The ghost of the Iron Curtain : the impact of life satisfaction on populist voting

(1)

of Life Satisfaction on Populist Voting

by

Vera Eline Verhoef

Master thesis Political Science

Political Economy

Graduate School of Social Sciences

University of Amsterdam

Student number: 12271187

Supervisor: Dr. Mike Medeiros

(2)

Abstract

Verhoef, V. 2019. The Ghost of the Iron Curtain: The Impact of Life Satisfaction on Populist Voting. Master of Science (Political Economy), University of Amsterdam, Amsterdam.

While the rise of populism in the western contemporary political climate has been a hot topic of discussion for some time now, researchers have been unable to pinpoint the factors that drive an individual to support a populist political party. Identifying and understanding these drivers could have a great impact on countries’ policymaking. It would allow the established elites to adjust their governing structure in order to maintain their supporter base and possibly even appeal to populist voters. An especially understudied motivational aspect of populist voting behaviour is life satisfaction. This thesis uses Germany as a case study to assess whether individuals who report lower levels of life satisfaction will be more likely to express support for populist political parties. Data from the German Institute for Economic Research is used to run a binary logistic regression analysis for the impact of life satisfaction and several (demographic and attitudinal) covariables on populist party preference in Germany for the year 2017. Results indicate that a significant negative relationship can be found between life satisfaction and populist support, but that East-West German residence and concerns about the country’s economy have an even bigger impact on populist voting tendency amongst German citizens. This would suggest that it is in a government’s best interest to ensure that their citizens are overall satisfied with their own lives, but also do not feel like a disadvantaged group in society, and trust established national (economic) institutions.

(3)

1. Introduction

Populism, a political ideology that considers society to be separated into two groups, the pure people and the corrupt elite, and which argues that politics should be an expression of the will of the people, has experienced a surge in popularity in recent years. Democracies all across the world are seeing populist parties on the rise and witness them gain an increasing number of seats within their national governments. Just last month, the three populist radical right groups in the European Parliament increased their seat share from 21% to 23% as a result of the 2019 European elections (Charlemagne 2019). This is a relatively small expansion considering the current European political climate, but an expansion nonetheless. Why is it that populism has become so popular and widespread in developed countries? There are countless different theories that take a shot at explaining this phenomenon. Many focus around changes in the international environment, such as globalization, economic crises or the recent migrant crisis. Others are focussed around people’s personal dissatisfaction with established institutions or individuals’ socio-economic status in society. Identifying and understanding the drivers of populist support could have a great impact on future policymaking of countries’ ruling parties. It would allow the established elites to adjust their governing structure in order to maintain their supporter base and possibly even appeal to populist voters.

This thesis will provide new insight into the study of populism and what motivates people to support populist ideology. It will do so by looking at the influence of subjective well-being, or more specifically life satisfaction, on populist voting behaviour. Research into this domain is limited. There are a plethora of studies which assess the motivational drivers that get people to cast a vote during an election (Bahry and Silver 1990; Flavin & Pacek 2014; Ng, Vaithilingam & Rangel 2017; Owens, Videras & Willemsen 2008; Veenhoven 1988; Weitz-Shapiro & Winters 2011; Zhong & Chen 2002) and there is even some research that has evaluated how people with different levels of life satisfaction vote in line with a particular political party ideology, such as conservative or liberal (Flavin & Pacek 2014; Liberini, Redoano & Proto 2013). However, surprisingly few researchers so far have looked at the relationship between subjective well-being and populist party preference in particular (Elchardus & Spruyt 2016). This relationship, which is further explored in this thesis, is also based on an observation of recent German figures. When looking at the geographical distribution of votes for the 2017 German federal election, what stands out is the fact that a much larger percentage of people living in states that in the past were part of the communist German Democratic Republic (GDR) voted for a populist party, than their fellow countrymen who live in the states previously falling under the western Federal German Republic (FGR). A map containing life satisfaction data from recent years for every German federal state, showed a similar contrasting distribution. Those residing in the West have a noticeably higher

(5)

average level of life satisfaction than those living in the East. A comparison of these two figures resulted in the following research question: Can the contemporary populist political divide between Germany’s

former East and West be explained by the East-West gap in overall life satisfaction?

Based on this research question, this thesis will, besides providing new insight into the study of populism, also contribute to the explanation for the political divide between Germany’s former East and West, exactly three decades after the unification process commenced. As has been depicted, some literature exists on the topic of the impact of subjective well-being on political party preference, but it is scarce. This study will therefore make a useful contribution to the field. In order to test the theory, a quantitative analysis is employed. It used a dataset provided by the German Institute for Economic Research, called the Socio-Economic Panel study, containing information on several thousand German citizens for the year 2017.

The thesis proceeds as follows: It first examines the concept of populism and aims to explain the contemporary popularity of this phenomenon, based on the existing literature and recent world events. The following chapter defines life satisfaction as an important measure for subjective well-being. In both these chapters Germany is used as a case study. Its former division between East and West make for an incredibly interesting case and is what instigated this research in the first place. Chapter 4 combines the findings from the two previous chapters and introduces the main theory of the thesis, that is, the impact of life satisfaction on political preference and populist parties in particular. Based on this, the following hypothesis is presented: Individuals who report lower levels of

life satisfaction will be more likely to express support for populist political parties. Subsequently, the

research design and data that will be used to test this hypothesis are outlined. Only data from the year 2017 will be used, since this was the year of the most recent German Federal Election, whose outcomes were an instigator for the main research question. The study first makes use of an independent t-test in order to confirm the assumption that life satisfaction levels differ between East and West Germany for the year 2017. The second analysis is a binary logistic regression, testing the influence of life satisfaction and ten other established predictors on populist voting. The findings indicate that life satisfaction does indeed have an impact on vote choice, where individuals who are less satisfied with their life are more likely to be attracted to populist parties. However, controlling for individuals’ concerns about immigration and the economy unquestionably reduces life satisfaction’s significance as an indicator for populist vote choice. Interestingly, a fear for the general state of the country’s economy appears to be a highly influential predictor for populist support. Finally, it is concluded that the findings support the argument that the contemporary populist political divide between Germany’s former East and West can be, at least in part, explained by the East-West gap in overall life satisfaction. Future research is encourages to try and replicate this experiment for numerous different countries in

(6)

order to test the robustness of the link between life satisfaction and populist vote choice. In doing so, it is recommended to include additional variables that measure societal declinism and trust in established political institutions, such as the government and the democratic system in general. This could help build a better predictive model for deciphering what factors ultimately drive people to decide to vote for populist parties. Knowledge and understanding of these factors could help established political parties to keep the spread of populism in their country in check, by focussing on the most detrimental causes for people to cast an anti-establishment populist vote, and allow governing politicians to change their policy accordingly.

(7)

2. The Rise of Populism

Nowadays, one encounters the word populism on a nearly daily basis. However, this term can be interpreted in a multitude of ways. It is therefore important to clearly define it at the beginning of this thesis. Well known political science researcher Cas Mudde has clearly identified the concept in the following manner: “I define populism as an ideology that considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ versus ‘the corrupt elite’, and which argues that politics should be an expression of the volonté générale (general will) of the people” (Mudde 2004, 543).

According to researchers Vehrkamp and Wratil, populism can have three different dimensions: anti-establishment, anti-pluralism, and pro-popular sovereignty. Accordingly, a core characteristic of populism is thus the critiquing of institutions and elites that form the societal establishment (Vehrkamp & Wratil 2017, 9). The straightforward core elements, the “us” versus “them” mentality, and the conviction that the people’s ideas are better than those of their rulers, has earned populism the reputation of being a “thin ideology”, since there does not appear to be much more to it (Elchardus & Spruyt 2016, 113; Mudde 2004, 544; Muis & Immerzeel 2017, 910). Primarily, populist parties direct their focus toward the established ruling parties and politicians, calling for a greater influence of the people. These parties can be on either side of the political left-right spectrum, as long as they believe in “societal disputes as conflicts between the “one” people and the “corrupt” political elite” (Vehrkamp & Wratil 2017, 9).

Support for populist parties is commonly explained not only by people’s political discontent, but also their economic and cultural grievances, such as opposition to immigration. However, Mudde (2004, 544) stresses that “populism is moralistic rather than programmatic.” There are some researchers who believe that feelings of societal pessimism or societal decline play an important part in determining populist voting behaviour as well. They regard support for populism as a result of a very negative view of the evolution of society (Elchardus & Spruyt 2016, 111; Harteveld & Steenvoorden 2018, 28). Similarly, Sartori stressed that politics is not just about what people think, but also about what they believe and feel. According to him, populism in particular strongly relies on an emotional, or as some would say, “gut” appeal (Sartori 1976, 132-3). Moderate populists strive for more power to the people in political decision making, whereas radical populists will often call for radical reforms to the political sphere, where they believe they should be the ones representing and protecting the will of the people. The former is a healthy aspect of modern democracy, but the latter can grow into a real threat to the democratic system itself. Therefore, it is important to acknowledge populist sentiment when it rears its head and to attentively consider the reasons for its emergence.

(8)

Some academics argue that the concept has existed since the time of Aristotle, and that populism is by no means a new occurrence in the political sphere (Kuzminski 2008, 3). However, most scholars of populism believe that it is truly a modern phenomenon that emerged in the late nineteenth century in Russia and the United States. By now, populism has spread into almost all continents and political regimes, but is most commonly witnessed in the European and American democracies (Mudde & Kaltwasser 2017, 21). Many countries in South America have experienced several waves of populism during the past 90 years, which were fuelled by severe economic downturns and anti-imperialist sentiment (Mudde & Kaltwasser 2017, 28). One of the most recent South American examples of rising populism, is the election of President Jair Bolsonaro (Campoy 2018). In Southeast Asia, the 1997 Asian financial crisis caused the emergence of several populist politicians, who campaigned against the neoliberal globalization trends. These politicians were at first rather successful in countries such as the Philippines, South Korea, Taiwan and Thailand, but none of their populist regimes were long-lived (Mudde & Kaltwasser 2017, 38). Populism is rather uncommon in Africa, since many countries are still led by an authoritarian ruler or are at best host to a flawed electoral democracy. On this continent, populism has been embodied by the campaigns of future heads of state. Examples of this are the campaigns by Ugandan president Yoweri Museveni in the 1980s and Zambian president Michael Sata in 2011 (Mudde & Kaltwasser 2017, 38). On the other hand, Middle Eastern countries with more established democracies, such as Turkey and Israel, have seen populism become a more integral part of their politics in recent years with leaders Benjamin Netanyahu and Recep Tayyip Erdogan currently in power (Mudde & Kaltwasser 2017, 39). A recent increase in popularity for populism could be witnessed in the United States as well. The 2016 US presidential election encountered expressions of populist sentiment on the side of both the Democrats (Bernie Sanders) and the Republicans (Donald Trump). Both these men were running on anti-establishment platforms and criticised free trade deals. Trump’s eventual victory coincided with a similar trend of rising populism in Europe.

Over the past decades, populist movements have gained popularity all across Europe (Muis & Immerzeel 2017, 9). Last year alone, populist parties secured 63% of the vote in Hungary, the Five Star Movement formed a populist coalition with Lega Nord in Italy, and in the Czech Republic populist parties won over 40% of the votes, which was ten times as much as 20 years ago. Furthermore, The Netherlands has been home to a strong populist party, the Party for Freedom, for over ten years, and the success of the relatively new Forum for Democracy party in this year’s provincial elections has once again resulted in a surge in populism. Not to mention that populism in France has had a strong foothold for decades now, with the Front National as one of the most successful of all populist parties in Western Europe (Henley 2018). European populism is mostly of a xenophobic nature and “derives from a very specific conception of the nation, which relies on an ethnic and chauvinistic definition of the

(9)

people” (Mudde & Kaltwasser 2017, 34). The focus of these populist parties is often purely cultural and aimed at immigration and (European) integration policy. Accordingly, economic aspects often go undiscussed in these parties’ manifests and election statements.

The case of Germany, another European country where populism has been making great strides in the political scene in recent years, will be focused on in greater detail throughout the rest of this thesis, since it makes for an interesting contemporary case study. The Berlin Social Science Research Centre’s populism barometer found that over 30% of eligible German voters held populist attitudes on either the left or right side of the political spectrum in 2018. This was an increase of 4% compared to the previous year and thus an indication that such attitudes have increased throughout the German population (Vehrkamp & Wolfgang Merkel 2018, 3). According to the report, this rise in vote share for populist parties has benefitted the right-populist Alternative for Germany (Alternative für Deutschland, AfD) the most, followed by the left-populist The Left (Die Linke). The official website PopuList, which offers an overview of populist parties in Europe since 1988 and is supported by the Amsterdam Institute for Social Science Research, The Guardian, and the ECPR Standing Group on Extremism and Democracy, recognises the AfD and Die Linke as Germany’s only two current populist parties (Rooduijn et al. 2019, 1). Moreover, scholars broadly agree that the AfD is a radical right populist party. However, Die Linke’s populist ideology has been a topic of dispute for some. Based on the PopuList, Die Linke’s own manifesto (Die Linke 2017, 7), and studies showing that the more populist individuals are, the more likely they are to feel affinity for Die Linke, identify with the party, and vote for it (Vehrkamp & Wratil 2017, 54-60), this thesis will recognise it as a populist party.1_{The following}

two sections will further describe the formation and ideology of these two parties.

2.1. The Alternative for Germany

The AfD was founded in 2013 by a group of elite conservatives that largely consisted of former CDU supporters (Eddy 2017). Coming from the largest centre-right party, spearheaded by Chancellor Merkel, this group claimed to be unsatisfied with the centrist direction the party was moving in. They were especially frustrated with Merkel’s “Euro rescue package,” the decision to help Greece out of its economic crisis (AfD 2017, 5). In its first year of existence, the AfD received an impressive 4.7% of the vote in the federal elections, falling just short of the 5% electoral threshold needed to take seat in the federal parliament, or Bundestag (Lees 2018, 297). Starting off as a liberal conservative party with a Eurosceptic view, more specifically with the Euro crisis as its initial mobilization catalyst, the migrant crisis of 2015 caused the party to radicalize, moving even more toward the right side of the political

1_{Small German parties which portray populist sentiment, such as the Pirates Party (Piratenpartei Deutschland)}

and the National Democratic Party (Nationaldemokratische Partei Deutschlands, NPD), have been excluded from this study, since they have never been able to cross the 5% threshold in votes needed to secure a seat in the Bundestag.

(10)

spectrum (Lees 2018, 308). The anti-immigrant attitudes became much more prominent in mobilizing support than the previously vital anti-EU sentiments. At this point in time, the party also started to significantly increase its populist rhetoric, expressing predominantly xenophobic and nativist ideologies, as the European debt was no longer a dominant issue for German society (Hansen & Olsen 2019, 3). 2015 was a turning point for the AfD: “[Its] Eurosceptic narrative became increasingly nested within a more populist and critical approach to the entire German political settlement and the institutions, procedures, and practices that underpin it. At the same time, the ordo-liberal and pro– free market elements of the party’s programme began to be over-written by a more orthodox right-wing populist agenda” (Lees 2018, 305). This ideological shift could for a large part be attributed to the change in AfD leadership, from the original economically liberal spokesperson Berndt Lucke to Frauke Petry. Petry steered the party in its populist direction by strongly opposing Merkel’s open door policy to Syrian refugees. Four months before the 2017 federal elections, Petry stepped down as AfD’s candidate for Chancellor after being accused of harming the party’s popularity by causing internal tumult. She still ran for an AfD seat in the Bundestag in the September elections, but left the party only days after the election, and was replaced by Alice Weidel and Alexander Gauland. Whereas Weidel could be seen as a classic liberal, Gauland was more of a radical (Lees 2018, 305). Because of its shifts in ideology over time, the AfD has a rather diverse membership. Some of its original supporters, national conservatives and discontent Christian Democrats, have remained loyal to the party, whilst its radicalisation has drawn right-wing extremists and populist voters to their ranks as well (Siri 2018, 141). The AfD’s 2017 manifesto opened with the slogan: “Courage to stand up for Germany – We are not subjects, but free citizens” (AfD 2017, 5). Furthermore, it states that “citizens should have the right to true political alternatives, not only those presented by the political class” and that “the time to stand up and act has arrived.”

2.2. The Left

Die Linke came into existence in 2007 with the merging of two German left-wing parties: the PDS (Party of Democratic Socialism) and the WASG (Labour and Social Justice – The Electoral Alternative) (Coffé & Plassa 2010, 722). The PDS was the successor of the Communist Party and was mostly concerned with topics such as employment and job security. However, this party was most well-known by its reputation as a defender of East German regional identity and being “the central political force of lingering cultural divide in united Germany” (Doerschler 2015, 377). In order to rid themselves of this image and reach a broader voter base, the PDS was keen to merge with the WASG in 2007. The WASG was erected not long before that in 2005 and held support from a wide range of people, including unionists, dissatisfied SPD members, communists, and radical green leftists, who found each other through their opposition to the economic and labour market reforms of Chancellor Schröder

(11)

(Coffé & Plassa 2010, 722). The final product of the merge of these two parties was called Die Linke, or the Left Party. This ambiguous name has helped it retain its base of left-wing voters and Die Linke is currently a catch-all party for the left, consisting of leftist extremists, union members, working class members, and some discontent social democrats. What brings all these different groups of people together under one party is Die Linke’s “critique of neoliberalism and a strong denunciation of economic elites and big business” (Doerschler 2015, 382). Die Linke has made the shift from its former PDS-East German supporter base, to becoming an advocate for the ordinary people against the established elites, thus embracing a populist stance. This was helped along by the change in leadership after the 2012 presidential election, when Gesine Lötzsch stepped down as chairwoman. Since then, Katja Kipping and Bernd Riexinger have been the party’s co-chairs (Mannewitz 2017, 227). Their populist ideology is evident from their manifesto, in which they claim to stand up against ruling politicians and fight for the representation of all people in the democratic process (Die Linke 2017, 7). In the 2013 federal election, Die Linke managed to receive 8.6% of the German votes. This landed them 64 seats in the Bundestag and made them the third biggest party in the coalition, one place up from 2009 (The Federal Returning Officer 2019).

2.3. The 2017 Federal Elections

The outcome of the 2017 German federal elections came as a shock to many, but especially to the German political class. Many scholars were taken by surprise as well, as they were not expecting such an extreme surge in populist votes (Chazan 2017; Hansen & Olsen 2019; Lees 2018; Siri 2018). Shortly before the election, scholars Vehrkamp and Wratil published a study with results indicating that, although populist attitudes in Germany were quite widespread, they remained “comparatively moderate politically.” They even went as far as to state that the federal election year would be “quite far from being a populist moment” (Vehrkamp & Wratil 2017, p. 7). However, there were others who predicted a change to the established political structure. Mere weeks before the election, Copley (2017) gave warning of particularly those living in the weakest regions, people who might feel ‘left-behind’, standing the risk of adopting radical attitudes against the ruling political parties. Mannewitz even forewarned the “spread of far-right ideology” by the AfD (2017, 225). Regardless, no one had expected the excessive populist support within the German public that was uncovered by the elections.

2.3.1. The Results

The elections were held on the 24th_{of September and 76.2% of eligible German voters cast}

their ballots in order to elect the new members of the Bundestag. The centre-right Union Parties, Christian Democratic Union (CDU) and Christian Social Union (CSU), collectively managed to rake in 33% of the votes and therefore remained the largest party grouping in the Bundestag. However, they achieved some of their worst results in recent years. Compared to the previous elections in 2013, they

(12)

lost 8.5% of their vote share. Many voters seem to have shifted their support from the CDU/CSU to smaller parties, with most of these votes going to the Free Democratic Party (FDP) and the AfD. These are in fact the two parties positioned right of the CDU/CSU in the political spectrum. Gaining 12.6% of the votes, the newly formed AfD drew away votes from all the big parties and managed to win the support from the most non-voters in the previous election. In doing so it landed 94 out of the 709 seats in the Bundestag and became its third largest party, with the Social Democratic Party (SPD) coming in second holding just over 20% of the votes. The other big parties, the FDP, the Greens and Die Linke all performed slightly better than they had in previous years (FT Reporters 2017). In figure 1 you can see the percentage of total vote share the six biggest parties managed to secure and whether their party support has either grown or shrunk since the previous federal election in 2013. It is clear to see that, whereas the CDU/CUS lost the most supporters to other parties, the AfD gained the most. Government formation ensued immediately after the results were announced. Because CDU/CSU still held the most Bundestag seats, namely 246, they took on the leading role of Formateur in the coalition negotiations. However, since the AfD managed to obtain such a large portion of seats, but was seen as an unacceptable coalition party by the centre due to its extremist beliefs, it was difficult for the CDU/CSU to find another small party to join into coalition with in order to reach a majority in the Bundestag. After a failed attempt to form a coalition with the FDP and the Greens in what is called a “Jamaica Coalition”, a Grand Coalition with the SPD was eventually formed in 2018 in order to avoid having to call for new elections. This put the role of main opposition party on the AfD (Lees 2018, 296).

Figure 1: Percentage of votes won by the six biggest German parties in the 2017 federal election and

(13)

The AfD has increased its vote share in all 299 German constituencies since the previous election, but the success of this radical right populist party was most astounding in the new Länder (federal states) that previously made up the communist German Democratic Republic (GDR)2_(FT

Reporters 2017). Here, the party won 22.5% of the votes, more than double of those in the West. Even after controlling for large sets of other socio-economic factors, researchers found that this East-West divide remains visible (Roth & Wolff 2017). When also taking the votes won by Die Linke into account, it can be concluded that nearly four in ten voters in East Germany voted for populist parties (Chazan 2017). Die Linke has traditionally always been strongest in the East, since it grew out of the GDR’s PDS and used to be a strong advocate of Eastern identity (FT Reporters 2017). However, when Oskar Lafontaine – former Minister President of Saarland, leader of the SPD, and Federal Minister of Finance – became part of, and from 2007 to 2010 even leader of, Die Linke, the party’s popularity soared in Saarland. This is the only Western federal state in which the results of Die Linke have ever surpassed those of the East German states. Die Linke has managed to reduce its East-West voting gap in recent years, but remains a distinct Eastern party even now (Mannewitz 2017, 226-7). The strong eastern success of the two populist parties in the 2017 election is clearly represented in figure 2 below.

Figure 2: The AfD and Die Linke are strongest in the East, share of party vote in % in 2017 federal

election (FT Reporters 2017)

(14)

2.3.2. Explanations for populist success (in East Germany)

On the day after the fall of the Berlin Wall in 1989, former West German Chancellor Willy Brandt spoke the following famous words: “what belongs together will grow together” (Willy Brandt Biography, 2019). This statement was of grave importance, since at the time of unification, many people questioned whether East Germans, who had been living in an autocratic and oppressive state for forty years, could be integrated into the western democratic society. However, Brandt was an optimist and successfully rallied the country behind his idea of a swift reunion. Consequently, studies have shown that there was strong support for democracy amongst easterners in the early 90’s. However, these people seemed to grow sceptical towards the democratic process over the years, due to the social and economic differences that endured between East and West (Doerschler 2015, 377). Furthermore, party affiliations that had had time to develop within a long democratisation process in the Federal Republic of Germany (FRG), were imposed on East Germans almost overnight. Simply transferring the political system to the new federal states, it was unable to grow naturally as people did not have the opportunity to shape it in ways westerners had over the years (Mannewitz 2017, 227). These and other factors could explain why East Germans seem to have a stronger disposition towards radical voting than those in the West (Mannewitz 2017, 223).

When looking at the causes of populist movements in societies, many scholars tend to focus on the notions of relative deprivation, or group grievances in general. According to Gurr (1993, 189), “group disadvantages – political and economic differentials, poverty, discriminatory treatment – have a major impact on grievances articulated by communal groups […]. Economic disadvantages, especially those associated with discrimination and poverty, are consistently correlated with economic and social grievances and with demands for greater political rights.” Likewise, Muis and Immerzeel (2017, 912) believe that populist voters are motivated by a “perceived loss of culture and economic deprivation.” Moreover, populisms central ideology of “us versus them” assumes that its supporters experience feelings of relative deprivation. This stems from a person’s vulnerability and weak position compared to others in society, where they feel like they are discriminated against and “never get what they deserve” (Elchardus & Spruyt 2016, 116). In this regard, people with populist attitudes in the 21st

century are often referred to as “the losers of globalization” (Elchardus & Spruyt 2016, 114; Kriesi 2006, 929) These people fear that their society is in decline as a result of the new challenges that are posed by the increasing internal diversification and detraditionalization of their country that comes with globalization. They blame the established political elites for this and are likely to turn to more radical political parties that advocate a change to the current system and a return to nationalist culture. In Western European countries, populist parties play into such people’s fears. In their manifestos they usually promise to combat the Islam, decrease diversity, strengthen the borders, reduce the European

(15)

Union’s influence, and return the country to its old norms and values (Elchardus & Spruyt 2016, 125-6). However, there are also those scholars that do not believe that populist voters are so called losers of globalization. Muis and Immerzeel (2017, 912) state that radical voters are often properly socially integrated and appear as perfectly normal people. Especially in the case of AfD voters in Germany, Hansen and Olsen (2019, 15) and Lengfeld (2017) stress that these voters were not really losers of globalization, but rather were driven by anti-establishment and anti-immigrant sentiments. Twenty-five years ago, Gurr already pointed out that “the growing flood of economic and political refugees from Third- and Second-World countries to the First World is creating new minorities who differ in ethnicity, culture, and often in religion from their host societies” (Gurr 1993, 190). He rightfully predicted what politically volatile situations such immigrant flows could lead to. Some believe that because there is a very low concentration of immigrants living in eastern Germany nowadays compared to the West, anti-immigration is a bigger concern among easterners (Burn-Murdoch 2017). This is because of “their perception of western German cities where large-scale immigration has caused a host of social problems” (Chazan 2017). Easterners fear that by taking in more immigrants in their new federal states, they as Germans will risk economic and social decline. In sum, the theory of relative deprivation as an explanation for populist tendencies appears to hold rather well for East German voters, but the notion that they are all “losers of globalization” is not generally believed to be true.

Based on previous literature, the theory of relative deprivation seems to be a well-established cause for populist voting, but why would this feeling be more palpable in the East than in the West? The main explanation for this is that even though we nowadays think of Germany as a single country, the gap in living standards and well-being between the East and West remains (Doerschler 2015, 383). Even though this gap has grown smaller over the years since unification, the economic differences are still clearly visible. Unemployment is higher and GDP is lower in the East, in job sectors without collective agreements employees in eastern Germany receive lower pay than their counterparts in the West, and large industrial plants and corporate headquarters are more commonly located in western federal states (Pollok 2017). The same trend can be witnessed for well-being factors. German citizens all throughout the country have reported to be more satisfied now than they were at the time of reunification, but the satisfaction levels of easterners still lag behind those of the West (Pollok 2017). Many East Germans still feel left behind, both socially and politically. This can be traced back to the time of unification. Many Germans belief that the Western standards may have been imposed too abruptly on the people living in the former communist GDR, especially since “forty years of living under separate political and economic systems had produced separate identities and cultural values between East and West” (Doerschler 2015, 381). During a public debate in 2018, a man asked Petra Köpping,

(16)

Minister of Integration in Saksen, why she always talked about immigrants. He directly asked her “Why don’t you first integrate us?” 3_{(Eijsvoogel 2018). A lot of easterners still see the reunification period}

and the first couple years that followed it as a traumatic experience. On the one hand, people were happy to be given the chance to start a new life in a free country, but at the same time, a hundred thousand people lost their jobs almost overnight when many companies were relocated to the West. They not only lost their way of making a living, but also their sense of dignity, by being relieved from their day jobs. Politicians expected such feelings to subside over time and disappear completely with the arrival of a new generation, but the bitterness remained. Research and public sentiment have shown that even people aged 18 to 27 still feel like second class citizens (Eijsvoogel 2018). A 2012 Gallup poll showed that easterners were twice as likely to describe their situation as “struggling” when asked about their prospects compared to westerners (Kahn 2015). Köpping herself believes that 21st century disgruntled East Germans do not wish for the return of the GDR, “but the people here have the feeling that their problems are not recognised by politicians, let alone solved” (Eijsvoogel 2018).

The unrelenting economic, social and political grievances in the East, which have built up over the years since reunification, have laid the foundation for eastern discontent with the established elites. Merkel’s response to the immigrant crisis in 2015 seems to have been the final straw. The opening of Germany’s borders to about a million migrants in 2016 did not fare well with many easterners. These people, already feeling like a deprived group in society, feared that refugees would overrun the country and take over their resources and jobs. The AfD saw an opportunity to play on this public angst and adopted a strong anti-immigrant sentiment in their 2017 election campaign, moving on from their originally dominant anti-EU position. In fact, a study by Hansen and Olsen (2019) found that the dominant factor impacting vote choice for the AfD in 2017 was anti-immigrant sentiment. On the contrary, Die Linke’s eastern support has always been strong, seeing as the party started off as an advocate for East German identity in the face of the region’s relative deprivation. After unification, many easterners have felt like second-class citizens and an outgroup of German society. According to Doerschler, “those who vote for Die Linke are those who see their standard of living threatened, and those who, while themselves being upwardly mobile take a critical view of overall social developments” (2015, 383). Furthermore, seeing as the SPD had experienced a rightward shift in recent years and there was no strong far-right movement before the AfD took to the political stage, Die Linke had been able to rake in many of the German protest votes (Doerschler 2015, 395). In the 2017 federal elections, Die Linke and the AfD were each other’s main competitors for protest votes, especially in the East. They were both populist, welfare state friendly, anti-establishment parties that were rather similar in their dissatisfaction with the country’s elites and current democratic system.

(17)

This competition has seen a clear winner: the AfD. It drew nearly half a million voters in East Germany away from Die Linke’s 2013 support (Hansen & Olsen 2019, 15). As the election results showed, no other party managed to increase its vote share as much as this right-wing populist party. In sum, the 2017 federal elections caused a shock to the German government formation. This was caused by having large parts of the population, but especially those residing in the new federal states, voicing their discontent with the established parties through populist voting. The fact that populism has found such traction amongst eastern voters can be explained by the regional disparities that persist between the two sides of the country and the fact that the party system is much more anchored in the West.

(18)

3. Life Satisfaction

As the previous section has shown, there are different explanations for the recent surge in populist sentiment. However, a large contributing factor that jumps out, is the feeling of being “left behind” and an evident discontent with the current state of affairs. These negative feelings about the political, economic and social sphere, which are particularly evident in the case of Germany’s eastern federal states, would indicate lower levels of subjective well-being (SWB). This section will continue to explain life satisfaction as a trustworthy and robust measure of well-being and furthermore provide evidence that there is in fact a considerable discrepancy between individuals’ life satisfaction levels in East and West Germany.

3.1. The Life Satisfaction Measure

Subjective well-being refers to people’s own evaluation of their lives. This evaluation can be split into two different categories: happiness and life satisfaction (Ng, Vaithilingam & Rangel 2017, 153). The former also carries the name affective happiness. It is seen as “an outcome of positive experiences, particularly influenced by close personal relationships”. The latter can also be referred to as evaluative happiness. It is seen as “an outcome of an evaluation process including social and material aspirations and achievements” (Hadler & Haller 2006, 169). Life satisfaction appears to be the more calculating component of subjective well-being and specifically refers to what people really think of their own lives when reflecting on it as a whole (Deaton & Kahneman 2010, 16489). Good life satisfaction scores are influenced by factors such as adequate trust in the government, good overall health and a sustainable income (Helliwell 2005, 7; Helliwell, Layard & Sachs 2012, 7). It is important to realise that happiness and life satisfaction are two different aspects of well-being. This was empirically proven for the first time by Bradburn and Caplovitz (1965, 19), who found that a person can be depressed or worried about a certain issue and still be generally very satisfied with life at the same time. Although they are not identical, the two measures have in fact been proven to be closely related to each other (Ortiz-Ospina 2017). Considering scope of this project, this thesis will only focus on the measure of life satisfaction in its analysis. This decision has been made on the basis of the OECD’s guidelines for the collection of subjective well-being data. These state that “if there can only be a single measure, preference is given for a life evaluation measure, in the form of a life satisfaction question on a 0 to 10 scale” (Helliwell, Huang, Grover & Wang 2014, 2).

Since the measure of life satisfaction has been introduced into scientific research, some doubts have been raised on its robustness and trustworthiness. In general, observed behaviour is always preferred over subjective data, which often stems from self-reported answers to questionnaires or interviews. As Diener notes, people may respond to subjective well-being questions in socially

(19)

desirable ways. If a person believes that answering that he is unsatisfied with his life will make him look like a failure, he will be inclined to report higher measures of life satisfaction (Diener 2000, 35). Furthermore, Di Tella, MacCulloch and Oswald believed that individual respondent’s answers on a well-being questionnaire are unlikely to be reliable, since there is no general standard for happiness or satisfaction. According to them this would make it impossible to allow for cross-person comparison in small sample groups (2003, 812). Schwarz and Strack stress three more aspects which they believe make subjective well-being data unreliable. First, they state that such self-reported measures can be contaminated by bias. Second, they note that these measures can be influenced by a respondent’s mood or other situational factors at the time of answering the question. Lastly, they underline the framing effect, which means that the order of the questions on a questionnaire can influence reports of well-being (1999).

Nevertheless, all of these doubts have been challenged over time with the help of empirical research. Schwarz and Strack’s second argument against SWB as an unreliable measure has been disproven, since researchers have found that people tend to adapt to changes rather well and that even important life events, such as marriage or becoming crippled, have surprisingly little long-term impact and do not have an extreme influence on a person’s average life satisfaction (Ortiz-Ospina 2017). Richard Easterlin’s 1974 study underlined this, showing that, at least in developed countries, when an individual’s income increases, self-reported well-being levels do not rise with it. In regards to the validity of the SWB measurement, Diener himself showed in a 1984 study that people who say they are happy are rated by the people close to them as happy as well, and he stated in a 1996 paper that SWB measures could be correlated to related measures, such as optimism (Diener, Lucas & Suh 1996). A 1991 study by Pavot et al. found that people who report themselves as happy do in fact smile more. Having expressed their doubts, Di Tella et al. in fact find that higher levels of national reported well-being are associated with lower national suicide rates (2003, 812). Moreover, Sutton and Davidson found that subjective well-being measures correlate with physiological responses, such as cortisol levels, and with activity in the brain associated with satisfaction and pleasure (1997). In general, various surveys have confirmed that people who rate themselves as happy also tend to sleep better and more frequently verbally express positive emotions (Ortiz-Ospina 2017). Furthermore, it is important to mention that the reliability of subjective well-being measures has been assured by confirming its possibility for replicability (Helliwell, Layard & Sachs 2012, 17). Even economists have of late begun recognising measures of subjective well-being as acceptable indicators. Kahneman and Krueger go as far as to state that measures of satisfaction could be relevant for future decisions (such as vote choice), since “retrospective evaluations are relevant for some subsequent choices, so measures of satisfaction may be relevant for future decisions” (2006, 7). In sum, it can be concluded

(20)

from the evidence that the survey-based measure of life satisfaction is in fact a consistent and reliable measurement. An added benefit is that the life satisfaction question is very easily included in all sorts of surveys or questionnaires and that therefore data on this topic is widely available.

3.2. Lower Life Satisfaction Levels Linger in East Germany

Germany, together with other Western European countries, usually ranks very high in global surveys of life satisfaction. Therefore, the country seems to be doing rather well in taking care of the well-being of its people. However, the German national average score does not show the large inequalities in life satisfaction. These regional disparities are truly remarkable, especially between the East and West (Ortiz-Ospina & Roser 2017). These inequalities are a known fact (see figure 3). Even in the most recent Government Report on Wellbeing in Germany, a distinction between the two regions is made in almost every sub-section, showing the results of East and West separately in the corresponding figures (Die Bundesregierung 2017). The fact that this is still necessary nearly thirty years after the fall of the Wall and German unification, shows that the country is far from united. Chancellor Helmut Kohl’s famous statement “Wir sind ein Volk!” (“We are one nation!”) at the time of unification remains part of a hopeful future (Connor 1993, 382), since three decades later, Germany has still not been able to fully close the gap between living standards and well-being between the East and the West (Doerschler 2015, 383).

Reported life satisfaction is on average significantly lower in the eastern states compared to the western federal states. After unification, those living in the East have experienced a steady increase in levels of life satisfaction, with the exception of a sharp drop of 0.6 points (on a 0-10 scale) in 1990. This sudden drop preceding the improving levels was due to the inevitable initial adjustment phase, which brought changes to the labour market and caused a rapid rise in unemployment for easterners. Similarly, life satisfaction for West Germans saw a slight decrease in the years just following the unification period, which can be attributed to an adjustment to the new order. However, on average the western levels did not change much. As a result, the country has witnessed a convergence between the two regions’ levels of life satisfaction that was strongest in the early 90’s, but continues until this very day (Pfeifer & Petrunyk 2015, 2). Figure 3 portrays how the East-West gap in life satisfaction still exists.

(21)

Figure 3: Life satisfaction in German federal states, measured between 2012 and 2016 (Ortiz-Ospina

(22)

Researchers have looked at whether this life satisfaction gap between the East and the West cannot simply be explained by economic and socio-demographic factors. Pfeifer and Petrunyk (2015) found that only about half of the gap can in fact be explained by differences in macroeconomic conditions between the regions, such as the GDP per capita and unemployment rates. Age was found to have somewhat of an impact as well, as life satisfaction levels seem to converge increasingly for younger age-cohorts. However, gender does not appear to have had a significant influence on life satisfaction levels here. Having accounted for these factors, they concluded that the gap must for some part also be attributed to “individual non-monetary and unobserved differences between people living in East and West Germany” (Pfeifer and Petrunyk 2015, 14). Such differences would include the before mentioned feelings of relative deprivation and a loss of identity that resonates in easterners. This idea is supported by the fact that older generations, those who have lived in communist GDR at least for some time, are more likely to report lower satisfaction levels. Diener’s famous work on well-being underlines this, as he believed that “cultural and societal differences are likely to be important determinants of international4_{differences in subjective well-being” (Diener 2000, 39).}

In his book Ethnic Groups in Conflict, Horowitz (1985) states that the division of a population into groups only takes a few differences and that “groups can form quickly on the basis of simple division into alternative categories. Once groups have formed, group loyalty quickly takes hold. In interactions between groups, favouritism toward ingroups and discrimination against outgroups are demonstrated. What group members seem to desire is a positive evaluation of the group to which they belong. A favourable evaluation is attained by comparison to other groups in the environment” (Horowitz 1985, 144). The division of Germany during the Cold War has created exactly such a situation, the consequences of which can be seen even now. Easterners still compare themselves to their fellow citizens in the West and see that they are economically better off and generally more well-integrated into society. Even in something as mundane as the German football league such regional disparities show, seeing as last year only one member of the national team originated from an eastern federal state (Urlaub 2018). The East German outgroup feels like they have a competitive disadvantage compared to their western countrymen on many levels in society and constantly have to compete for scarce resources, such as property rights and jobs (Lake & Rothchild 1996, 44). In sum, the feeling of relative deprivation, which comes from the evident cultural and societal differences that remain between the two sides, has had a considerable influence on East German levels of life satisfaction.

4_{Although in this case we talk about regional and not international differences in well-being, the fact that the}

(23)

4. Theory: The Impact of Life Satisfaction on Political Preference

This chapter will use the information from the previous two to support this thesis’ main argument, namely that life satisfaction and populist party preference are (negatively) correlated. First, an overview of the existing literature on the relationship between individual well-being and political participation and party preference will be presented. This is followed by examples from the literature which show that a correlation between life satisfaction and party choice has been found in the past. The chapter ends with a specific focus on the motivation of populist voters.

4.1. Findings up until now

There is a substantive gap in the literature when it comes to research on the relationship between subjective well-being and political attitudes. Although there is plenty of empirical research into the political causes of well-being, its political consequences and possible effects on political attitudes and behaviour remain largely understudied (Flavin & Pacek 2014, 2). This is rather surprising, considering that studies, such as the one by Kahneman and Krueger mentioned before, have proven that an individual’s well-being can be of considerable influence on his or her future decisions. This naturally also includes political participation and voting (Ng, Vaithilingam & Rangel 2017, 151). Furthermore, thanks to recommendations of the EU, OECD and the US National Research Council, among others, many countries have started collecting self-reported well-being data in order to measure societal progress and guide policymaking (Ward 2015, 2). The data are therefore becoming increasingly widely accessible and applicable to societal affairs. As a matter of fact, research on the relation between subjective well-being and political participation does exist more broadly. Veenhoven (1988) was one of the first to link life satisfaction with political participation. He worried that “happiness would lead to an emptying of democracy” (p. 342) since he was concerned that citizens might become nonchalant towards politics when they were already highly satisfied with their own lives. However, he was able to refute this hypothesis and eventually found that the more satisfied citizens are with their lives, the more they become concerned with social and political issues. Several more recent studies found results along these lines in different environments. Flavin and Pacek’s study (2014) used an analysis of 60 countries to show that happier people are generally more likely to vote and participate in politics in other ways. Weitz-Shapiro and Winters (2011) performed a similar study, but did so solely for a sample of younger democracies in the developing world. From a sample of eighteen Latin American countries, they concluded that individual happiness is a likely cause of political participation. Interestingly, Zhong and Chen (2002) have even found a positive correlation between life satisfaction and political participation in authoritarian China and Bahry and Silver (1990) found corresponding results for the former communist Soviet Union. Furthermore, research has been done

(24)

in the opposite direction as well, by looking at whether participating in politics increases citizens’ levels of life satisfaction. Here, results showed that “democracy is positively correlated with individual levels of being. The opportunity to participate in the political process [is] related to individual well-being” (Owens, Videras & Willemsen 2008, 1001). In sum, it seems safe to say that Veenhoven’s original assumption has been adequately disproven and that happiness does not lead to an emptying of democracy, but in fact strengthens it.

However, this thesis surpasses the focus on political participation. It does so by looking at political party preference as a consequence of life satisfaction. There are a few studies that have examined this effect. Using cross-country panel data, Ward found that “a country’s aggregate level of subjective well-being is able to account for more of the variance in government vote share than standard macroeconomic variables” (2005, 1) from which one can argue that the average life satisfaction level in a country is therefore a robust predictor of election outcomes. Besides merely looking at political participation, Flavin and Pacek (2014) used their data from 60 countries to examine whether people who are more satisfied with their lives are more likely to express political preferences on the liberal or conservative side of the political spectrum. They build upon the idea of economists, that happier people are more risk averse and more reluctant to change. Such happy people tend to save more and spend less and are more sensitive to loss. In accordance to this, their study showed that individuals with higher self-reported levels of life satisfaction are more likely to support the maintaining of the political status quo and are therefore more likely to identify with conservative leaning political parties. Ng et al. found similar effects for the people of Malaysia concerning the status quo, and draw the conclusion that “respondents with higher levels of life satisfaction will have a higher probability of voting for the ruling party” (2017, 149). A similar conclusion was reached by Liberini et al. (2013). They used the British Household Panel Survey Data to find a positive relationship between life satisfaction and voter preference for supporting the ruling party. In sum, previous research has vastly shown that political party preference towards status quo ideology and ruling parties is positively linked to individual life satisfaction levels.

This thesis focuses on individuals’ motivation to vote for a populist party and whether this motivation is influenced by life satisfaction. There are some studies that try to explain populist voting by looking at the levels of trust and cynicism in society. It is generally believed that a lack of trust in institutions leads to “a loosening of the social and political fabric” and can therefore be blamed for populist mobilisation (Fieschi & Heywood 2004, 292) This is supported by the idea that populist voting is motivated by economic and cultural grievances. The only existing study to date that has a similar research question to the one in this thesis is by Elchardus and Spruyt (2016). It examines the support for populism as a consequence of dissatisfaction with personal life. The researchers found that populist

(25)

support is not directly influenced by low levels of life satisfaction. According to them, “support for populism appears foremost as a consequence of a very negative view of the evolution of society – declinism – and of the feeling of belonging to a group of people that is unfairly treated by society” (Elchardus & Spruyt 2016, 111). However, the results of their study are the opposite of what this thesis expects to find. The prediction that has been made here, is in fact based on the many studies that have shown that higher life satisfaction motivates people to vote for the status-quo, which when reversed, can lead one to believe that lower life satisfaction anti-establishment populist voting.

4.2. The theory

Section 4.1. has brought together the concepts of populism and individual overall life satisfaction. In doing so, it has laid the foundation for the theoretical concept that will be explored throughout the rest of this thesis: the impact of life satisfaction on populist party choice. The link between these two factors is based on an observation of recent German figures. Is it a coincidence that average life satisfaction scores are considerably lower in contemporary East Germany than in West Germany and that double the percentage of East Germans voted for a populist party in the most recent federal election compared to citizens of West Germans? Possible explanations that have been introduced by researchers for the latter observation are focussed around the socio-economic differences that still exist between the two sides of the modern united country. At the core of these differences appears to be a sentiment among easterners of being left behind, being discriminated against, and feeling like an out-group in their own country. These are all explanations for the lower levels of life satisfaction that persist between Germany’s East and West, even thirty years after the fall of the wall. Section 4.1. has shown that a link between life satisfaction and political party preference has been found in numerous studies on the subject. Now, based on the outlined existing research, the following hypothesis will be tested in the subsequent chapters:

H1: Individuals who report lower levels of life satisfaction will be more likely to express support for

populist political parties.

The reason that life satisfaction is expected to have a negative influence on populist party preference, originates from the realisation that East Germans experience both substantially lower levels of life satisfaction and higher populist voting behaviour, in comparison to their fellow citizens in the West (see figures 2 and 3). The causes for both these occurrences seem to overlap surprisingly well on several different factors, such as a lack of trust in the government and a feeling of relative deprivation (e.g. immigrants taking jobs, lower GDP and higher unemployment rates). The argument made in this thesis extends on previous research, using it as a stepping stone to explore a new area of the academic field of predicting populist voting sentiment, using the case study of modern East and West Germany to do so.

(26)

5. Research Design

This research section outlines the data and the empirical analysis that was used in an attempt to prove the previously stated hypothesis correct for the case of Germany’s 2017 federal election. The study tests whether higher life satisfaction scores reported by German citizens in 2017 are negatively correlated to the likelihood of expressing support for the populist AfD or Die Linke, instead of another non-populist German political party in that same year. The study is carried out based on data from an existing dataset and thus relies on the accurate collection of this data by other parties. For the main analysis, only data collected in the year 2017 is included. The amount of relevant observations (N) for each variable described in the data section are therefore for that single year alone.

5.1. Data

The data used in the analysis have been procured from the German Institute for Economic Research (DIW Berlin). The specific dataset that is used is called the soep.v34. This dataset contains all the results from the Socio-Economic Panel (SOEP) study, a wide-ranging representative longitudinal study of private households that is conducted every year amongst over 25,000 individuals (33,265 in 2017). For ethical reasons, the datasets have been completely anonymised by the DIW. The most recent version, number 34, has a collection period of 1984-2017 and contains weighted information on thousands of variables. These variables are divided over 513 different data files. The following analysis will only make use of a maximum of ten variables of all the possible usable indicators provided. I have merged these indicators into a single datafile and have in some cases had to recode the variables. All of the variables are described in section 5.1.1. up until and including section 5.1.3., and include an explanation as to why they were selected. Since an answer to all the used indicators is provided for only some of the respondents, the number of observations used to run the analysis comes down to a total of 1,090.

5.1.1. Dependent Variable

The dependent variable in this study is individuals’ political party preference. The SOEP survey includes an item where respondents are asked which political party they lean toward. This is Question 145: “Which party do you lean toward?” followed by a list of active political parties from that year in Germany. The soep.v34 dataset thus provides the information to determine respondents’ partisan inclination in the 2017 federal election year. A person’s political partisanship has been proven to be a solid indicator of the party they tend to vote for, as long as partisanship and actual voting behaviour are recorded with little time in between (Whiteley 1988, 961). Since I use the 2017 results from the SOEP survey, party preference is measured close to the 2017 federal election, hereby creating a good indicator of respondents’ real voting behaviour. In order to measure for populist voting sentiment, the

(27)

variable is coded into a populism dummy variable. Here, populist voting tendency includes the responses of people who have stated that they lean towards voting for either Die Linke or the AfD. Logically, the non-populist voting category includes all other party choices. This grouping is based on the literature study previously presented.5

5.1.2. Main Independent Variable

The independent variable of this study is individuals’ self-reported life satisfaction. The very last question (Question 175) of the SOEP survey asks respondents to report how satisfied they are with their life in general on a scale of 0-10, where 0 means completely dissatisfied and 10 means completely satisfied. This measurement is called Cantril’s Self-Anchoring Ladder of Life Satisfaction (Cantril 1965). It is the OECD’s recommended measure of life satisfaction, since research has shown it to be the purest measure of life evaluation (Deaton & Kahneman 2010, 16492). It is entered into the analyses as a continuous variable.

5.1.3. Control Variables

I will be controlling for a number of different variables, which could possibly influence the impact life satisfaction has on populist party choice. These control variables have all been found in previous studies to have at least some influence on people’s decision to vote for a populist party and are listed below. The control variables are divided into two groups: demographics and attitudinal controls. The number of observations that are reported for each variable’s possible categories represent the ones used in the analysis for 2017 and only include the observations for which every single variable was reported.

Demographic

Age of respondents has been proven to be an important demographic factor in predicting

political allegiance (Burn-Murdoch 2017). It looks at the generational divide within the population. I have computed the variable ‘Approximate age in 2017’ by subtracting the registered year of birth from 2017, hereby creating a new continuous variable6_{. Respondents range in age from 18 to 71, with an}

average age of 46. Elchardus and Spruyt (2016, 124) have found no effect whatsoever of age on populist voting preference, but researchers Roth and Wolff (2017) found contradictory evidence, which suggests that older people are more likely to harbour support for a populist party. I therefore

5_{I have also run the analyses with the DV constructed in different manners: (1) including NDP and the Pirate}

Party in addition to Die Linke and AfD, and (2) with the DV only including the AfD. Results (not reported) were essentially the same.

6_{Running the analysis with age as a categorical variable, divided into different age cohort groups, did not change}

the outcome. I have therefore chosen to keep it as a continuous variable in order to keep the overall outcome from becoming too cluttered.

(28)

expect age to either not have any influence on my dependent variable at all, or to slow a slight populist preference amongst the older age cohort.

Education has been found to negatively correlate with populist voting in Germany, where

people with higher school leaving degrees were less likely to vote for a populist party (Elchardus and Spruyt 2016, 123; Roth & Wolff 2017). The explanation behind this effect is that people with low levels of education are more vulnerable to changes in society and also miss out on some of the socializing effects that come with higher education (Elchardus and Spruyt 2016, 117). Education is therefore included as a control variable in this study. None of the SOEP’s datasets provide an indicator of highest obtained educational degree in survey year 2017. I therefore recoded the variable for tertiary degree into: people who have obtained a university degree, doctorate or some other form of tertiary education (N = 371), and those who never completed any sort of tertiary education (N = 719).

Employment status will be controlled for as well, just as it was in the study of Flavin and Pacek

(2014), since it is an important indicator of one’s economic position. Previous research by Roth and Wolff (2017) has found that unemployed persons are, surprisingly enough, less likely to vote for the populist AfD in Germany. On the other hand, other research has found no effect of employment status on populist preference (Elchardus and Spruyt 2016, 125). Based on these contrasting findings, employment status is an important indicator to take into consideration in this study. A dummy variable has been created from the original employment variable, creating a group of people who are regularly either full-time or part-time employed (N = 972) and a group that is not, either because they are in vocational training, employed in a sheltered workshop, or simply unemployed (N = 118).

Gender is a dichotomous variable, with women (N = 495) and men (N = 595) being the only two

groups reported in the data. Empirical studies have demonstrated that populist parties, on both the left and right side of the political spectrum, are more strongly supported by men than by women (Spierings & Zaslove 2017, 839). I therefore expect to find a similar influence of gender in my results.

Income, like education, was found to be negatively associated with populist political party

preference (Roth & Wolff 2017). However, other research has found no effect of income, which is part of one’s economic status, on populist preference (Elchardus and Spruyt 2016, 125). Based on these contrasting findings, income is another important indicator to take into consideration for my analysis. I have transformed the continuous gross income variable from the SOEP dataset into a categorical dummy variable with low (€0 – €1,410, N = 234), middle (€1,411 – €2,640, N = 272) and high (€2,641