social Democratic ideology was operationalised in two variants: “Toughness” refers either to the most right-leaning party (column (2)) or to the average of all Social Democratic party positions, weighted by the respective party’s share of the vote (column (3)).¹⁴ However, the way Social Democratic ideology is measured makes virtually no difference.

According to this second set of estimates, the trend towards more Extreme Right voting is slightly less pronounced¹⁵ once the additional contextual variables are taken into consideration. Nonetheless, given its wide range time still has the strongest effect amongst all covariates.

The level of welfare state protection as measured by the OECD’s standardised wage replacement rate for the unemployed also has a strong positive effect on the probability of an Extreme Right vote. Raising the standards from the first to the third quartile of its empirical distribution (see table 4) will almost quadruple the odds of a right-wing vote. Given the Extreme Right’s rediscovery of centre-left leaning policies, this could be interpreted as a result of “welfare chauvinism” and (perceived) ethnic competition (Bélanger and Pinard, 1991) over a resource that is still plentiful. However, an alternative explanation is at least as plausible: Only if the welfare state is seen as safe and can be taken for granted, workers will turn from Social Democratic parties towards the Extreme Right.

Another factor that has a strong effect on the electoral prospects of the Extreme Right is the salience of their issues for other parties (excluding the Social Democrats). The more statements other parties make on questions of immigration, national identity and the like, the better the Extreme Right does in the polls, irrespective of the direction of these statements. Since objective factors such as unemployment and new asylum applications (which have weak or insignificant effects) are statistically controlled for, this finding can be interpreted as evidence for an agenda setting effect (Arzheimer, 2009a).

Ideological variation in the manifestos of other parties has a moderate negative effect on right-wing voting, whereas ideological “toughness” (i.e. attempts by mainstream parties to steal the immigration issue) does not shift the balance between the Extreme Right and the Social Democrats.

Taken together, the effects of salience and ideological variation indicate that a strategy of issue diffusion could be viable in principle, if (and only if, as the Social Democrats can hardly shape political discourse singlehandedly) the other mainstream parties co-operate.

---

	min	p25	mean	p75	max
XR vote	0.00	0.00	0.12	0.00	1.00
Male	0.00	0.00	0.60	1.00	1.00
Time	-131.00	-47.00	1.99	55.00	130.00
Toughness (max SD)	-11.71	-2.01	-0.12	1.51	13.68
Toughness (mean SD)	-11.71	-2.37	-1.02	1.12	7.45
Ideology Salience (SD)	0.00	3.45	6.83	9.19	16.08
Toughness (other)	-4.54	0.59	4.84	7.92	27.54
Ideological Variance (other)	0.00	1.87	17.18	16.50	244.60
Ideology Salience (other)	0.50	5.08	8.95	12.41	31.25
New Asylum Applications	-0.98	-0.61	0.16	0.58	4.46
Unemployment	-4.91	-1.31	0.35	1.69	12.29
Replacement Rate	-31.62	-4.19	4.07	18.48	32.96
AT	0.00	0.00	0.05	0.00	1.00
BE	0.00	0.00	0.06	0.00	1.00
DE-E	0.00	0.00	0.06	0.00	1.00
DE-W	0.00	0.00	0.19	0.00	1.00
DK	0.00	0.00	0.17	0.00	1.00
ES	0.00	0.00	0.03	0.00	1.00
FI	0.00	0.00	0.02	0.00	1.00
FR	0.00	0.00	0.12	0.00	1.00
GR	0.00	0.00	0.04	0.00	1.00
IT	0.00	0.00	0.05	0.00	1.00
LU	0.00	0.00	0.00	0.00	0.00
NL	0.00	0.00	0.10	0.00	1.00
NO	0.00	0.00	0.05	0.00	1.00
PT	0.00	0.00	0.04	0.00	1.00
SE	0.00	0.00	0.01	0.00	1.00
	19663

Table 4: Full model: summary statistics

---

While this test of the “defuse” strategy might be somewhat indirect, the efficiency of the “hold” and “adopt” strategies can be more readily assessed by looking at the estimates for the “toughness” and salience variables that refer to Social Democratic parties. Neither of them has a significant effect on the odds of voting for the Extreme Right. Put differently, in this core constituency of the Centre Left, it does not make a difference whether the Social Democrats stick to their traditional positions on immigration or whether they try to toughen up their policies. Either way, their fortunes vis-a-vis the Extreme Right are largely determined by external factors and an overall negative trend.

The null effect of salience provides an interesting correlate. This variable takes a value of zero if Social Democrats completely ignore the issues of the Extreme Right, which is equivalent to a very radical “defuse” strategy, whereas positive values represent attempts to engage with the issue by making affirmative and/or critical statements. The insignificance of the coefficient provides further evidence for the assertion that a “defuse” strategy is only viable if pursued in concert.

5 Conclusion

After World War II, parties and movements of the Extreme Right were most closely associated with the petty bourgeoisie. Over the last three decades, however, the propensity of workers to vote for the Extreme Right has risen significantly. This “proletarisation” is the result of the interplay between a long-term dealignment process and increasing worries amongst the European working classes about the immigration of cheap labour. As a result, Western European Centre Left parties may find themselves squeezed between the New Right on the one hand and the New Left on the other.

The analyses in the previous section have shown that there is no obvious strategy for dealing with this dilemma. Staying put will not win working class defectors back. Toughening up immigration policies is unpalatable for many party members, does not seem to make Social Democrats more attractive for working class voters, and might eventually alienate other social groups.

That leaves what Bale et al. have called the “defuse” option, i.e. efforts to downgrade the immigration issue. In democracies, however, a single party can not normally sustain control over the political agenda. Any attempt to de-politicise immigration would therefore require some sort of agreement amongst mainstream parties. Given that Centre Right (Bale, 2003) and (for completely opposite reasons) even New Left parties might have a strategic interest to keep the debate on immigration alive, this is not a very likely outcome. In all probability, the working class parties “of a new type” will keep poaching voters from the Social Democrats.

References

¹I will treat these two terms as interchangeable through the remainder of this chapter.

²For a slightly different account of these developments see van der Brug and van Spanje (2009), who claim that European parties’ actual policy proposal can still be arranged on a single vector even though parties and voters operate in a two-dimensional space.

³Gallagher, Laver and Mair subsume five party families under this label: Christian democrats, Conservatives, Liberals, Agrarian and Centre parties, and the Far Right.

⁴Other authors, however, have highlighted the strategic opportunities that the rise of the new party family may present for the right as a whole if and when the Extreme Right can be brought into a coalition (Bale, 2003).

⁵Consequently, the rise of the Extreme Right has sometimes been framed as a “silent counter-revolution” (Ignazi, 1992) against the growing influence of the New Left and their post-materialist electoral base.

⁶These feelings are related to, but not identical with xenophobia and racism (Rydgren, 2008).

⁷See e. g. van der Brug, Fennema and Tillie (2000) and Arzheimer (2009b) for reviews of the importance of ideology and Arzheimer and Carter (2009) for the nexus between class and attitudes.

⁸The OECD does not provide Standardised Unemployment Rates for Luxembourg. Thus, the country had to be excluded from the series of models presented in section 4.

⁹Homemakers were coded according to the occupation of the householder, if available.

¹⁰For the construction of the two latter variables, positions were weighted with the parties’ shares of the vote. In some cases, elections were contested by two or more parties codes as Social Democratic by the CMP. See section 4 for details.

¹¹East and West Germany are treated as two separate polities.

¹²ρ equals the proportion of total variance contributed by σ_u.

¹³To ease the estimation and interpretation, a number of interaction effects and relatively stable macro variables were dropped. Moreover, all attitudinal and most socio-demographic variables were dropped, since they do not vary much in this subset of working class voters. The findings for many variables are somewhat different from those reported in Arzheimer (2009a) because they apply to a more limited choice set and a subsample of the original data.

¹⁴The salience variable was always constructed as an weighted average over all Social Democratic party positions in the respective election (if applicable).

¹⁵The estimated factor change in the odds is exp(0.007 × 261) = 6.

What earned me the invitation was that I have written a tiny contribution to the book so many years ago that I had forgotten about it. So here, for your edification, are my two cents on radical attitudes.

Radical Attitudes
Kai Arzheimer

Like many other concepts in political science, the notion of radicalism harks back to the political conflicts of the late 18^th and 19^th century. Even then, its content was depended on the political context and far from well defined. Consequentially, being “radical” has meant different things to different people in different times and countries. Moreover, radicalism is closely related, if not identical to a number of (equally vague) concepts such as extremism, fundamentalism, and populism. As of today, there is no universally accepted definition of radicalism, and, by implication, radical attitudes

Authors version. The final text appeared in Badie/Berg-Schlosser/Morlino, International Encyclopedia of Political Science. IPSA/Sage 2011.
Radical Attitudes PDF

There is, however, a core meaning of radicalism: radicals are willing to challenge the ground rules of politics to get to the root (Latin: radix) of what they perceive as the most pressing political problems. In any given context, radicals will confront the political establishment and will support policies whose implementation would trigger systemic change.

Radicalism in the 18^th and 19^th century

In the last third of the 18^th century, a heterogeneous group of philosophers, writers and politicians began to campaign for a thorough reform of Britain’s political system. Amongst the goals of this movement were the abolition of the slave trade, a reform of the electoral laws and a better protection of citizens’ rights. They soon gained support from the emerging middle and working classes. The parliamentarian Charles James Fox is often credited with coining the name for this new movement when he demanded a “radical reform” of the electoral system in 1797, and by 1819, the “radicals” had established themselves as a separate political force that inspired the Chartist movement and played an important role in both the creation of the Liberal and the Labour Party.

Similarly, after the restoration of the monarchy in 19^th century France, supporters of republican principles called themselves “radicals”. Over the last third of the century, they drifted to the left and were instrumental in the foundation of the country’s first modern left-wing party, the “Republican, Radical and Radical-Socialist Party” in 1901.

In Germany, “radical” was initially a political label chosen by those liberals who, in the spirit of the French Revolution, demanded civil liberties, universal male suffrage and parliamentary representation. In the second half of the 19^th centuries, this label was applied those members of the workers’ movement who favoured a revolutionary change of government (i.e. an end of the authoritarian monarchist regime). In a similar fashion, in many other European and Southern American countries “radicalism” became shorthand for a subtype of liberalism that could be located either to the left or to the right of the political centre. To the present day, “radical” parties exist in many countries including Argentina, Chile, Denmark, Ecuador, France, Italy, Paraguay, and Switzerland. Most of them are today classified as either liberal or socialist/social-democratic.

Radicalism in the 20^th century

The gradual spread of liberal democracy and its crisis during the interwar period changed the meaning of the concept. In the wake of the events in Germany, Italy, Russia, and many other European countries, radicalism became a collective term for the forces at the poles of the political spectrum that had formerly be known chiefly as “ultras” and threatened to overthrow liberal democracy: Communists on the on side, Fascists and National Socialists on the other. Consequentially, radicalism was transformed into a primarily spatial term (location on the left-right axis) with a connotation that was directly opposed to its original meaning. While the original radicals had been champions of freedom and democracy, the radicals of the 20^th century were, by virtue of their ideological preferences, opposed to these values. Under the post-war consensus of the 1950s, this perspective on radicalism became dominant.

However, less than two decades after the end of the Second World War, Seymour Martin Lipset challenged the prevailing view of the connection between centrism and support for democracy. In his seminal study Political Man (1960), Lipset claimed that Fascism and National Socialism were neither left- nor right-wing ideologies. Rather, they constituted an “extremism of the centre”. While this statement is problematic if interpreted in purely sociological terms – Fascism and National Socialism appealed both to the middle and to the working classes – it reflects the ambiguous location of these regimes on the traditional Left-Right Spectrum. On the one hand, they violently suppressed the left-wing unions and parties. On the other hand, they were hardly champions of a free market economy: Fascism and National Socialism insulated farmers and small businesses from competition, engaged in large-scale economic planning and raised government spending on welfare to unprecedented levels.

More generally, Lipset argued that attitudes towards the economy and attitudes towards democracy could vary independently. In his view, any position on the Left-Right spectrum – radical or centrist – can be combined with “the repression of difference and dissent, the closing down of the market place of ideas”. This “tendency to treat cleavage and ambivalence as illegitimate” is what Lipset called extremism.

Lipset fruitfully applied this concept to right-wing extremism in the United States. In his view, the insistence on free-market principles makes this particular breed of extremisms “right-wing”, whereas anti-semitism, homophobia, racism, religious intolerance and xenophobia are simply manifestations of the same underlying generic phenomenon. Indeed, in separate work he convincingly demonstrated that these traits are also prevalent amongst members of the working class, whose criticism of free market principles marks them as left-wingers.

Lipset’s notion of extremism is so broad that it resonates with even more general concepts that were developed around the same time by psychologists such as Hans Jürgen Eysenck (“tough-mindedness”) and Milton Rokeach (“closed mindedness”, “dogmatism”) and refer to a tendency to unconditionally accept norms, prejudice and authorities. Like Lipset, Eysenck, Rokeach and many other scholars treat political preferences in general and political radicalism in particular as an essentially two-dimensional phenomenon. However, while Lipset argued that left-right ideology and support for democratic values and institutions can vary independently, other authors disagree.

In work that is partly inspired by Lipset, Uwe Backes and Eckhard Jesse claim that there is a U-shaped link between ideological radicalism and anti-democratic extremism. While they acknowledge that radicalism and extremism are conceptually different, they argue that radical ideological positions have implications that render them incompatible with liberal democracy as defined by the core values of the French revolution: liberty, equality, fraternity. According to Backes and Jesse, left-wing radicalism (Communism) overemphasises equality to the detriment of freedom whereas traditional European right-wing radicalism (Fascism) as well as American right-wing radicalism disregards equality in favour of either fraternity or liberty. In Backes’ and Jesse’s view, centrism is conducive to liberal democracy while radicalism is a necessary and sufficient condition for extremism. In a sense, the 20^th century view of radicalism has come full circle in their work, which has influenced many European scholars directly or indirectly. However, empirical evidence for the U-shaped link between radical ideological positions and opposition to liberal democracy is sparse.

Measurement issues

If radicalism is interpreted in a purely spatial sense, it simply refers to the endpoints of the ideological spectrum. The most common instrument in this context is the general left-right scale that has been employed in countless comparative and single-country studies. Since the left-right scale is still interpreted chiefly in economic terms, other, more specific scales which refer to the appropriate degree of government intervention in the economy, state control of prices and wages, or the importance of trade unions have also been used. On the other hand, more inclusive attempts at measuring radicalism include preferences on the “postmaterialist” issues such as the environment, minority rights, and direct democracy.

Logical implications of extreme positions not withstanding, most researchers would, however, agree that a position at the endpoints of any policy scale is in itself of little importance because people frequently hold inconsistent and contradictory attitudes. Therefore, a number of items and scales have been proposed to directly capture support for liberal democracy.

Arguably, the most influential amongst these were developed by Herbert McClosky in his work on democratic values. In his 1964 article, McClosky distinguishes between three sub-dimensions of democratic values: respect for the “rules of the game” on the one hand and support for freedom of expression as well as support for political, economic, social and ethnic equality on the other. McClosky’s first dimension primarily refers to formal compliance. As long as a majority of citizens has internalised these rules, they will support democratic institutions even if their grasp of the underlying principles is patchy. His second and third dimension, however, refer precisely to these principles.

A model (liberal) democrat should subscribe to both the principles and rules, whereas an anti-democrat would despise both. Real-world citizens usually find themselves somewhere in between those two poles: they agree with the rules and abstract principles, but sometimes struggle with their application. Some items on McClosky’s scale were specifically designed to capture these conflicts. For instance, 90 per cent of his respondents believed in “free speech for all no matter what their views might be”, yet 50 per cent agreed that books containing “wrong political views” did not deserve to be published and 25 per cent were ready to suspend due process for “dangerous enemies like the Communists”.

To the present day, McClosky’s work has a tremendous impact on the field, but there are some basic problems with his and all subsequent attempts to measure support for democratic values. First, the items inevitably reflect the political and historical context for which they were devised. For McColsky and many of his successors, Communism was the main threat to liberal democracy. With the advent of new ideological challenges such as Islamism and Right-Wing Populism, this is obviously not longer true. Second, the rules and sometimes even the principles that constitute liberal democracy are bound to change gradually over time. Political behaviours and issues from the New Politics agenda that were considered “radical” in the 1960s – minority rights, the environment, sit-ins and human chains etc. – are now well within the political mainstream. Therefore, finding items that work well in all countries at all times is conceptually and empirically next to impossible. Third, even if these attitudinal scales generate measurements that are valid across time and space, they lack a natural cut-off point. At best, they are able to identify the most radical persons in society. However, where the boundary lies between democrats and radicals is an entirely different question.

Kai Arzheimer (University of Essex)

See also Communism, Democracy, Theory of, Democracy, Types, Fascism, Fundamentalism, Ideology, Islamist Movements, Left-Right Spectrum, Liberalism, Peasants’ Movements, Political Attitudes, Populism, Postmaterialism

Further Readings

Backes, Uwe. 2007. “Meaning and Forms of Political Extremism in Past and Present,” Central European Political Studies Review 9 (4): 242–62.

Lipset, Seymour Martin. 1959. “Democracy and Working-Class Authoritarianism,” American Sociological Review 24: 482–501.

Lipset, Seymour M. 1960. Political Man. The Social Bases of Politics. Garden City: Doubleday.

Lipset, Seymour Martin, and Earl Raab. 1971. The Politics of Unreason. Right-Wing Extremism in America, 1790-1970. London: Heinemann.

McClosky, Herbert. 1964. “Consensus and Ideology in American Politics,” The American Political Science Review 58 (2): 361–82.