EMPIRICAL SOCIAL RESEARCH Tarik Abou-Chadi, Reto Mitteregger, Cas Mudde LEFT BEHIND BY THE WORKING CLASS? Social Democracy’s Electoral Crisis and the Rise of the Radical Right

The Friedrich-Ebert-Stiftung The Friedrich-Ebert-Stiftung(FES) is the oldest political foundation in Germany with a rich tradition dating back to its foundation in 1925. Today, it remains loyal to the legacy of its namesake and campaigns for the core ideas and values of social democracy: freedom, justice and solidarity. It has a close connection to social democracy and free trade unions. FES promotes the advancement of social democracy, in particular with: – political educational work to strengthen civil society; – think tanks; – international cooperation with our international network of offices in more than 100 countries; – support for talented young people; – maintaining the collective memory of social democracy with archives, libraries and more. Authors Tarik Abou-Chadi is Assistant Professor at the Department of Political Science at the University of Zurich and the Centre for Democracy Studies Aarau. His research focuses on elections and political competition in advanced democracies. He has published widely on social democracy, the radical right and green parties. Reto Mitteregger holds a BA and MA from the University of Zurich and is a doctoral student and assistant at the Department of Political Science of the University of Zurich. He is part of the research project”European social democracy and the trade-offs of party competition in post-industrial societies”. Cas Mudde is the Stanley Wade Shelton UGAF Professor in International Affairs and a Distinguished University Professor at the University of Georgia(US), as well as a Professor II at the Center for Research on Extremism(C-REX) at the University of Oslo(Norway). His latest books include The Far Right Today(2019) and(with Cristóbal Rovira Kaltwasser) Populism: A Very Short Introduction(2017). Responsible for this publication at the FES Jan Niklas Engels is responsible for the area”Trend and Empirical Social Research” within the Division for Analysis, Planning and Consulting of the Friedrich-EbertStiftung.

Tarik Abou-Chadi, Reto Mitteregger, Cas Mudde LEFT BEHIND BY THE WORKING CLASS? Social Democracy’s Electoral Crisis and the Rise of the Radical Right 3 EXECUTIVE SUMMARY 4  1 INTRODUCTION 6  COMMON NARRATIVE: FROM SOCIAL DEMOCRATS TO THE RADICAL RIGHT 8  APPEAL OF THE RADICAL RIGHT 9  POST-INDUSTRIAL SOCIETY AND THE SOCIAL DEMOCRATIC DILEMMA 11  WORKING CLASS IN THE TWENTY-FIRST CENTURY 16  WORKING CLASS AND THE RADICAL RIGHT 19  7 HOW HAVE SOCIAL DEMOCRATIC ELECTORATES CHANGED? 22  8 SOCIAL DEMOCRATIC RESPONSES 26  9 SUMMARY 27 10 TOWARD A SOCIAL DEMOCRACY FOR THE TWENTY-FIRST CENTURY 30 Figures and tables 31 References

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 2

LEFT BEHIND BY THE WORKING CLASS? 3 EXECUTIVE SUMMARY 1. Social democratic parties in Western Europe have been experiencing an electoral decline of historic proportions in the first two decades of the twenty-first century. At the same time, radical right parties have been experiencing unprecedented electoral success. This has paved the way for a common narrative that claims that social democratic parties have lost their working-class voters to radical right parties either because of their shift to the right on the economic dimension(“Third Way”) or because of their shift to the left on the cultural dimension(“identity politics”). This narrative is wrong in both its assumptions and its assertions. 2. The working class in Western Europe in the twenty-first century is not a monolithic white male block with authoritarian and nationalist attitudes. A considerable and growing share of the working class is female and has a migration background. A large share of the working class holds progressive positions on questions such as LGBT rights and immigration. 3. The radical right is not the new home of former social democratic voters. While radical right parties have gained proportionally more support from working class voters, they should in no way be characterized as the new workers’ parties. Only a small proportion of the Western European working class supports the radical right. In addition, only a small proportion of radical right voters used to be supporters of social democratic parties. 4. Social democratic parties have not lost voters mainly to radical right parties. Data from various sources and countries in Western Europe show that only a small minority of former social democratic voters have left for radical right parties. Social democratic parties have lost voters mainly to Green and mainstream right parties. 5. The loss of the(white) working class is not the driving force behind the electoral decline of social democratic parties. In fact, social democratic parties have disproportionately lost voters among the educated middle class. The lion’s share of more highly educated voters have left social democratic parties for Green and social-liberal parties. These parties have been particularly successful in attracting this growing class of voters by offering more progressive positions, particularly on the cultural dimension, than social democratic parties. 6. Social democrats risk further aggravating these developments if they take up left-nationalist strategies, that is, if they pivot right on the cultural dimension. These policy positions deter potential and former voters, who may vote for more progressive parties instead. Centrist strategies (on both dimensions) could be equally bad for their electoral fortunes as they blur the ideological profile of social democratic parties. In the medium term, this leads voters to choose parties that have more distinct positions on either of the two political dimensions. 7. New left and old left strategies show the highest level of support among potential social democratic voters: experimental and survey data shows that potential social democratic voters prefer old left and new left programmes over centrist and left-nationalist strategies. Based on our analysis, and given the dilemmas they are facing, combining aspects of old and new left strategies constitutes the most promising avenue forward for social democratic parties.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 4 1 INTRODUCTION “Wer die Arbeiter des Rust Belt verliert, dem werden die Hipster in Kalifornien auch nicht mehr helfen.” 1 Sigmar Gabriel There can be little doubt that social democratic parties are currently facing a fundamental electoral crisis. In the past decade, social democratic parties in several Western European countries have scored among the worst results in their electoral history. It started with PASOK, once near-hegemonic in Greece, which lost nearly three-quarters of its support in the May 2012 elections and has been able to survive since then only as part of broader electoral coalitions. Consequently,“Pasokification” has become a term used to refer to the broader decline of social democratic parties in European countries since the 2010s. Five years after PASOK, the Dutch PvdA and the French PS dropped into single digits in their national elections, losing 19.1 and 21.9 percentage points, respectively. The same year, the German SPD received its worst result in the post-war era and has since further declined in the polls. Even the British Labour Party under Jeremy Corbyn, which not long beforehand had been hailed as the potential future of European social democracy, scored its worst result since 1935 in December 2019. Looking at the reactions to social democratic parties’ electoral decline among commentators and pundits, a common narrative has emerged to explain the current electoral crisis. The basis for this narrative is exemplified in Figure 1, which shows the development of vote shares of social democratic parties and radical right parties in Western Europe since 1990. While social democratic parties have lost a lot of votes in the past 30 years, radical right parties have gained substantial electoral support in the same period and have become an established political force across the region. Looking at these two developments together, it is not surprising that commentators have made a connection between the decline of social democratic support and the rise of the radical right. 1 If you lose workers in the Rust Belt, the hipsters in California won’t be able to help you.

LEFT BEHIND BY THE WORKING CLASS? 5 Vote share (parliamentary elections) Figure 1 Average Vote Share for Social Democratic and Radical Right Parties in Europe, 1990 – 2020 30 20 10 Year 1990 Source: parlgov.org 1996 2002 2008 2014 2020 Radical-Right Social-Democratic

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 6 2 THE COMMON NARRATIVE: FROM SOCIAL DEMOCRATS TO THE RADICAL RIGHT The common narrative about the decline of social democratic parties can be summarized as follows: social democratic parties have seen a strong decline in their vote share because the working class, their former core constituency, has abandoned them and has found a new home with the radical right. The explanation of why working class voters have abandoned social democratic parties usually comes in two different flavours: economic and cultural. The first narrative points to the economic and social policy changes that happened in the era of Britain’s New Labour and die Neue Mitte(New Centre). Pursuing the idea of a Third Way, politicians such as Tony Blair and Gerhard Schröder shifted their parties’ policies to the economic centre. While there is an ongoing debate about the policy consequences of Third Way policies, many commentators see the electoral decline of social democratic parties as a working class backlash against the“neoliberal” economic policies of the New Labour era. A second narrative emphasizes social democratic positions on cultural issues such as immigration, European integration, gender equality and the environment. Working class voters, so the narrative goes, hold decidedly more authoritarian and nationalist positions on these issues. As social democratic parties are now associated with progressive cultural positions, working class voters have rejected them and turned instead to radical right parties, which promise tougher policies on immigration, less global integration and the protection of ­so-called“family values”. Both the economic and the cultural narrative thus frame the electoral crisis of social democratic parties as a response to shifts in their policy positions that have alienated working class voters who, in turn, have switched their support to the radical right. These narratives are now so widespread that they have become received wisdom in much of the media, politics and even some academic work. Every time a social democratic party loses an election, a flood of articles, blog and social media posts are published that reinforces these narratives. As they are brought forward not only by pundits and commentators, but also by(former) politicians, these alleged explanations have had a major impact on the public discourse and on the(declining) electoral fortunes of social democratic parties in Western Europe. In addition to advancing their proposed narrative of electoral decline, these contributions often, implicitly or explicitly, include a call for action to halt this development. In interviews with The Guardian(22 November 2018), Hillary Clinton, Tony Blair and Matteo Renzi all sounded the same note:“Europe must curb immigration to stop rightwing populism”, in Clinton’s words. According to his former head of the European Secretariat, Sir Stephen Wall, Blair ­already identified immigration as his Achilles heel in 2001, allegedly saying“The one thing that could lose me the next election is immigration”(Watt and Wintour 2015). This view is hardly confined to Clinton and Blair, however. Many US ­liberals have made similar statements since Trump’s unexpected victory. New York Times columnists, from Frank Bruni to Thomas Edsall, were at the forefront of this pushback from“woke” liberals. In fact, Clinton herself has a long track record, in both speech and votes, against illegal immigration and for“border security,” including a“fence” at the border with Mexico(Kruzel 2018). But many prominent European social democrats have also been calling for“immigration realism” since the late 2000s. In recent years, for instance, Belgian SP.A chairman John Crombez has stated, that“migration to Europe must decrease” ( Nieuwsblad, 6 January 2019), while former SPD leader Andrea Nahles has said that Germany“cannot accept all” asylum seekers that arrive at its borders( Der Spiegel, 26 May 2018). None has taken it as far as the Danish Social Democrats, who are becoming increasingly less distinguishable from the far right Danish People’s Party(DF), as they decry alleged “parallel societies” of Muslim immigrants in Denmark, and argue that immigration“undermines” the Danish welfare state(Agustín and Jørgesen 2019). Probably, no one has summarized this narrative as well as former SPD leader and German Minister of Foreign Affairs Sigmar Gabriel, who, in an article in Der Spiegel(18 December 2017) wrote:“If you lose workers in the Rust Belt, the hipsters in California won’t be able to help you.” This narrative, and the calls for action based on it, has also been supported by a number of academics and think-tankers, who have offered similar advice: the centre-

LEFT BEHIND BY THE WORKING CLASS? 7 left has to take care again of the“left behinds” and“somewheres” by limiting immigration and departing from an emphasis on post-material policy positions that emphasize gender equality, LGBT rights or the environment(Goodhart 2017; Lilla 2018). Some academics now even go so far as to openly defend white identity politics(Kaufmann 2018). Interestingly, even though supported by a number of academics, there exists very little work by these scholars that is academic and provides empirical evidence to support these claims. In fact, as we will outline later, most academic work speaks against this narrative. In this report, we investigate the political competition ­between social democratic parties and radical right parties in Western Europe. We empirically investigate the three core assertions of the dominant narrative. First, is the radical right really the new working class party? Second, have social democratic parties really lost primarily, or at least disproportionately, working class voters? And, third, have(working class) voters really left social democratic parties for radical right parties? We find that the transformation of the Western European political space indeed constitutes a dilemma for social democratic parties. Because their former core constituency continues to shrink, social democratic parties have been forced to appeal to other socio-demographic groups to try to retain the same size of the electorate. Building these new alliances inevitably comes with trade-offs: appealing to new groups necessarily entails the risk of alienating old ones. However, these trade-offs do not necessarily play out in the way described in the dominant narrative. This is in part because the narrative wrongly assumes that the working class is a monolith, based on the stereotypical“working class authoritarian” white man of the twentieth century(see Lipset 1960). We demonstrate that the working class today is a sociodemographically diverse group and holds a broad range of attitudes, not limited to authoritarian and nativist positions. While the radical right does appeal to a considerable share of the working class, in no way does a majority of the working class support a radical right party in Western Europe. Moreover, with regard to the claimed competition between radical right parties and social democratic parties, we show that only a small share of today’s radical right voters previously supported social democratic parties. Similarly, social democratic parties have lost relatively few of their voters to the radical right. In sharp contrast to the dominant narrative, a large share of former social democratic voters have gone to more progressive parties, such as the Greens or social liberal parties. We demonstrate that more restrictive positions on immigration, as well as generally less progressive positions, do not help social democratic parties to attract more working class and lower educated voters. However, they do alienate a larger share of their potential electorate. Our findings show that the largest potential for social democratic parties lies in economically left-wing and culturally progressive positions. Before discussing in detail the three building blocks of this narrative of social democratic decline, we shall briefly outline(i) the appeal of radical right parties and(ii) the sociostructural transformations within European societies that form the basis of the social democratic dilemma.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 8 3 THE APPEAL OF THE RADICAL RIGHT In order to understand why people support radical right parties, it is necessary to distinguish two parts of their appeal. On the one hand, these parties appeal to voters through their policy positions on issues such as crime, corruption, ­immigration and multiculturalism. On the other hand, their potential to mobilize anti-elite sentiment lends them additional appeal. The core ideology of(populist) radical right parties consists of a combination of nativism, authoritarianism and populism(Mudde 2007). Nativism combines nationalism and xenophobia, which is best captured by the infamous slogan “Germany for the Germans: Foreigners Out!”. Nativists strive for a monocultural country, in which only“natives” live(or have full citizenship) and where everything“alien”(that is, non-native), be it ideas or people, is seen as threatening. ­Authoritarianism is the belief that the state should strictly enforce order upon a naturally chaotic population by teaching discipline in schools and imposing severe law and order policies. Populism, finally, is an ideology that sees society as ultimately divided into two homogeneous and antagonistic groups, the pure people and the corrupt elite, and that believes that politics should generally follow the“will of the people”. At its core, radical right ideology is monist and rigid, seeing simplistic and“natural” distinctions between people. This puts it in direct opposition to the dynamic pluralism of liberal democracy. While it has long been argued that the radical right is a“normal pathology” of Western democracies – that is, a set of values that are fundamentally detached from mainstream values and supported by only a small minority of the population – it is better seen as a“pathological normalcy”, that is, a radicalization of the mainstream(Mudde 2010). Many surveys show that large minorities, and sometimes even majorities, support some of the main positions of the radical right on issues such as crime, European integration, immigration and the political establishment. In most ­European countries, populist radical right parties have so far tapped into only an(admittedly growing) part of their potential electorates. In the past two decades, the core issues and frames of the radical right have become part of mainstream politics in many European countries. The most painful expression of this process of mainstreaming and normalization is the framing of immigration as a“threat” to national(and European) identity and security, particularly in the wake of the so-called “refugee crisis” of 2015–2016(Mudde 2019). For instance, the 2019 manifesto of the mainstream right European People’s Party(EPP) discussed immigration under the heading“A Europe that preserves our way of life”, while the new European Commission tried to introduce a“Commissioner for Protecting Our European Way of Life”, who would be responsible for, among others, immigration policy. While the mainstreaming of their issues and frames has helped to increase the salience of some of the key issues of radical right parties, notably immigration and security, it has only partially benefitted these parties electorally. First of all, radical right parties now have electoral competition from mainstream(right-wing) parties, which at times present light versions of their policies and programmes. Second, despite a right-wing turn of the political debate, the population is turning left on many key issues, in particular on immigration and integration. This is largely a generational effect, related in part to education(O'Grady 2020). That said, in many European countries radical right parties remain the preferred option for voters who hold negative attitudes towards diversity, European integration, immigration and immigrants, or the political establishment. Despite the copy-cat behaviour of mainstream parties, they continue to prefer the original over the copy, as FN leader Jean-Marie Le Pen already noted many decades ago.

LEFT BEHIND BY THE WORKING CLASS? 9 4 THE POST-INDUSTRIAL SOCIETY AND THE SOCIAL DEMOCRATIC DILEMMA To understand the competition between social democratic and radical right parties, as well as the electoral crisis of social democratic parties more generally, it is necessary to evaluate these developments against the backdrop of the socio-structural changes of post-industrial societies. Western European societies have been – and still are – experiencing distinctive transformations of their occupational and social structures that have led to a substantial shift in the political space(Beramendi et al. 2015b; Gingrich and Häusermann 2015; Inglehart 1977; Kitschelt 1994). Overall, this transformation has been characterized by three major developments. First, educational expansion, with increasing numbers of people going into higher education. This has led to an average increase in both educational levels and wealth. Second, a still ongoing tertiarization – driven by processes of globalization, automation and digitalization – has led to a“de-proletarianization” of the workforce. Third, family roles and gender norms have changed substantially since the end of the Second World War. With large numbers of women entering the labour market, a progressive feminization of the workforce has been taking place(Beramendi et al. 2015a). Altogether, this has led to a substantial transformation of these countries from industrial into“post-industrial” democracies. While the number of well-educated people working in the service sector has increased significantly over recent decades, both the relative and the absolute size of the working class has been shrinking continuously. But the working class has changed not only in terms of relative size but also in term of socio-structural composition. As a consequence, the working class today is significantly more diverse than in the 1960s and 1970s(see below). These processes have transformed not only the occupational structure but also the political space in post-industrial societies. They have fundamentally changed political values, as well as the salience of political dimensions. Among so-called “socio-cultural professionals”(Oesch and Rennwald 2018), who are often well educated and used to interacting with others, as well as open-minded and sensitive to others, material concerns are not the most important political issues. ­Instead, they often focus on post-materialist values, such as environmental and feminist issues, which have consequently entered the political agenda of all advanced capitalist democracies and continue to gain in relevance(Häusermann and Kriesi 2015; Kitschelt and Rehm 2014). Therefore, party competition and voter attitudes no longer move along a single left–right, economic dimension in these countries in the twenty-first century. Several studies have identified at least two dimensions that are relevant for the positions of both voters and parties. First, the economic dimension(often also labelled“left–right”) divides voters and parties into pro-redistributive, state-interventionist proponents(left), and those preferring the“free” allocation of goods and market liberalism(right). Second, a cultural dimension(often labelled“liberal–authoritarian”) has become an equally salient cleavage in post-industrial societies. While the left–right dimension refers mainly to questions about economic distribution, the libertarian–authoritarian dimension concerns issues related to societal organization. By cross-cutting the existing left–right dimension, it has transformed the unidimensional political space into a two-dimensional space(Bornschier 2010; Hooghe, Marks, and Wilson 2002; Kitschelt 1994). The growing importance of the latter political dimension, with its“liberal-progressive” and“authoritarian-conservative” poles, has also laid the foundation for new political parties that have entered the political arena in most West European countries. Green or New Left parties offer political programmes promoting the liberal-progressive pole of this new dimension. They have come to attract a big part of the growing group of socio-cultural professionals(Oesch and Rennwald 2018). Even though there are considerable n­ ational differences, social democrats have also adopted many of these positions, while simultaneously trying to stick to their main issues of economic justice and redistribution. However, they are often less outspoken and more ambiguous than their“new left” counterparts on these cultural positions(AbouChadi and Wagner 2019; Kitschelt 1994). The success of the radical right can be regarded as the mirror image of, and counterreaction to, the ascent of Green and New Left parties(Bornschier 2010; Ignazi 1992). As ­outlined above, radical right parties also prioritize the cultural ­dimension over the economic one but seek to occupy its

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 10 ­authoritarian-conservative pole instead. In promoting issues and positions linked to this pole, such as Euroscepticism and opposition to immigration, they have managed to disproportionally attract voters belonging to the working class and the “petit-bourgeoisie”(Oesch 2008; Oesch and Rennwald 2018). For social democratic parties, the transformed political space has created a number of electoral trade-offs. In order to remain electorally relevant, they have had to adjust their positions to appeal to new groups of voters. The core tradeoff in this changing political environment, then, results from the question of how much social democratic parties can appeal to new electoral groups that are more educated, work in socio-cultural professions, and live in urban regions, without alienating their former core constituency, the working class(Abou-Chadi and Wagner 2019; Kitschelt 1994). Looking at this trade-off, it seems reasonable to assume that radical right parties constitute an attractive alternative for social democratic supporters, especially among the working class. However, it is ultimately an empirical question of how much the radical right has attracted working class and former social democratic voters. We know that some in the working class have always supported parties on the right rather than the mainstream left(Lipset 1960). So, it is equally plausible that today’s radical right support largely stems from this group. To understand these dynamics better, it is essential to reassess what“working class” means in the twenty-first ­century.

LEFT BEHIND BY THE WORKING CLASS? 11 5 THE WORKING CLASS IN THE TWENTY-FIRST CENTURY Many narratives around the decline of social democratic parties focus on the working class or lower educated voters. In both the material and the cultural alienation story workers have turned their backs on social democratic parties in response to their policy choices. Perceived as the main benefactors of this shift, radical right parties are now increasingly presented as the new workers’ parties, or as“working class parties 2.0”(Arzheimer 2018). These accounts often portray the preferences of the working class through a very specific lens, that of white, male industrial workers. However, while this focus might have been justified in the 1960s, the socioeconomic transformations of the past half century have fundamentally reshaped the working class. In this section, we outline what it means to be working class in the twentyfirst century. What is their socio-demographic profile? And what policy preferences do members of the working class have? Although the term“working class” has been used for over a century, its meaning has always been contested and its structure shifting. There are both objective and subjective interpretations of“working class” and the two do not always overlap. Simply stated, a lot of people have“working class jobs” but do not self-identify as“working class”, while there are also people who self-identify as working class but do not have working class jobs. In other words, in life and politics, (working)“class” is as much a self-defined identity as it is an externally ascribed category of employment. The common(Marxist) understanding of working class has always been relatively narrow and refers mainly to industrial workers, such as dockers, factory workers and miners. The popular image of“the worker” was a white man, which has always been at best only partly accurate, as a lot of(white) women worked in factories, too(particularly during the wars). The so-called Post-Industrial Revolution of the 1960s, which shifted the focus of the economy from industry to services, did not just change the(absolute and relative) size of the working class in Western Europe, which halved between 1970 and 2017, but also profoundly changed its composition(Benedetto, Hix, and Mastrorocco 2020; Beramendi et al. 2015a). In this report, we mainly use data from the European Social Survey(ESS), one of the leading social science survey projects in Europe, which is often described as the gold standard in terms of its sampling and methodology. 2 Following a broader categorization of eight class groups(Oesch 2006), we define the working class in this paper as a combination of both“production workers” and“service workers”. 3 Simplifying this distinction only slightly, production workers can be seen as the “old working class”, reflecting both the economic model of the Industrial Era and the still dominant, if outdated, popular representation of the working class. This“old working class” is predominantly male and white in most countries but has become more diverse in terms of ethnicity and gender since the mid-twentieth century. In this slightly simplified understanding, service workers can be seen as the“new working class”, that is, the workers of the Post-Industrial Era. They are the workers of the now dominant tertiary sector, active in service industries such as customer services, health care and hospitality. Service workers have always been much more diverse than production workers, traditionally including a sizeable group of women and increasingly also of ethnic minorities. Figure 2 demonstrates this diversity of the working class in four West European countries. 4 It shows how production and service workers differ in their composition, based on gender and migration background(being born in a foreign country or having at least one such parent). First, there is indeed a strong gender difference between production and service workers. Whereas a majority of production workers are male, a majority of service workers are female. Second, it shows that a considerable(and growing) share of the working class 2 For this project, we either use all or the most recent of the 9 waves of the ESS, which include the years between 2002 and 2018, and focus specifically on Northwestern Europe, a region that has most profoundly experienced the socio-economic and political developments that we described in the previous section. Our findings are thus based on the following countries: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), France(FR), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE). 3 The eight“Oesch-Classes” are as follows: Self-employed professionals, small business owners,(associate) managers, socio-cultural(semi-) professionals, technical(semi-) professionals, office clerks, production workers, and service workers. 4 Using data from the 2018 round of the European Social Survey in four countries(CH, DE, FRA, NL)

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 12 Figure 2 Migration background and gender in class groups France Production workers Service workers % 0 20 40 Netherlands Production workers Germany Production workers Service workers 60% 0 20 Switzerland Production workers 40 60 Service workers % 0 Service workers 20 40 60% 0 10 20 30 40 Men, No mig. back. Men, mig. back. Women, No mig. back. Women, mig. back. (Reading example: Almost 20 per cent of production workers in Germany are male with a migration background, while over 40 per cent of production workers in Switzerland are male with a migration background) Source: Migration background and gender among service workers and production workers(CH, DE, FR, NL; ESS 2018) has a migration background, topping 25 per cent in both categories in all four countries. In Switzerland, a majority of both production and service workers have a migration background. The working class also varies strongly in its attitudes. Implicit in the“common narrative” is that the working class is a monolithic block supporting authoritarian/nationalist positions. Interestingly, in his seminal work on working class ­authoritarianism in the mid-twentieth century, American political sociologist Seymour Martin Lipset already pointed to the diversity of the working class in terms of their preferences on non-economic issues, including internationalism and immigration(Lipset 1959; 1960). But while he noted that more conservative or authoritarian positions can be found among the working class, he also clarified that the working class as a whole is not authoritarian. Additionally, and important to this paper, Lipset(1959) argued that working class authoritarian voters are not much represented within the electorates of social democratic parties. In the following section, we present the political preferences of the contemporary working class. We focus on second-­ dimension issues, with regard to which supposedly nativistauthoritarian attitudes are a large driver behind the changing political allegiances of the working class. We focus on two sets of issues that are representative of the main conflicts on this dimension, attitudes toward homosexuality and attitudes toward immigration(Kitschelt and Rehm 2014). All the following graphs present the findings for service and production workers(in the left panel), as well as attitudes based on a respondent’s education(right panel). Figure 3 shows attitudes of production and service worker toward homosexuality. These attitudes concern a typical issue on the cultural dimension and correlate strongly with other socalled“morality issues” as well as attitudes on gender equality. We use the item“Gays and lesbians should be free to live the way they wish,” which has been used in a multitude of studies investigating attitudes toward homosexuality(Abou-Chadi and Finnigan 2019).

LEFT BEHIND BY THE WORKING CLASS? 13 Figure 3 Gays and lesbians should be free to live the way they wish Class Production workers Education Lower Sec Service workers % 0 10 20 30 40 Higher Sec Tertiary % 0 Strongly Agree 2 3 4 Strongly Disagree 20 40 60 (Reading example: Over 30 per cent of production workers strongly agree with the statement that gays and lesbians should be free to live the way they wish, while around 5 per cent of production workers strongly disagree with the statement) Source: Attitudes of educational groups and workers towards homosexuality(pooled data from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), France(FR), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE); ESS 2002–2018) We can see that working class respondents hold broadly positive attitudes toward the gay and lesbian population. More than three-quarters of both production workers and service workers(strongly) agree with the statement that gays and lesbians should be free to live the way they wish. Only a small minority disagree with the statement. Support is somewhat higher among service workers, confirming longstanding research that this group generally holds more progressive positions than production workers(Kitschelt and Rehm 2014). In the right panel we can see how attitudes toward homosexuality are distributed over educational groups. Unsurprisingly, attitudes are more positive among the more educated. However, while we can see that those who strongly agree are more represented among those with tertiary education, other differences are quite marginal. Most important for this study, we see that respondents with lower levels of education also hold largely positive attitudes toward the gay and lesbian population. In fact, almost 80 per cent(strongly) agree with the statement that gays and lesbians should be free to live the way they wish. We now move on to attitudes toward immigration. Immigration has not only become a defining issue of the new dividing line in the politics of post-industrial societies, but attitudes toward immigration also constitute the main determinant of voting for the radical right(Ivarsflaten 2007; Mudde 2007; Rooduijn 2018). Figure 4 shows how the various groups evaluate the statement“A country’s culture is undermined or enriched by immigration”, while Figure 5 does the same for the statement“Immigration makes the country a worse or better place”. Both items are widely used to study attitudes toward immigration and have been demonstrated to be strong predictors of radical right voting. While attitudes toward immigration are less progressive than those toward the gay and lesbian population, we see far from uniformly authoritarian-nativist attitudes within the working class. In fact, over 50 per cent of service workers and around 45 per cent of production workers(strongly) agree that a country’s culture is enriched by immigration. In both cases, the percentage of workers who think that a country’s culture is undermined by immigration is(significantly) lower. This is slightly different with regard to the question of whether immigration makes a country a better or worse place to live in, however. More production workers think it makes a country a worse place, while for service workers views are about even. Still, only minorities of both groups think immigration makes a country a worse place; roughly 40 per cent of production workers and some 35 per cent of service workers. In line with decades of social science research, we find that attitudes toward immigration strongly correlate with education, with people with the highest level of education (tertiary) showing the most favourable attitudes towards immigration. That said, only a minority of respondents with the lowest level of education(Lower Sec) thinks that a country’s

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 14 Figure 4 Country's culture underminded or enriched by immigration Class Production workers Education Lower Sec Service workers % 0 5 10 15 20 25 Higher Sec Tertiary Undermined 2 3 4 Enriched % 0 10 20 30 40 50 (Reading example: 20 per cent of production workers strongly agree with the statement that the country’s culture is enriched by immigration, while around 15 percent of production workers strongly disagree with the statement). Source: Attitudes of educational groups and workers towards immigration(1)(pooled data from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), France(FR), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE); ESS 2002–2018) Figure 5 Immigration makes country a worse or better place Class Education Production workers Lower Sec Service workers % 0 Higher Sec 10 20 30 Tertiary Worse 2 3 4 Better % 0 10 20 30 40 (Reading example: Around 15 per cent of servie workers strongly agree with the statement that immigration makes a country a better place, while around 15 per cent of service workers strongly disagree with the statement Source: Attitudes of educational groups and service/production workers towards immigration(2)(pooled data from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), France(FR), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE); ESS 2002–2018)

LEFT BEHIND BY THE WORKING CLASS? 15 culture is undermined by immigration or that immigration makes a country a worse place. In short, the working class constitutes an increasingly diverse and progressive group in the twenty-first century. Although women and people with a migration background are more prominent among the“new working class” of service workers, where they collectively constitute a majority, a significant minority of the“old working class” of production workers is neither white nor male. And in terms of attitudes, the vast majority of both production and service workers hold progressive views on LGBT issues, while a large share holds progressive or neutral(and therefore not very strong) attitudes on immigration.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 16 6 THE WORKING CLASS AND THE RADICAL RIGHT Former Trump advisor Steve Bannon used to boast that he and his(former) boss had turned the Republican Party into a “working class party”( The Guardian, 17 December 2019). He hereby confirmed a broader narrative that has been increasingly popular since the 1990s, at least in Western Europe, in which radical right parties are portrayed as the new representatives of the working class. This is in line with academic research that showed that, as some radical right parties started to gain better results in the 1990s – between 5 and 15 per cent – their electorates started to develop a slightly more distinct profile. The few moderately successful radical right parties of the early 1980s had attracted both blue-collar and white-collar workers, but had been particularly strong among the self-employed and small business owners(the so-called“petite bourgeoisie”). It was the German political scientist HansGeorg Betz who first noted a process of“proletarization”, in which the more successful populist radical right parties, such as the French FN and the Austrian FPÖ, were able to hold on to their lower middle class voters, while increasing mainly their working class support(Betz 1994). This process of proletarization was followed and strengthened by a shift in the propaganda of radical right parties, from instrumental support for neoliberal policies(such as lower taxes and privatization) to staunch support for welfare chauvinism, that is, social benefits should go(only) to“our own people”(De Lange 2007). As a consequence of this programmatic change, the more successful radical right parties were able to attract a disproportionate number of working class voters(Harteveld 2016). In both academic and popular writings radical right parties became increasingly referred to as“workers’ parties” or “a new type of working-class party”(Oesch 2012: 48). The stereotypical radical right voter in Western Europe was now young(ish), lower educated, male and working class (Arzheimer 2018). Moreover, today’s radical right supporters are often portrayed as the former support base of social democratic parties. However, is this an accurate representation of the radical right’s electorate? In this section, we first document the class basis of the radical right electorate and then show where their voters came from. Based on the already described class scheme developed by Oesch(2006), Figure 6 shows patterns of party support for different class groups from the European Social Survey of 2002–2018. 5 For each class it shows which party receives the highest level of support. We can see that there is some support for radical right parties in all class groups but there is also considerable variation. Support for radical right parties is lowest among socio-cultural professionals, that is, people who work in interpersonal occupations, often have higher levels of education, but are not managers or part of the entrepreneurial elite. The three class groups that show the highest level of radical right support are indeed the groups already mentioned: small business owners and the old and new working class, that is, production and service workers. However, it is important to note that while(new and old) workers vote somewhat disproportionately for radical right parties, these parties should in no way be described as ”workers’ parties”. Social democratic parties remain the party family that receives the highest support among both production and service workers. The second largest(and in some countries the largest) group among working class voters votes for mainstream right parties, such as Christian democratic and conservative parties. Hence, it is very clear that the radical right is in no way the voice of the working class. Their increasing success has indeed been based on larger appeals to this class but overall, only a small minority of working class voters(roughly 15 per cent) has been supporting radical right parties in Western Europe in the twenty-first century. Because radical right parties have seen an increase in working class support while social democratic parties have lost support, it is not surprising that many commentators have regarded these new radical right voters as former social democratic voters. But how does this pattern look empirically? Where do these new radical right voters come from? 5 Pooled data, that is, we made one dataset from all countries and years, to maximize the number of voters.

LEFT BEHIND BY THE WORKING CLASS? 17 Figure 6 Party Choice by Class Group Party choice by class group Self-employed professionals Small business owners (Associate) managers Office clerks Technical professionals Production workers Socio-cultural professionals Service workers % 0 20 40 60 80 100 Social Democrat Mainstream right Radical right Green Radical left (Reading example: 40 per cent of service workers voted for the mainstream left, while 30 per cent voted for the mainstream right) Source: Party choice by class group(pooled data from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), France(FR), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE); ESS 2002–2018) Figure 7 Vote Switching to Radical Right Parties since 2000 Vote switch from Social Democrat Mainstream right Liberal Green Radical left Other Non voting % 0 10 20 30 40 (Reading example: Almost 40 per cent of the voters who switched to a radical right party voted for the mainstream right before they switched) Source: Vote switching to radical right parties(National Election Studies 2000–2018 from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE))

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 18 Figure 8 Vote Switching to Radical Right Parties since 2000 Below UpSec UpSec/Voc Tertiary % 0 20 40 60 80 100 Social Democrat Mainstream right Other Non voting Radical right Green Radical left (Reading example: Less then 20 per cent of voters who switched to the radical right and who have an educational degree below upper-secondary education had previously voted for the mainstream left) Source: Vote switching to radical right parties by educational groups(National Election Studies 2000–2018 from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE)) Figure 7 shows the different party families that new voters of radical right parties supported in the election before they switched to the radical right. 6 We can see that the largest share of voters that switched to radical right parties in the past 20 years previously supported a mainstream right party. The second largest group are non-voters. Less than 20 per cent of new voters of radical right parties switched from a social democratic party. Hence, the idea that radical right parties largely drew support from disenchanted former social democrats does not find any empirical support. It might be of course that voters do not switch to radical right parties directly but abandon social democratic parties, first for a mainstream right party, or abstain and only later switch to a radical right party. This question is much more difficult to investigate empirically as it requires long-running panel data to track the same people over a longer time period. Bischof and Kurer(2021) use data from the German Socio Economic Panel(SOEP)(as well as panel data in Switzerland and the United Kingdom) to achieve exactly that. Germany is an ideal case to test this question as AfD support is a fairly recent phenomenon, so we can track the voting history of these mostly new voters and see which parties they used to support in previous elections. Bischof and Kurer (2021) show that only a very small share(less than 20 per cent) of 2017 AfD voters previously supported the Social Democrats. Thus, even when using panel data to track voters over time, we do not find any evidence that today’s radical right voters are former supporters of social democratic parties. Moreover, Figure 8 shows that these patterns largely do not differ over educational groups. Among voters with low levels of education, too, we find that only a small share of radical right voters previously supported a social democratic party. So, in sharp opposition to the assertions and assumptions of the common narrative, we have shown that the contemporary working class(1) is not a monolithic block(of white men);(2) does not overwhelmingly hold authoritarian-nativist positions; and(3) does not primarily support the radical right. In reality, the working class is increasingly diverse in terms of gender and ethnicity, holds outright progressive views on the LGBT community and diverse views on immigration(only a minority hold anti-immigration attitudes). And while radical right parties do disproportionately draw support from working class voters, a large majority of the working class supports other parties. In addition, the largest share of today’s radical right voters come from mainstream right parties or from those who previously did not vote, not from social democratic parties. 6 In order to answer these questions, we rely on two different data sources, as well as existing research. Building on Abou-Chadi and Wagner (2020b), we use data from national election studies in several countries to analyse vote switching from one election to the next. For the period 2000 to 2018, we can thus document which parties the new voters of the radical right supported before they switched

LEFT BEHIND BY THE WORKING CLASS? 19 7 HOW HAVE SOCIAL DEMOCRATIC ELECTORATES CHANGED? The common narrative assumes that the loss of lower educated and working class voters to radical right parties lies at the heart of the electoral crisis of social democratic parties. We have already documented, however, that former supporters of social democratic parties constitute only a small part of the radical right electorate. In this section we shift the focus and look at the electorates of social democratic parties. Specifically, we show how social democratic electorates have changed in the past twenty years, the period that has seen the most severe decline in electoral support for social democratic parties. Figure 9 Changes in Social-Democratic Party Support since 2000 by Educational Groups The upper graph presents the size of different educational groups in the social democratic electorate and their change in party support. The lower graph shows the educational groups as a share of the total group by education. Below UpSec UpSec/Voc Tertiary % 0 10 20 30 40 50 Below UpSec UpSec/Voc Tertiary % 0 20 40 60 80 100 Stay Attract Leave (Reading example Graph 1: Since 2000, around 30 per cent of social democratic voters have an upper-secondary/vocational educational degree and still vote for the social democrats, while around 5 per cent have a tertiary educational degree and have left the social democrats.) (Reading example Graph 2: Since 2000, 60 per cent of social democratic voters with an upper-secondary/vocational educational degree still vote for the social democrats, while around 25 per cent of those with a tertiary educational degree have left the social democrats.) Source: Changes in social democratic party support since 2000 by educational groups(National Election Studies 2000–2018 from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE))

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 20 Using vote switching data described earlier, we look at which socio-demographic groups have stayed with, abandoned or joined social democratic parties in the past two decades (Abou-Chadi and Wagner 2020b). We are again particularly interested in the class structure of the voters who have abandoned social democratic parties. Unfortunately, the data from national election studies that allow us to investigate vote switching do not include information on occupation, so we rely on education instead. Figure 9 presents the same information in two different ways. The upper panel shows the size of the different groups (stayers, leavers and joiners) in the social democratic electorate by level of education. We can see that people with upper secondary education and/or vocational training make up the biggest share of the social democratic electorate in the twenty-first century, with about similar shares for people with below secondary education and tertiary education. First, we can observe that in all groups social democratic parties have lost(far) more voters than they gained. Second, in absolute terms, social democratic parties have lost most voters among the middle category in terms of education, not among lower educated voters. The lower panel shows the same groups as a share of the total group by education. Irrespective of group size, it shows us the groups in which social democratic parties have lost the highest share of voters. Again, we can see that proportional losses are quite evenly distributed over all groups. Interestingly, and against much of the public narrative, we can see that social democratic parties have lost most among the highly educated and least among the low educated. Hence, we find no support for the idea that the decline of social democratic parties is driven primarily by less educated voters. On the contrary, social democratic losses are, in absolute and relative terms,(much) higher among people with higher levels of education. Another related question concerns which parties social democratic parties have lost their voters to. As we have shown in the previous section, in sharp contrast to the common narrative, radical right voters do not come primarily from social democratic parties. But this does not necessarily mean that social democratic parties do not disproportionally lose votes to radical right parties. Figure 10 shows the parties to which social democratic voters have moved. Several points seem noteworthy. First, social democratic parties have lost voters in all directions – no one type of transition can be singled out as solely responsible for their decline. Second, social democratic parties have lost by far the most voters to mainstream right and green/left-libertarian parties. Especially in the 2010s, social democratic parties lost the largest share of former supporters to more progressive parties(Abou-Chadi and Wagner 2020b). Third, voteswitching to the radical right is responsible for only a very small share of the losses of social democratic parties. From the people who have left social democratic parties for other parties in the first two decades of the twenty-first century, only a little more than 10 per cent switched to the radical right. Figure 11 shows that these patterns are roughly the same for the different educational groups. As might be expected, people with tertiary education primarily left social democratic parties for Green and New Left parties. However, even among the less educated, switching to the radical right constitutes only a small part of the losses of social democratic parties (less than 5 per cent). Figure 10 Vote Switching from Social Democratic Parties since 2000 Vote switch to Mainstream right Liberal Green Radical left Radical right Other % 0 10 20 30 (Reading example: Since 2000, 30 per cent of social democratic voters who switched from the social democrats to another party switched to a Green party.) Source: Vote switching from social democratic parties(National Election Studies 2000–2018 from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE))

LEFT BEHIND BY THE WORKING CLASS? 21 Figure 11 Vote Switching from Social Democratic Parties by Educational Groups Below UpSec UpSec/Voc Tertiary % 0 5 10 15 Mainstream right Liberal Green Radical left Radical right Other (Reading example: Since 2000, almost 15 per cent of voters who switched from the social democrats and had an upper-secondary/vocational educational degree switched to a Green party) Source: Vote switching from social democratic parties by educational groups(National Election Studies 2000–2018 from: Austria(AT), Switzerland(CH), Germany(DE), Denmark(DK), Finland(FI), the United Kingdom(GB), the Netherlands(NL), Norway(NO), and Sweden(SE))

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 22 8 SOCIAL DEMOCRATIC RESPONSES We have shown that(1) working class preferences should not be reduced to authoritarian/nationalist positions; that(2) social democratic parties have not disproportionally lost working-class voters; and that(3) only a small share of former social democratic voters have switched to the radical right. We will now address the main question that we can derive from the common narrative: how do social democratic policy positions affect their own electoral support and which ones potentially affect voter flows to the radical right? What different strategies could social democratic parties pursue and which ones do they propose? Following previous research(Abou-Chadi et al. 2021), we distinguish four ideal-typical issue profiles for social democratic parties: new left, old left, centrist, and leftnationalist(see Table 1). As discussed, one popular narrative on the electoral decline of social democratic parties identifies the supposedly progressive-liberal shift from social-democratic parties on the liberal-authoritarian dimension as the major reason for their electoral decline. In this view, the adaption by social democratic parties of progressive positions on“new” topics such as immigration and integration, gender equality, and LGBTQ*-rights is thought to have alienated the working class(Eatwell and Goodwin 2018; Lilla 2017).” While these voters may still be attracted to the economic positions of social democratic parties, they prefer to vote for radical right parties because of their more authoritarian and nativist stances on the increasingly more salient second dimension. Thus, some observers have argued that social democratic parties should take up a left-nationalist position. By doing so, the argument runs, they would be able to win back their lost electoral backbone, the(white) working class, which is believed now to constitute the electoral core of the radical right. The rise of radical right parties is thus seen not only to correlate with the social democratic electoral decline, but also as a cause of it. Accordingly, social democratic parties have been urged to take up less progressive positions regarding immigration or to at least downplay the importance of the cultural dimension. This is particularly displayed in a narrative that claims that“identity politics” is the main cause of the social democratic decline Berman 2016; Berman and Snegovaya 2019; Lilla 2017, 2018). In the mirror-image of the left-nationalist approach, proponents of a new left strategy argue that social democratic parties should take up progressive stances to attract the growing group of highly-educated voters, as well as parts of the working class that are not authoritarian-nativist(including Table 1 Idealtypical Issue Profiles for Social Democratic Parties Electoral target group New-Left New Middle Class Old-Left Working Class Centrist Middle Class Left-Nationalist Working Class Key competitor party Key policy change Green/Social Liberal Radical Left Mainstream Right Radical Right More progressive stances on the cultural dimension More leftist stances on the economic dimension More moderate stances on both dimensions More authoritarian stances on the cultural dimension Source: Own representation based on Abou-Chadi et al. 2021

LEFT BEHIND BY THE WORKING CLASS? 23 workers with a migrant background). New left strategies are characterized by a combination of“investment-oriented” social policies(for example, childcare expansion, active labour market policies) and progressive positions on issues such as gender equality, the environment and immigration. They aim primarily to win back/over voters from Green and social liberal parties(Abou-Chadi et al. 2021). Proponents of the third approach emphasize the electoral success of the“Third-Way” politics of the mid-1990s and early 2000s and call for a“return to the centre”. This centrist approach proposes moderate positions on both economic and cultural dimensions. By pursuing a centrist strategy, social democratic parties target the coveted“median voter” and aim to attract moderates. They are thus competing mainly with the mainstream right for voters of the middle class, who are likely to switch between these two parties. Finally, supporters of an old left strategy argue that this shift to the political centre is the main reason for the electoral decline of social democratic parties in recent decades(Evans and Tilley 2017). While this strategy may have led to electoral success in the short term, it has alienated large parts of (potential) social democratic voters – particularly belonging to the working class – who favour a more leftist economic programme. An old left strategy thus proposes to take up positions that strongly favour redistribution and consumptionoriented policies, such as the expansion of pension and unemployment benefits, but does not propagate authoritariannativist stances regarding second dimension policies. Rather, the old left strategy argues that the importance of those policies should not be emphasized. These ideal-typical strategies all come with different tradeoffs and make different assumptions about which voters the electoral appeals of social democratic parties should be directed towards. It is ultimately an empirical question how these strategies play out for the competition between social democratic and radical right parties and how they specifically affect the electoral fortunes of social democratic parties. We have already seen that the assumptions underlying some of these approaches do not hold up to empirical scrutiny. We have demonstrated that while the working class is surely electorally relevant for the radical right, social democratic parties have not primarily lost voters to the radical right. The bulk of social democratic losses over recent decades have gone to other parties of the left, such as the Greens, and parties of the mainstream right. In comparison, voter losses to radical right parties have been fairly modest. Hence, the left-nationalist narrative overestimates the electoral relevance of the working class vote for the electoral decline of social democracy and also ignores the socio-structural diversity of the working class today. Focusing on the working class as the core electorate, let alone reducing the working class to white men who hold authoritarian-nativist positions, bears the risk of tying the electoral success of social democratic parties to an ever-decreasing minority of the electorate. After all, the working class has significantly decreased in size in recent decades and the subset of(white male) workers with authoritarian-nativist views have become a shrinking minority within that decreasing minority. With progressive automation, digitalization and educational upgrading, there is little reason to assume that this downtrend will stop anytime soon. In addition, the left-nationalist strategy flies in the face of the positions of the well-educated middle class, the fastest-growing class group within West European societies. Our analyses show that social democratic parties have not disproportionately lost working class or anti-immigrant voters, as the left-nationalist strategy asserts. On the contrary, if anything, educated voters have been overrepresented in social democratic losses. Besides that, social democratic parties are struggling with an ongoing generational replacement of voters. While their electorates are getting older, they have not been able to consistently gain new generations of voters. Social democratic vote losses may therefore, to some extent, have been caused by and contributed to a progressive“greying of Europe’s social democracy”. Less progressive positions, particularly on the cultural dimension, are notably unpopular among the youngest generations(Abou-Chadi and Wagner 2020a), which are still being socialized in terms of political ideologies and partisan identification. Consequently, social democratic parties risk losing newly emerging politicized generations permanently and persistently. This is not a new phenomenon, however. In past decades, newly emerging generations of potential social democratic voters have decided disproportionally to vote for leftist competitors instead(notably Green and radical left parties). This development could be further aggravated if social democratic parties started to promote more authoritarian stances or downplay issues related to the liberal-authoritarian political dimension. We can, of course, assess more directly and systematically how social democratic strategies affect competition with the radical right, as well as their own electoral support. Research has shown that more anti-immigrant positions among mainstream parties do not help to reduce the support of the radical right(Krause, Cohen, and Abou-Chadi 2021). If anything, more anti-immigrant positions lead to higher vote-switching to the radical right. More specifically, research finds that when mainstream parties become more anti-immigration, the level of voter exchanges between mainstream and radical right parties increases. However, overall, the radical right is the net winner of these increased exchanges. Using original survey data from Germany and Austria, 7 we can also demonstrate that left-nationalist strategies do not help social democratic parties to reduce radical right support (Abou-Chadi et al. 2021). Figures 12 and 13 show the probability of choosing a social democratic programme over a profile that contains policy positions of the radical right conditional on people’s left–right self-placement. We can see how this probability varies between a decidedly left-nationalist social democratic profile and a control profile in which differ7 For the comparison between two social democratic programs, respondents were asked to rate two programs on a 1 to 7 scale. The programs contained information regarding the parties’ stances for nine different policy areas. The respondents were also directly asked to choose which program of the two that were shown they preferred. The surveys then also contain a survey experiment(“conjoint experiment”), where voters were asked to compare random hypothetical programmatic profiles of social democratic parties and then to indicate which they preferred. This experiment allows us to evaluate which policy items have a significant negative or positive impact on support for social democratic parties’ programmes.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 24 Figure 12 Predicted Probabilities of Choosing a Social Democratic Program over a Radical Right Program By Left-Right Self-Placement(Germany) 1 Predicted probability SD choice .8 .6 .4 .2 0 0 2 Left-right self placement 4 6 8 10 Random Left Nationalist (Reading example: German voters who place themselves at 0 on the left–right scale have a probability of 80 per cent of choosing a random or left-nationalist social democratic programme over a radical right programme) Source: Predicted probabilities of choosing a social democratic programme over a radical right programme by left–right self-placement(Unique Survey Data, Germany 2020) ent social democratic positions are assigned randomly. While the probability of choosing a social democratic over a radical right profile declines the more someone self-identifies as more right-wing, there is no significant difference between a left-nationalist and a random social democratic profile. Put differently, more left-nationalist positions do not help to keep people from switching to the radical right. In addition to asking how social democratic party strategies affect the radical right, we can of course also investigate how they affect the electoral fortunes of social democratic parties themselves. A centrist strategy aims primarily to reach voters in the centre, who are not stable social democratic voters but have a certain probability of voting for a social democratic party. But this strategy risks losing larger parts of the already mobilized, effective core electorate to other parties in the left spectrum – especially to Green or radical left parties. Research has shown that such a strategy electorally harms social democratic parties, especially in the long term (Karreth and Polk(2013). Pursuing a centrist strategy may help to win elections in the short term, as happened in the 1990s, but could do more harm in the medium and long term. This is because such a strategy, with its rather moderate stances, succeeds neither in winning over many centrist voters nor in winning new generations and thus younger voters. In fact, such a strategy bears the risk of losing left-wing voters to leftist competitor parties that offer a more distinct political programme on those issues, be they Green, and even social liberal, parties on the cultural dimension or radical left on the economic one. Thus, while the“leftist-nationalist” and“centrist” strategies seem intuitively appealing, their potential electoral risks are very likely to outweigh the potential electoral benefits. Those risks may be even higher from a long-term perspective, particularly when looking at new cohorts of voters, who are crucial for social democratic parties’ electoral future. Indeed, the data show that left-nationalist programmes and centrist programmes are popular across the electorate in West European countries, but significantly less popular than new left or old left programmes among potentially social democratic voters(Abou-Chadi et al. 2021). In fact, potential social democratic voters prefer programmes that offer progressive stances on immigration, gender and environmental issues. Furthermore, they also substantially prefer programmes that support the expansion of redistribution and social in-

LEFT BEHIND BY THE WORKING CLASS? 25 Figure 13 Predicted Probabilities of Choosing a Social Democratic Program over a Radical Right Program By Left-Right Self-Placement(Austria) 1 Predicted probability SD choice .8 .6 .4 .2 0 0 2 Left-right self placement 4 6 8 10 Random Left Nationalist (Reading example: Austrian voters who place themselves at 0 on the left–right scale have a probability of 90 per cent of choosing a random or left-nationalist social democratic programme over a radical right programme) Source: Predicted probabilities of choosing a social democratic programme over a radical right programme by left–right self-placement(Unique Survey Data, Austria 2020) vestment. Hence, proposing less progressive stances on immigration and less expansive policies on social investment is likely to lead these voters to reject social democratic parties. Research shows that there is no trade-off between old left and new left strategies within the left-wing electorate, as opponents of“identity politics” like to claim, and that social democratic parties are most successful when they combine social investment policies with more progressive positions on the second dimension(Abou-Chadi and Wagner 2019, 2020; (Abou-Chadi et al. 2021). In sum, old left and new left strategies resonate best among potential social democratic voters. There is no evidence of the often-suggested trade-off between these strategies among the potential electorate. In contrast, left-nationalist and centrist strategies are not likely to result in electoral gains that would outweigh potential losses. Even though these strategies may be favoured by some voters, they are not appealing to those who are most likely to consider voting for a social democratic party. Thus, proposing a centrist, let alone left-nationalist strategy risks keeping many potential voters away and is unlikely to win over sufficient numbers of voters to compensate for these potential losses(Abou-Chadi et al. 2021). In the long term, these strategies look even less promising, as less progressive positions among social democratic parties are associated with less support among younger people(Abou-Chadi and Wagner 2020a). Hence, taking into account that social democratic parties need to build electoral coalitions for the future, left-nationalist and centrist strategies are most likely to lead to further electoral decline.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 26 9 SUMMARY The first two decades of the twenty-first century have been rough on social democratic parties in Western Europe. What first became evident with the implosion of PASOK in Greece has turned into an electoral decline of historic proportions across Western Europe. At the same time as social democratic parties have been losing big, radical right parties have entered a“fourth wave”(Mudde 2019) in which they have been achieving unprecedented electoral successes. It is no wonder that these two processes have become connected in the political and public debates. The common narrative has become that social democratic parties have lost their working-class voters to radical right parties because of either their shift to the right on the economic dimension(“Third Way”) or their shift to the left on the cultural dimension(“identity politics”). Although this narrative has become received wisdom in media and politics, including within social democratic circles, little empirical evidence has ever been provided to back it up. In this report, we have put the common narrative to the test and found that neither its assumptions nor its claims are supported by empirical evidence. First, radical right parties are not the alleged“new workers’ parties”, as a(large) majority of workers still support other parties. Second, social democratic parties have not primarily lost workers. Third, social democratic parties have not lost most of their voters to radical right parties. In fact, they have mainly lost middle class voters to mainstream right(in the 2000s) and to Green parties(in the 2010s). This should not really come as a surprise, as the, often implicit, key assumption is empirically incorrect, too. The working class is not monolithic – white, male and with authoritarian-nativist views. Since the mid-twentieth century, the working class has both shrunk and diversified significantly. Women and people with a migrant background form an ever-growing part of the working class, in particular service workers. Moreover, most workers hold progressive views on libertarian issues(such as gay rights) and only a minority hold anti-immigrant opinions. These facts have important implications for the various suggested strategies for social democratic parties to stave off further decline and regain enough ground to lead government coalitions. The left-nationalist strategy, in which social democratic parties pivot right on the cultural dimension(primarily on immigration) to compete with radical right parties, will attract only a small group of the(lost) electorate. At the same time, it will push away a much larger electorate, including the young and the fast-growing group of higher educated professionals. Similarly, a centrist strategy, in which social democratic parties take more centrist positions on both the economic and the cultural dimension, will probably attract few people and keep away the same groups. Both the old left strategy, in which social democratic parties pivot left on the economic dimension, and the new left strategy, in which they pivot left on the cultural dimension, are the most promising. In both cases the gains significantly outweigh the losses. Moreover, the two are not mutually exclusive, as is often assumed and asserted. Economically left-wing voters prefer culturally progressive positions and cultural progressives are in favour of redistributive policies.

LEFT BEHIND BY THE WORKING CLASS? 27 10 TOWARD A SOCIAL DEMOCRACY FOR THE TWENTY-FIRST CENTURY With an ever declining and greying electorate, it is clear that social democratic parties will have to make fundamental changes to remain relevant in the twenty-first century. But these changes should be guided by solid empirical evidence and grounded in genuine social democratic ideology. Many social democrats have bought into the common narrative and have called for either a playing down of“identity politics” or even for pivoting right on cultural issues such as immigration and integration. Leaving aside that this has further strengthened the mainstreaming and normalization of the radical right (Mudde 2019), it has not halted the electoral decline of social democratic parties or the electoral rise of radical right parties. We end this paper by discussing four points that we believe should guide discussions about how to help make social democratic parties important political actors in the twentyfirst century. They deal with the role of the social democratic party, the rejuvenation of its electorate, the continued need for “identity politics”, and the building of a new electoral coalition. If social democratic parties want to stay electorally relevant, they also need to appeal to young, educated, female and non-white voters. These voters are not the“urban elite”. The “average voter” today is not, say, a miner or a factory worker, but rather a customer service representative, graphic designer or project manager. Normatively, social democracy in the twenty-first century needs to go beyond the mere representation of the white working class, too. By today’s standards, the factory worker of at least the mid-20th century had a privileged position, a well-protected and well-rewarded permanent position. In a world of globalization, and within the knowledge economy, many of these classic working-class positions have disappeared, or they are being replaced by precarious positions that pay less and/or lack protection and security. It is one of the great challenges of contemporary social democracy to find a way to integrate this socio-demographically diverse“precariat” and new working class into a broader movement for economic and social justice in the twenty-first century. 1.  SOCIAL DEMOCRATIC PARTIES ARE NOT THE PARTY OF THE WHITE WORKING CLASS Social democracy in the twenty-first century cannot understand itself as providing the party of the white working class alone. Electorally, a strategy based on appealing mainly to white working and lower middle class people will eventually mean reducing social democracy to a party with a niche appeal and an increasingly old(and dying) electorate. We cannot emphasize enough how much social structures have changed in the knowledge economy. The share of people employed in routine tasks has declined dramatically in the past three decades, while the share of people working in non-routine cognitive tasks stands at around 50 per cent in most Western European countries(Häusermann et al. 2021). In Germany, for example, the share of the cohort attending university lies at 55 per cent and has strongly increased in the past 20 years. 8 8 https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/BildungF o r s c h u n g- Ku l t u r/ H o c hs c h u l e n/ P u b l i k a t i o n e n /_ p u b l i k a t i o n e n- i n n e n- h o c h schulen-kennzahlen.html 2.  REJUVENATING THE ELECTORATE The electoral dynamics analysed in this report are important not only with regard to past voters, but also in relation to future voting behaviour. Emerging cohorts of voters are more in favour of progressive positions and are likely persistently to prefer parties that offer a clear and progressive profile in this dimension. Pursuing left-nationalist or centrist strategies may thus harm social democratic parties, not only in the short term, but also – and particularly – in the long-term. With already aging electorates, social democratic parties are destined for electoral insignificance if they do not find a way to rejuvenate their electorates. This generational decline should be a warning sign for social democratic parties. In many European countries, social democratic parties have little appeal among younger generations, who rather support social liberal or Green parties instead. Without attracting new voters and appealing to younger generations other parties will soon become the leading force of the Left – such developments are already visible in

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 28 France, Germany and the Netherlands, for example. Hence, party strategy needs to go beyond the next election and ask how a long-term, durable progressive coalition can be(re-) attached to social democratic parties. This strategy puts social democratic parties primarily in competition with Green and social liberal parties, which should at the same time be their most likely coalition partners. The best way to stand out from these two competitors is to combine the shared progressive cultural policies with more unique, bold and innovative left-wing economic policies. 3.  THE CONTINUED NEED FOR ‘IDENTITY POLITICS’ The future of social democratic parties in Western Europe (and North America) is increasingly discussed under the label of“identity politics”. In its original meaning, identity politics is of course at the heart of the history of social democracy, the labour movement, and the struggle for democratic representation. Originally coined by the Combahee River Collective (1977), the idea of identity politics signifies that oppressed groups are themselves best equipped to signify and represent their own preferences. Even a cursory look at the history of social democratic and socialist party formation should make it obvious that this thought lies at the heart of the democratic class struggle and social democracy itself. In the current debate, the meaning of identity politics has become more diffuse, however. Identity politics has become an umbrella term for political and social demands to address structural discrimination based on race, gender or sexual identity. A core idea is that structures are maintained and reproduced through performative acts, such as speech. The ways we talk and act affect perceptions of social categories and create boundaries of inclusion and exclusion. Political demands often labelled“identity politics” focus on making these mechanisms of structural discrimination visible and counteracting them by changing the way we speak and act, but also by overhauling some social and political institutions altogether. In the debates most relevant for social democracy, opponents argue that identity politics(1) are the agenda of an urban educated elite and(2) divide the redistributive coalition. As this report has shown, these claims are largely based on a misperception of who the working class is and what it wants, as well as on a wrong idea of who the redistributive coalition and the support groups of social democracy are. Large shares of the new middle-class groups that support left-wing parties are very much in favour of redistribution (sometimes even more than the working class)(Abou-Chadi and Hix 2021). Moreover, we have shown that, contrary to the assertions of opponents, many policies associated with identity politics are supported by the overwhelming majority of working class and lower educated voters, while only a minority of these groups truly oppose them(even immigration). Rather than seeing“identity politics” as in competition with“class” politics, they should be considered intrinsically intertwined. As long as the institutional structures of heteronormativity, patriarchy and white supremacy remain intact, the material advancement of marginalized groups will not make them equal citizens. At the same time, the destruction of these institutional structures alone will not change the economic, and therefore political, marginalization of these groups. Moreover, the material advancement of marginalized groups will make them into a more powerful(or less powerless) force against these institutional structures. For the bulk of potential voters of social democratic parties, there is no trade-off between more progressive positions on the cultural and the economic dimension. In fact, we find support for more outspoken progressive positions on both dimensions. In other words, a strategy of redistributive economic policies and new“identity politics” is most likely to attract the largest share of potential social democratic voters, including those from the working class. If social democratic parties do not want to give up a broad class appeal, this combination of positions seems to be the only viable strategy for the long run. 4.  BUILDING A NEW PROGRESSIVE COALITION Social democratic parties should not just build a new electoral coalition, but also a new progressive political coalition. This includes finding new ways of engaging and incorporating membership, as well as strengthening ties with civil society actors. We have already documented how the electorate of social democratic parties has become significantly older. This is even more the case for their members. Members play a key role not only in mobilizing during campaigns but also in generating links between political parties and different groups in society. If social democratic parties want to be more than technocratic policy providers, attracting young and diverse members is essential. Social democratic parties in many countries also need to re-cultivate their relationship with trade unions. Labor unions themselves have of course suffered from many of the same developments outlined in this report and have struggled to represent a growing precariat. But a revival of this relationship should be in the interest of social democratic parties and trade unions themselves.“Work” and“labour” will remain at the core of social democratic ideas and policy positions. The coming decades will fundamentally challenge what the concept of work means, how wage labour is structured and what types of tasks will still be done by humans. Social democratic parties are best placed to actively shape these transformations and prevent them from exacerbating social and economic inequalities. Focusing on the future of work is also a way for social democratic parties to distinguish themselves from Green and social liberal competitors. This project can work only in alliance with labour unions. At the same time, social democratic parties need to rethink and redevelop their relationship to social movements. Failing to incorporate the new social movements of the 1970s and 1980s is one of the main reasons for the electoral dilemma that social democratic parties are facing today. If social democratic parties repeat these mistakes with progressive movements such as Fridays For Future or Black Lives Matter they will lose further generations of potential supporters.

LEFT BEHIND BY THE WORKING CLASS? 29 In general, parties of the Left need to redevelop their relations with each other. The German case is the prime example: the refusal of the SPD and the Left Party to cooperate has resulted in coalitions with the SPD as the junior coalition partner of the mainstream right Union parties. These types of coalitions make it impossible for social democratic parties to make the necessary changes for a strategy for the coming decades. The often affectively driven animosities between different groups on the Left will prevent coalitions that are able to implement profound and durable progressive change. Other countries, such as Portugal, demonstrate the potential of cooperation between different parties on the Left. With some exceptions, the time of big parties in advanced capitalist democracies is over and party systems are more fragmented than ever before. As society has become more diverse and less structured, electoral politics has become two- or even three-dimensional. As a consequence, voters have become more dispersed and volatile, and most party systems now have at best one party able to attract more than one-third of the vote. Even if social democratic parties re-­ establish their position as one of the biggest parties, they will still require the support of one and more likely two(or even more) other parties to be able to create coalitions that will enable them to implement some of their key policies. Hence, they need the cooperation of other progressive parties, from radical left to social liberal, and the support of progressive civil society organizations such as trade unions. It is therefore essential that these parties and progressive organizations stop seeing each other as potential threats to their power but rather as potential allies in their political struggle.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 30 List of tables and figures 5 Figure 1 Average Vote Share for Social Democratic and Radical Right Parties in Europe, 1990 – 2020 12 Figure 2 Migration background and gender in class groups 13 Figure 3 Gays and lesbians should be free to live the way they wish 14 Figure 4 Country's culture underminded or enriched by immigration 14 Figure 5 Immigration makes country a worse or better place 17 Figure 6 Party Choice by Class Group 17 Figure 7 Vote Switching to Radical Right Parties since 2000 18 Figure 8 Vote Switching to Radical Right Parties since 2000 19 Figure 9 Changes in Social-Democratic Party Support since 2000 by Educational Groups 20 Figure 10 Vote Switching from Social Democratic Parties since 2000 21 Figure 11 Vote Switching from Social Democratic Parties by Educational Groups 24 Figure 12 Predicted Probabilities of Choosing a Social Democratic ­Program over a Radical Right Program By Left-Right Self-Placement(Germany) 25 Figure 13 Predicted Probabilities of Choosing a Social Democratic ­Program over a Radical Right Program By Left-Right Self-Placement(Austria) 22 Table 1 Idealtypical Issue Profiles for Social Democratic Parties

LEFT BEHIND BY THE WORKING CLASS? 31 References Abou-Chadi, Tarik, and Ryan Finnigan. 2019.“Rights For Same-Sex Couples And Public Attitudes Toward Gays And Lesbians In Europe.” Comparative Political Studies 52(6): 868–895. Abou-Chadi, Tarik, Silja Häusermann, Reto Mitteregger, Nadja Mosimann, and Markus Wagner. 2021.“Old Left, New Left, Centrist Or Left Nationalist? Determinants Of Support For Different Social Democratic Programmatic Strategies.” Unpublished Manuscript. Abou-Chadi, Tarik, and Simon Hix. 2021.“Brahmin Left Versus Merchant Right? Education, Class, Multiparty Competition, And Redistribution In Western Europe.”. The British Journal of Sociology 72(1): 79–92. Abou-Chadi, Tarik, and Markus Wagner. 2019.“The Electoral Appeal Of Party Strategies In Post-Industrial Societies: When Can The Mainstream Left Succeed?” Journal of Politics 81(4): 1405–19. Abou-Chadi, Tarik, and Markus Wagner. 2020a.“Electoral Fortunes Of Social Democratic Parties: Do Second D­ imension Positions Matter?” Journal of European Public Policy 27(2): 246–72. Abou-Chadi, Tarik, and Markus Wagner. 2020b.“Losing The Middle Ground: The Electoral Decline Of Social Democratic Parties Since 2000.” Paper presented at the Annual Meeting of the American Political Science Association. Agustín, Óscar García, and Martin Bak Jørgenson. 2019.“Danes First, Welfare Last”. Jacobin, 31 January. https://www.jacobinmag.com/2019/01/ denmark-social-democrats-immigration-welfare(Accessed 21 June 2021). Arzheimer, Kai. 2018.”Explaining Electoral Support For The Radical Right”. In The Oxford Handbook of the Radical Right, eds. Jens Rydgren. Oxford: Oxford University Press, 143–165. Benedetto, Giacomo, Simon Hix, and Nicola Mastrorocco. 2020.“The Rise And Fall Of Social Democracy, 1918 – 2017.” American Political Science Review 114(3): 928–39. Beramendi, Pablo, Silja Häusermann, Herbert Kitschelt, and Hans-Peter ­Kriesi. 2015a.“Introduction: The Politics Of Advanced Capitalism.” In The Politics Of Advanced Capitalism, eds. Pablo Beramendi, Silja Häusermann, Herbert Kitschelt and Hans-Peter Kriesi. New York: Cambridge University Press, 1–64. Beramendi, Pablo, Silja Häusermann, Herbert Kitschelt, and Hans-Peter ­Kriesi eds. 2015b. The Politics Of Advanced Capitalism. New York: Cambridge University Press. Berman, Sheri. 2016.“The Lost Left.” Journal of Democracy 27(4): 69–76. Berman, Sheri, and Maria Snegovaya. 2019.“Populism And The Decline Of Social Democracy.” Journal of Democracy 30(3): 5–19). Betz, Hans-Georg. 1994. Radical Right-Wing Populism In Western Europe. Houndmills, Basingstoke: MacMillan. Bischof, Daniel, and Thomas Kurer. 2021.“Where Do Radical Right Voters Come From?” Unpublished Manuscript. Bornschier, Simon. 2010.“The New Cultural Divide And The Two-Dimensional Political Space In Western Europe.” West European Politics 33(3): 419 – 44. Combahee River Collective. 1977.“A Black Feminist Statement.” Eatwell, Roger, and Matthew J. Goodwin. 2018. National Populism: The Revolt Against Liberal Democracy. London: Pelican. Evans, Geoffrey, and James Tilley. 2017. The New Politics Of Class: The ­Political Exclusion Of The British Working Class. Oxford: Oxford University Press. Gingrich, Jane, and Silja Häusermann. 2015.“The Decline Of The Working-Class Vote, The Reconfiguration Of The Welfare Support Coalition And Consequences For The Welfare State.” Journal of European Social Policy 25(1): 50–75. Goodhart, David. 2017. The Road To Somewhere: The Populist Response And The Future Of Politics. London: Hurst. Harteveld, Eelco. 2016.“Winning The‘Losers’ But Losing The‘Winners’? The Electoral Consequences Of The Radical Right Moving To The Economic Left.” Electoral Studies 44: 225–34. Häusermann, Silja, Tarik Abou-Chadi, Reto Bürgisser, Matthias Enggist, Reto Mitteregger, and Delia Zollinger. 2021.“Wählerschaft, Wahlmotive und strategische Potenziale der Sozialdemokratie in Europa und der Schweiz.: Bericht zuhanden der Sozialdemokratischen Partei der Schweiz.” Häusermann, Silja, and Hans-Peter Kriesi. 2015.“What Do Voters Want? ­Dimensions And Configurations In Individual-Level Preferences And Party Choice.” In The politics Of Advanced Capitalism, eds. Pablo Beramendi, Silja Häusermann, Herbert Kitschelt and Hans-Peter Kriesi. New York: Cambridge University Press, 202–30. Hooghe, Liesbet, Gary Marks, and Carole J. Wilson. 2002.“Does Left/Right Structure Party Positions On European Integration?” Comparative Political Studies 35(8): 965–89. Ignazi, Piero. 1992.“The Silent Counter-Revolution: Hypotheses On The Emergence Of Extreme Right-Wing Parties In Europe.” European Journal of Political Research 22: 3–34. Inglehart, Ronald. 1977. Silent Revolution: Changing Values And Political Styles Among Western Publics. Princeton: Princeton University Pres. Ivarsflaten, Elisabeth. 2007.“What Unites Right-Wing Populists In Western Europe?: Re-Examining Grievance Mobilization Models In Seven Successful Cases.” Comparative Political Studies 41(1): 3–23. Kaufmann, Eric. 2018. Whiteshift: Populism, Immigration, And The Future Of White Majorities. London: Penguin. Karreth, Johannes, Jonathan T. Polk, and Christopher S. Allen. 2013.“Catchall Or Catch And Release? The Electoral Consequences Of Social Democratic Parties’ March To The Middle In Western Europe.” Comparative Political Studies 46(7): 791–822. Kitschelt, Herbert. 1994. The Transformation Of European Social Democracy. Cambridge Studies In Comparative Politics. Cambridge and New York: Cambridge University Press. Kitschelt, Herbert, and Philipp Rehm. 2014.“Occupations As A Site Of Political Preference Formation.” Comparative Political Studies 47(12): 1670–1706. Krause, Werner, Denis Cohen, and Tarik Abou-Chadi. 2021.“Does Accommodation Work? Mainstream Party Strategies And The Success Of Radical Right Parties.” Unpublished Manuscript. Kruzel, John. 2018.“Did Hillary Clinton Once Support A Border Wall With Mexico?”.Politifact, 27 June. https://www.politifact.com/factchecks/2018/ j u n /27/c a l- t h o m as /d i d- h i l l a r y- c l i nto n- su p p o r t- b o rd e r- wa l l- m e x i co. De Lange, Sarah L. de. 2007.“A New Winning Formula? The Programmatic ­Appeal Of The Radical Right Party Politics.” 13(4): 411–35). Lilla, Mark. 2017.“Die Linke hat sich selbst zerstört.” Neue Zürcher Zeitung, 2017. https://www.nzz.ch/feuilleton/identitaetspolitik-die-linke-hat-sichs e l bs t-ze r s to e r t- l d.1311079?re d uc e d= t r u e. Lilla, Mark. 2018. The Once And Future Liberal: After Identity Politics. Oxford University Press.

FRIEDRICH-EBERT-STIFTUNG – Division for Analysis, Planning and Consulting 32 Lipset, Seymour M. 1959.“Democracy And Working-Class Authoritarianism.” American Sociological Review 24(4): 482–501. Lipset, Seymour M. 1960. Political Man: The Social Bases Of Politics. Garden City, NY: Doubleday. Mudde, Cas. 2007. Populist Radical Right Parties In Europe. Cambridge: Cambridge University Press. Mudde, Cas. 2019. The Far Right Today. Cambridge: Polity. Mudde, Cas. 2010.“The Populist Radical Right: A Pathological Normalcy”. West European Politics 33(6): 1167–1186. Oesch, Daniel. 2006.”Coming To Grips With A Changing Class Structure: An Analysis Of Employment Stratification In Britain, Germany, Sweden And Switzerland.” International Sociology 21(2): 263–288. Oesch, Daniel. 2008.”Explaining Workers' Support For Right-Wing Populist Parties In Western Europe: Evidence From Austria, Belgium, France, Norway, And Switzerland.” International Political Science Review 29(3): 349–373. Oesch, Daniel. 2012. The Class Basis Of The Cleavage Between The New Left And The Radical Right: An Analysis For Austria, Denmark, Norway And Switzerland. London: Routledge. Oesch, Daniel, and Line Rennwald. 2018.“Electoral Competition In Europe’s New Tripolar Political Space: Class Voting For The Left, Centre-Right And Radical Right.” European Journal of Political Research 57(4): 783–807. O'Grady, Tom. 2020.“Age Cleavages Across European Countries And Issue Dimensions Since The 1980s.” SocArXiv.10.31235/osf.io/r8xbg. Rooduijn, Matthijs. 2018.“What Unites The Voter Bases Of Populist Parties? Comparing The Electorates Of 15 Populist Parties.” European Political Science Review 10(3): 351–68. Watt, Nicholas, and Patrick Wintour. 2015.“How Immigration Came To Haunt Labour.” The Guardian, 24 March. https://www.theguardian.com/ n e w s/ 2015/m a r/ 24/ h ow- i m m i g r a t i o n- c a m e- to- h a u nt- l a b o u r- i nsi d e- s to r y.

Imprint: © 2021 Friedrich-Ebert-Stiftung Publisher: Division for Analysis, Planning and Consulting Hiroshimastraße 17, 10785 Berlin, Germany www.fes.de Orders/contact: analyse-planung@fes.de The opinions expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung(FES). Commercial use of FES publications in any media without written permission from the FES is prohibited. Publications of the Friedrich-Ebert-Stiftung may not be used in connection with election campaigns. 978-3-96250-933-0 Copy-editing: James Patterson Title image:© picture alliance/blickwinkel/K. Thomas Design: www.stetzer.net Printing: www.bub-bonn.de

DIVISION FOR ANALYSIS, PLANNING AND CONSULTING ADDITIONAL PUBLICATIONS Transformation of the left. The resonance of progressive programs among the potential social democratic electorate/ Silja Häusermann, Herbert Kitschelt, Tarik Abou-Chadi, Reto Mitteregger, Nadja Mosimann, Philipp Rehm and Markus Wagner. Berlin: Friedrich-Ebert-Stiftung, International Policy Analysis, January 2021. ISBN 978-3-96250-802-9 http://library.fes.de/pdf-files/id/ipa/17384.pdf Transformation of the left. Strategic options for social democratic parties/ Herbert Kitschelt and Silja Häusermann. Berlin: Friedrich-Ebert-Stiftung, International Policy Analysis, January 2021. ISBN 978-3-96250-803-6 http://library.fes.de/pdf-files/id/ipa/17383.pdf Transformation of the left. The myth of voter losses to the radical/ Silja Häusermann, Herbert Kitschelt, Tarik Abou-Chadi, Macarena Ares, Daniel Bischof, Thomas Kurer, Mathilde van Ditmars and Markus Wagner. Berlin: Friedrich-EbertStiftung, International Policy Analysis, January 2021. ISBN 978-3-96250-801-2 http://library.fes.de/pdf-files/id/ipa/17385.pdf Download publications on Empirical Social Research in German from https://www.fes.de/lnk/sozialforschung Download and research all our publications from https://www.fes.de/bibliothek/fes-publikationen