• No results found

The Left-Right Scale : An analysis of its connection to preferences on economic issues

N/A
N/A
Protected

Academic year: 2021

Share "The Left-Right Scale : An analysis of its connection to preferences on economic issues"

Copied!
51
0
0

Loading.... (view fulltext now)

Full text

(1)

Linköping University | Department of Management and Engineering Bachelor’s thesis, 15 credits| Political Science 3 Spring 2019| LIU-IEI-FIL-G--20/02219--SE

The Left-Right Scale

-An analysis of its connection to preferences on economic issues

Anton Karlsson

Supervisor: Mikael Blomdahl Examinator: Mikael Rundqvist

Linköpings universitet SE-581 83 Linköping

(2)

Abstract

This thesis deals with the nature of the Left-Right scale. Theories and ideas about the Left-Right scale have been tested by a mixture of quantitative and qualitative methods. The research questions are, in short, firstly if voters’ preferences on political issues, where economic issues are tested in this specific thesis, can consistently explain voters’ Left-Right self-placement, secondly if this level of explanation can vary depending on changes in national political discourse, and finally if a high level of correlation between issue and Left-Right self-placement facilitates the matching process of parties and voters of similar ideological conviction. A regression is run on data from the World Values Survey and the European Values Study, data which consists of survey questions about attitudes towards economic preferences and self-placement on the Left-Right scale. The selection is restricted to old democracies, as there is some discrepancy between new and old democracies with regards to the capacity of the electorate to relate to the Left-Right scale. The independent variable in this regression is attitudes toward different economic issues, while the dependent variable is self-placement on the Left-Right scale. The first question is answered by the regression, which is if preferences regarding issues can explain self-placement on the Left-Right scale. The answer that is given is that there is indeed a consistent relationship between preferences on economic issues and self-placement on the Left-Right scale, over time. Albeit it is higher for some countries, like Scandinavia, and lower for other countries, like Japan and Ireland. The two remaining questions are answered by case studies, selected through the method of least likely and most likely cases. These are Netherlands, Sweden, Japan and France. In essence, these cases show that the Left-Right scale is indeed flexible, and can adapt to the current political discourse, and that a high relationship between preferences on political issues and self-placement on the Left-Right scale facilitates the matching of voters and parties of similar ideological conviction.

Keywords: Party Linkage Model, Left-Right Scale, GAL-TAN, Ideology-Voting , Voting-Behavior, Political Parties, Japan, Sweden, Netherlands, France

(3)

Table of content

1.Introduction……….1 1.1 Introduction………...1-2 1.2 Purpose and Research Question………2-3 1.3 Limitations and Delimitations………...3-4 1.4 Previous Research……….4-5 2.Theory………..6

2.1 Party Linkage Framework……….6-10 2.2 Theories About the Left-Right Scale……….10-16 3.Method……….17-19 4.Analysis……….20 4.1 Economic Preferences and the Left-Right Scale………20-23 4.2 France and Netherlands………..24 4.2.1 France……….24-28 4.2.2 Netherlands………29-31 4.3 Japan and Sweden………..32-39 5. Conclusion………...40-43 Appendix……….44 References………...45-48

(4)

1

1. Introduction

1.1 Introduction

For the public at large, there are perhaps few concepts of political science as renowned as the Left-Right scale. As Dalton et al. (2011, p.87) affirms, we can conclude that most people can relate and position themselves on the Left-Right scale. In fact, about 90 percent of the respondents across all participating countries have an idea of where they are located on the Left-Right scale, including respondents of recent and old democracies. Dalton (2006) reached a similar result when he looked at data from the World Value Survey.

This prominence that is enjoyed by the Left-Right scale should encourage us to study it closer. As the concept seems to transcend new and old democracies alike, it is crucial for us to know how people relate to, and understand, the Left-Right scale across the world. The fact that the Left-Right scale is such a global phenomenon implores me to underline that its meaning, and people’s understanding of it, will shift depending on what context it is used in. As Dalton et al. (2011, p.85) points out, the meaning of the Left- Right scale varies between nations, individuals and even classes within a country. The right might represent one thing for a Japanese citizen and another thing for a Swedish citizen. Likewise, the right might be one thing for a manual laborer and another thing for an economist.

The fact that the meaning of the Left-Right scale can fluctuate depending on context, as I have described above, is both its weakness and its strength. The perks of having such an ambiguous measurement as the Left-Right scale is that it can be applied in different circumstances. As is aptly remarked by Dalton et al. (2011, p.86), the Left- Right scale shows what specific issue is most prominent for a specific individual or country. On the other side of the coin, this might reduce the value of the Left-Right scale as a communicating vessel between parties and voters if the dominating issues are complex and hard to embody with a simple one-dimension (ibid.).

The prominence of the Left-Right scale is obviously not the only argument for why it needs to be studied, even though it is a compelling one. It is also an important heuristic tool according to some political party theories. One of those is the democratic linkage

(5)

model that hypothesizes that an ideological linkage must be fulfilled, with the aid of the Left-Right scale for example, in order to have a stable party system in a democracy (Dalton et al., 2011, p.7).

However, not all agree that a one-dimensional scale, such as the Left-Right scale, is enough to fully encapsulate political discourse. Some researchers, like Hooghe, Marks and Wilson (2002), mean that another dimension, like GAL-TAN, is required given the development of new political issues and post-material values. Other researchers mean that the Left-Right scale will do, like Dalton et al. (2011), as it is flexible and adaptable as mentioned previously. This thesis seeks to contribute to this discussion by discovering if the Left-Right scale is as flexible as some theories, and if the Left-Right scale can satisfy the conditions that the ideological linkage sets out.

1.2 Purpose and Research Questions

As was made clear in the introduction, the Left-Right scale must manage to perform several roles to function as desired. There are three main functions, that the Left-Right scale should fulfill, that this thesis looks at. Firstly, if people mainly talk and discuss issues pertaining to economic matters, such as income equality and free enterprise, then this economic dimension should be reflected by dividing the voters and parties along the Left-Right axis according to their preferences on these issues. Secondly, should other issues swing up in importance in a consistent manner, while old issues wanes in importance, then these new issues should theoretically replace the old ones when people relate to the Left-Right scale. In other words, a person who self-describe oneself as left wing previously, because of a desire for income equality, might self-describe oneself as right wing as the person relates to new issues and a new political discourse, for example climate change or post-material values. Finally, the Left-Right scale must be able to match the appropriate voter with the appropriate party, so that they share common interests. A party that wants to work for free enterprise must be able to match with voters of the same view, which is ideally done by both parties and voters orienting themselves according to the Left-Right scale. This concerns the ideological linkage that the Left-Right scale

(6)

3

parties should have an easier time finding voters with the same ideological conviction if the parties’ main issues are dominant on the Left-Right scale.

This discussion is summarized in the following three questions:

(1) To what degree can voters’ preferences about political issues explain their Left-Right self-placement?

(2) Can changes in national political discourse change the degree a certain political issue explains self-described Left-Right position?

(3) Does the Left-Right scale facilitate the matching process of parties and voters of the same ideological conviction?

The purpose of this thesis is to get an idea of how well the Left-Right scale fulfills the purpose that I have outlined here, by answering the questions above. Because, in the end, does one dimension make perfect, or is it not enough?

1.3 Limitations and Delimitations

The main limitation that has worked against this thesis is the lack of relevant questions posed in the surveys that the data is comprised of, and the fact that they changed from survey year to survey year and were limited to questions about traditional morality, religion and economics. Another contributing problem is participating countries varies from survey to survey, which makes it harder to track trends over time.

As for delimitation, I have chosen to focus on the traditional Left-Right dimension, largely related to economic issues and preferences. This is mainly due to time and space constraints. However, valuable insights can still be gathered from focusing on one dimension of issues. Several hypothesized functions of the Left-Right scale can be tested, such as if there are trends in regards to correlation between Left-Right self-position and preferences on a certain issue, if changes in national discourse can change this trend and if a high correlation can increase the matching of voters and parties of similar minds. Adding more issues would paint a more complete picture, and put the findings in a better context, but would in no way take away from the insights gained from looking at just economic issues. Thus, this thesis has, in some sense, traded depth in favor of some width,

(7)

and the ability to test more hypothesized functions of the Left-Right scale.

The focus on economic issues is also due to the fact that those questions were coded on a 1-10 scale, which you need in order to treat the data with statistical methods in a reliable way, at least with regards to the regressions I want to run, while questions pertaining to morality and religion had a narrower scale. Furthermore, questions about immigration and attitudes toward immigration were somewhat limited and suffering from a narrow scale as well, which contributed to the choice of limiting the focus to economic issues. This thesis is also limited to old democracies, as there is a difference in how well voters in old and new democracies can relate themselves to the Left-Right scale, as is explained previously and in the method section.

1.4 Previous Research

There are several studies related to the Left-Right scale. For example, Fuchs and Klingemann (1986) focus on individuals’ self-placement on the Left-Right scale in their paper as well. Perhaps the most interesting finding they present is that the meaning of the Left-Right scale can change over time. As new issues emerge in the public discourse, they will transform what people associate with the left and right. Rather than paving way for new measurements and terminology, new conflicts will rather conform to the Left-Right scale, which is in line with the findings of this thesis.

Perhaps the most relevant study for my work is Dalton et al. (2011, pp.91-96). The first key finding that is presented is that even if you put old and new democracies together, 90 percent of the public has a left-right position on average. They also inspired the model this thesis uses to correlate voters’ preferences on economic issues and their Left-Right self-placement. The authors also looked at other issues, such as cultural issues, several background variables and environmental attitudes, but found that cultural and economic preferences have the highest correlation. Another study, by Jou (2010), have a similar view that both the electorate of old and new democracies can relate to the Left-Right scale, and then support parties accordingly, when looking at democracies in, or in proximity, to East Asia. Although, old democracies scored better. Furthermore, Sigelman and Yough (1978) mean that the Left-Right scale indeed is flexible and can be adapted to the most salient issue in every democracy throughout the world. Laponce (1981) has a

(8)

5

similar view, as he means that the Left-Right scale has the quality to transcend cultures. Naturally, not everyone agrees that the future for the Left-Right schema is certain. Mair (2007) states that the Left-Right scale has been resilient, but that it risks becoming a complicated tool, and thus straying from its purpose of helping to simplify complex political issues and aiding in efficient communication. Stokes (1963) takes an even more critical stance. Essentially, he criticizes the notion that political thoughts and ideas could be boiled down to a single dimension, because political parties, voters and issues are not easily conformed to a unidimensional space.

In conclusion, much of the previous research have a positive answer to this thesis’ research questions. Voters can, throughout the world of new and old democracies, relate themselves and parties to the Left-Right scale, and with the aid of the Left-Right schema find parties of similar ideological conviction. There is an agreement that the content of the Left-Right scale can change meaning depending on national context and what issues are important for the particular time. What this thesis will add to previous studies is a deeper focus on the content of the Left-Right scale, and how that can change depending on national context and changes in what issues are most salient in the national political discourse. Case studies, chosen by the method of least likely and most likely cases, are used to get a greater understanding of this issue.

(9)

2 Theory

In this section, there is information about the theories that are at the center of this thesis. These are the ”party linkage model” and theories about the nature of the Left-right scale, which is devoted one subdivision each. The party linkage model has especially given me understanding of parties’ role in our society and what functions they must fulfill for our modern democracies to be legitimate and function properly. The theories about the nature of the Left-Right scale, on the other hand, are theories that work complementary to the party linage model. They paint the Left-Right scale as a tool to analyze voters’ policy preferences, and how they relate to other parties, that is warranted by the party linkage model. They claim that while the Left-Right scale does not always provide clarity, it is very flexible and has withstood the test of time. In other words, the theories about the Left-Right scale has inspired the research questions of this theory, by testing the properties that the theories claim the Left-Right scale has, while the party linkage model explains why a tool like the Left-Right scale is important in democracies dominated by a party system.

2.1 Party Linkage Framework

The Party Linkage Framework is a model that explains how parties should act and what its role are in our modern democracies. Its thesis is essentially that parties are an important phenomenon in our modern democratic systems. There are five linkages that connect the parties to the electorate: the campaign linkage, participatory linkage, ideological linkage, representative linkage and policy linkage. Furthermore, the theory claims that for the democratic system to function properly, these linkages need to be strong (Dalton et al., 2013, p.7).

(10)

7

Obviously, the Party Linkage Framework is not the only party model out there. Thus, to understand the Party Linkage Framework, one must understand the history of the field of party research. For a substantial amount of time, parties had entirely different conditions to operate in. According to Katz (1990, p. 143), political parties have played a central role in the formation and running of our modern democracies since their inception in the latter half of the nineteenth century, in the form of ”mass parties”. The aforementioned author also asserts that this role comprised of serving as a link, some sort of communication vessel, between the public at large, who did not actively participate in political organizations, and the political elite. The positive aspect for the political elite is that they can communicate with the local opinion leaders. Thus, they can both communicate their message as well as receiving crucial feedback in order to stay in touch with the people. For the masses, parties open opportunities to participate in the political process and provide further feedback than binary choices and other choices of limited scope that are provided at the ballot box. Katz (1990, p. 143) also theorizes that this development is linked to another contemporary trend, namely the expansion of the state’s social obligations and the popular participation in politics.

However, this trend of mass parties would come undone in the final decades of the 20th century due to several factors, which I will mention briefly. Based on Katz (1990, pp. 144-146), these factors are, among others, higher material wellbeing, decline in solidarity between social groups, the dominance of television in spreading political messages and financial support from the state. These explanations are simplified, but a deeper analysis can be found in Katz (1990). This fate of declining party membership, that I have described, can be observed through studying party membership statistic and mass surveys. Katz (1990, pp. 148-149) shows with this kind of data that party membership was generally in a clear decline in Europe since the 1970-ies.

This phenomenon has encouraged some researchers to draw the conclusion that the role of the party, in our modern democracies, is diminishing (Dalton et al., 2011, p. 9). Peter Mair is one of them. According to Mair (2005, p. 24), there is indeed a tendency of failing parties. He theorizes that it is caused by parties moving too close to the institutions of the state and abandoning their representative function. Mair (2005, p. 24) means that this could give parties temporary positive effects that could compensate for the consequence of being less representative, but that parties are doomed to fail if they stop fulfilling their representative function as it is required to legitimize their role. He calls

(11)

for a rethinking of democracy, both in theory and practice, in a landscape where the role of the party is diminished. Mair (2005, p. 24) phrases it as the emergence of ”democracy without the demos”. However, much as Dalton et al. (2011, p. 14) remarks, parties persist in our modern democracies. They adapt and change, rather than disappear. Thus, they mean that there is no strong indication that parties are failing. It is rather the mass party ideal that is becoming less applicable on the modern party.

With this discussion in mind, I have chosen to primarily use something that Dalton et al (2011, p. 16) refers to as ”party linkage framework” as a theoretical tool to analyze parties and their performance. Just like them, I believe that the role of parties in our modern democracies is not fading away. Rather, they change to tackle new circumstances. Although the unit of analysis in question of this study is not parties per se, but rather voters and how they relate to the Left-Right scale, the theory does put everything into context. The Party Linkage Framework provides invaluable insight into why it is important for tools like the Left-Right scale to function properly for a democracy to function well. I will elaborate on this further on, but first I will briefly establish exactly what the Party Linage Framework is.

Basically, the party linkage framework explains that in order for a democracy to function properly, it needs to fulfill a chain of five linkages. These linkages constitutes the so called ”chain of democratic linkage”. As can be seen by figure 2.1, the linkage is

Figure 2.1: The chain of democratic linkage

Citizen Policy

Source: Figured inspired by Dalton et al. (2011, p. 7) Note 1: The arrow shows that the process takes voters’ preferences and ideas, and then turn them into policies through the five linkages.

Comprised of the campaign linkage, the participatory linkage, the ideological linkage, the representative linkage and the policy linkage. These are explained by Dalton

Campaign -Candidate selection -Organize election -Structure debate Participatory -Mobilize voters Ideological -Offer policy choice -Voters choose Representative -Representative government are formed Policy -Deliver promised policies

(12)

9

Et al. (2011, pp. 7-9), which I will summarize shortly. The campaign linkage and participatory linkages are fulfilled when parties select and educated candidates for offices, organize the election and debates and mobilize the voter to go to the voting booth. The ideological and representative linkage concerns parties educating the populace and presenting them with choices to choose from. This must result in elections that achieve in a result that matches the wishes and preferences of the people. One could dive a lot deeper in each and every linkage, but that would steer away from the objective of this study and not contribute to further understanding of the problem at hand. The linkage that I look at particularly in this study is the ideological linkage, which I will elaborate on further.

Important to mention, concerning the ideological linkage, is that it relies on three assumptions. As is described in Dalton et al (2011, p.8), these assumptions are that people possess informed political preferences and policy choices. Furthermore, they make judgements about which party best represents these preferences. Finally, these perceptions guide voting behavior. These assumptions are essential for meaningful election to place, given that the objective of the election is to match voters’ preferences with the party that represent them.

The ideological linkage can be divided into three steps, or functions rather, as laid out by Dalton et al. (2011, p.18) who map out the following steps. Firstly, there must be a coherent way to assess the preferences and choices of the voters so that parties can relate to something tangible when proposing policies. It is also required for the obvious reason that the voters need to know what it is they want out of the political process, as stipulated by the three aforementioned assumptions. This is what mainly concerns my thesis. The Left-Right scale is a good example of how this is done, hence my focus of that analytical tool. Secondly, the voters need to be able to relate to the parties they can choose from, in order to make an informed choice based on their preferences. The final function is that there is actually a matching going on between voters and parties.

As mentioned earlier, this study focuses on the Left-Right scale, which concerns the first function of the party linkage framework as I described. I agree with Dalton et al. (2011, pp. 82-83) notion that as long as we allow the left-right scale to mean different things to different people and is changeable over time, we can see the left-right scale as a summary of parties’ and voters’ views on contemporary issues. Furthermore, I also share their view that voters are able to relate to this scale.

(13)

About 90 percent of all voters across new and old democracies can place themselves, from their personal understanding of it, on the left-right scale (a.a., pp. 86-87).

However, this is not a subject where consensus reigns. Some scholars believe that voters do not have the required attributes to make informed choices. Additionally, other preferences that are not strictly political in nature such as a party leader’s or a candidate’s look, or habitual party loyalties, might influence the voters’ decisions. There are also researches that claim that what policies parties stand for is becoming less predictable and less consistent, which would harm voters’ ability to choose from policy options and could end up with policies they in fact do not support (a.a., pp. 82-83). A lengthier discussion of the left-right scale and its positive and negative aspects is elaborated on in the next subsection.

2.2 Theories About the Left-Right scale

The Left-Right scale has its origin in the genesis of the French Revolution. Initially, the privileged classes in French society, priests and nobles more precisely, were placed on the right side of the General Estates. ”Right” has also been a distinguished word historically, with a positive connotation, along with words such as ”justice”. The right side is also distinguished in the Christian faith. (Goldmann et al, 1997, p.87)

During the chaotic days and months of the French Revolution, members of the assembly of common mind and opinion started to form groups, named after their position in the room, which would be along the traditional Left-Right division of the General Estates. This division would become the metaphor of our modern Left-Right scale, where the right shares common political thought, broadly speaking, as the groups that formed alongside the right side of the General Estates, while the left represent those with similar opinions of those groups that formed along the left side of the General Estates. Of course, political thought and ideologies have changed since then, but the terminology Left-Right is still the most commonly used dimension to assess the character of party systems, individuals and parties (Goldmann et al, 1997, p.87). Even to this date, both chambers of the French parliament is seated according to their ideological Left-Right position (Hague & Harrop, p.21, 2010).

(14)

11

The pervasiveness of the Left-Right scale, since its inception, is a testimony to its strength. It could describe something essential to political realities, thus withstanding the test of time. According to Dalton et al. (2011, pp. 84-85), most citizens in democracies of today, new ones as well as old ones, can place themselves on a Left- Right Scale.

Figure 2.2: The Left-Right dimension

Far-Left Middle Far-Right

Communism Socialism Liberalism Conservatism Fascism

Source: figure based on Heywood (2000, pp. 27-28)

With the Left-Right scale’s history in mind, one might wonder what the Left- Right scale looks like? As can be observed in figure 2.2, the Left-Right scale has a simple design by its nature. It is a one-dimensional plane, where ideas and ideologies gravitate towards the left side, the middle or the right side. Political scientists have tried to explain exactly what the Left-Right scale encapsulates. Dichotomies such as stability and change, ownership and need, upper class and lower class have all been presented as explanations to what the Left-Right dimension stands for, but none of the answers have been sufficient enough in itself to explain this puzzle (Goldmann et al, 1997, p.87).

(15)

Figure 2.3: The horse-shoe bent Left-Right scale

Far-Left Anarchism Far-Right

Communism Fascism

Socialism Conservatism

Liberalism Source: figure based on Heywood (2000, pp. 27-28)

Other political scientists, such as Giovanni Sartori, claim that the Left-Right scale is a layman’s political index and a simplification of political realities. However, the scale is still prevalent in party and election research today (Goldmann et al, 1997, p.87). Furthermore, as is pointed out by Heywood (2000, pp. 27-28), the extremes on the Left- Right scale tend to converge in their authoritarian tendencies. In other words, the extreme left and the extreme right tend to share more similarities, rather than less similarities, than the traditional left and right. Additionally, ”anarchism” could be both ultra-left-wing and ultra-right-wing. For this reason, one could bend the two-dimensional Left-Right scale into a horse-shoe-like shape. As can be seen in figure 2.3, the Left- Right scale is bent into a horse-shoe shape where the extremes, such as communism and fascism, converge. Alternatively, an additional dimension of Authority-Liberty is added when analyzing parties and ideologies to capture the authoritarian and libertarian aspect. This two-dimensional space, consisting of the Left-Right dimension and Authority-Liberty dimension, is captured by figure 2.4. In the figure, one can observe that both Social Democracy/Center-Left and Communism lean heavily to the Left in the traditional Left-Right scale. However, there is a stark difference regarding their authoritarian leaning. While Social Democracy adhere to more liberal values, some forms of Communism relies on authoritarian method and values. This difference

(16)

13

Figure 2.4: Two-dimensional, Liberty-Authority scale

Authority Communism Center-Right Left Social Democracy and Center Left Right Libertarianism Liberty

Source: figure based on Heywood (2000, pp. 27-28)

is considerable but would not be captured by the Left-Right scale alone. Therefore, some scholars argue that using an additional dimension is crucial to capture a party’s or an ideology’s views. Another version of the Authority-Liberty dimension is the Green-Alternative-Libertarian and Traditional-Authoritarian-Nationalist dimension (GAL-TAN). Hooghe, Marks and Wilson (2002) makes a compelling case that this relatively new way of categorizing political thought elegantly encapsulates the development of the most recent decades, where they claim questions pertaining to “lifestyle, ecology, cultural diversity, nationalism, and immigration” have emerged (Hooghe, Marks and Wilson, 2002, p.976). Furthermore, these authors mean that GAL should be associated with values such as ecology, participatory democracy, cultural diversity and libertarian values, and TAN should be associated with traditional values, opposition to immigration, and the defense of the national community (ibid.). There are other dimensions as well that is not necessarily captured by the traditional Left-Right scale, such as feminism, ecologism and

(17)

animal rights. In some countries, the development of ”third way politics” has also made the Left-Right scale outdated, as the main parties converges in their ideas about economic distribution and workers’ rights (Heywood, 2000, pp.27-28).

Although, if we should make an attempt at mapping out the content of the Left-Right scale, it is worth mentioning Budge (2009, p.429), who presents a formidable list of themes that are common in party programs. To produce this list, he has looked at 50 post-world war democracies and their parties and how often words and themes are mentioned in respective party’s program. He also checked if the words and themes were written in a negative or positive context. The themes and words that are mentioned most frequently is presented in figure 2.5. The main conclusion that is drawn from the list by Budge, is that three broad topics emerge for each side of the scale. The right side focuses on economic freedom, law and order and a strong military. The left on the other hand, focuses on expanding the role of the government, protecting labor and expanding labor laws and fights for peaceful internationalism. As Budge notes (2009, p.429), these three broad themes are not linked by any stringent logic, but rather pieced together by the parties themselves. In fact, there are instances where center parties, certain Christian parties for example, combine these themes. Thus, I share a similar viewpoint of this study as Hague and Harrop (2010, p.22), that it is not unimaginable that these combinations and focuses will change in the future. Another important remark, that is also pointed out by Dalton et al. (2011, pp. 85-86) is that the meaning of Left and Right can change depending on who you ask, and this person’s context, such as what class the person belongs to, or which country the person lives in. To a blue-collar worker in Denmark, the Left might mean social welfare, while it means something completely different to a college student, like multiculturalism and internationalism. The meaning of the Right might mean traditional values to a German citizen, but economic freedom for a Swedish citizen.

Furthermore, according to Goldman et al. (1997, pp. 87-88), there are several theses concerning the Left-Right scale that are backed up by scientific research. Basically, they tell us that everyone has an understanding of where the parties in the respective party system are located on the Left-Right scale, in relation to each other. This understanding is born out of political socialization and passive and active participation in the political

(18)

15

Figure 2.5: The contents of the Left-Right scale

Source: Inspired by Budge (2009, p. 429)

.

discourse. The Left-Right scale becomes a simplified tool that can measure and relate parties, factual questions and individuals themselves. This enables the individual to relate their individual position to the positions of the parties, thus making it simpler to sort out where one’s sympathies lie. Usually, the citizens have a normal distribution, in statistical terms, but data show that the electorate can polarize if, for example, post material questions are more prominent in the public debate. This behavior is also true for politicians, whose strategies are influenced by the Left-Right scale. Finally, the theses declare that one should be careful with the usage of the Left-Right terminology, as it is ambiguous and vague. This fact makes it unruly for politicians and political scientists to use.

Left

Right

Peace Regulated capitalism Military:Positive Freedom;human rights Internationalism Democracy Free enterprise Law and order

Nationalization Protectionism Social harmony Constitutionalism: Positive Expansion of welfare Economic incentives Effective Authority Military:Negative Decolonization Traditional Morality: Positive

(19)

The ambiguity of the Left-Right Scale is also stressed by Dalton et al. (2011, pp. 85-86) as explained earlier, regarding different people, in different contexts, having different understanding of the meaning of the Left and the Right. However, they also point out that this ambiguity could be a strength of the model, rather than a weakness. If we allow the meaning of the scale to be flexible, we can see the Left-Right dimension as the issue most salient to the individual voter and how they relate the political parties to this specific, individual, scale. This would enable us to compare the effectiveness of the Left-Right scale as a communicating vessel between parties and citizens across democratic nations. The importance of the Left-Right scale as a communicating vessel is explained by the Party Linkage model above. However, as is also underscored by Dalton et al. (2011, pp. 85-86), should multidimensional issues, that is hard to capture with one dimension, dominate public discord, then the Left-Right scale will be quite a blunt tool to use.

(20)

17

3. Method, Data and Operationalization

To analyze the problems at hand, this thesis uses a mixture of quantitative and qualitative methods. Firstly, two models will be used, that both relate to the relationship between economic preferences and the respondents’ self-placement on the Left-Right scale. The data is extracted from four waves of the World Values Survey (WVS) and three waves of the European Values Study (EVS). The waves of the EVS was gathered 1990, 1999 and 2008 while t h e w a v e s of the WVS was collected within the time spans 1995-1998, 1999-2004, 2005-2009 and 2010-2014. Only old democracies have been chosen, to be studied, from the surveys. The surveys are structured in similar ways, so they are comparable to some degree. However, I have constructed one model for each survey, as the questions differ a bit. The models look as following:

(3.1) YSP= β0+β1Xei+β2Xob+β3Xrp+β4Xaj+β5Xct+E

(3.2) Ysp = β0+β1Xeiβ2Xob+β3Xrp+E

Where:

*Ysp = Self placement on Left-Right scale * β0 = Intercept

* β1-β5 = Coefficients of independent variables * E = Error term

* Xei = Preference to equalize incomes

* Xob = Preference of state vs. private ownership of business

* Xrp = Preference of state vs. individual responsibility to provide

* Xaj = Preference of the right to refuse job when unemployed

(21)

What the models basically explains is that people’s preferences on economic issues determines their self-placement on the Left-Right scale. Model 3.1 is applied on the EVS, while model 3.2 is applied on the WVS. The variables are measured by the respondents having to place themselves on a scale from 1-10, which can be seen in the table below. Intuitively, someone who places themselves on 8 on the Left-Right scale, which would mean quite heavily Right leaning, should also place themselves on the Right side of the independent variables. In table 3.1, economic preferences are operationalized as questions regarding the respondents’ view on ownership of businesses, responsibility to provide for ourselves and equalize incomes, and so forth. In this way, one can determine if they align with an economic agenda aligned with the Right side of the scale, or Left side of the scale. This answer is correlated with their answer on where on the Left-Right scale they deem themselves to be. As can be discerned from model 3.1 and 3.2, this thesis treats the survey questions, of which all have an answer range from 1-10, as an interval scale. That is why a regression has been used to analyze the data, as it requires an interval scale. This interpretation is of course controversial, as an interval scale assumes that the distance between every option is equally large, which can be a tough assumption to make when the answers are provided by humans, who interpret and analyze the questions differently, from person to person. However, as I am in agreement with Teorell and Svensson (2007, pp. 109-111) that if there are five to six options, or more, then one can with greater certainty brand the scale as an interval scale

It is also important to have a discussion of the reliability of survey data. Naturally, they do not always have the scale that is required for the specific quantitative analysis, and a lot of desired questions are missing, thus hard to replace after the survey has been carried out, and there is a discrepancy of what questions are asked, and which countries that participate, from year to year. However, it is the most reliable and accurate data out there for this specific thesis. For obvious reason, a there has been no time or resources to tailor a unique survey for this thesis, although it would have helped a great deal.

Finally, four cases have been selected from the quantitative section. The selection criteria are most likely and least likely cases. The first question will mainly be analyzed by the quantitative model, to get a general idea of how voters relate to the Left-Right scale throughout the democratized world. The two other questions will be analyzed with case studies, partly because appropriate data th at is required to answer the questions is lacking,

(22)

19

and partly because diving more deeply into the two latter questions could help increasing the validity of the results, at cost of generalizability.

Table 3.1: Operationalization of economic preferences

Question Type of variable Scale

Political view left-right Dependent 1-10

Equalize incomes vs. incentives for individual effort Independent 1-10 Private vs. government ownership business Independent 1-10 Individual vs. state

responsibility for providing

Independent 1-10

Take any job vs. right to refuse job when

unemployed

Independent 1-10

Competition good vs. harmful for people

Independent 1-10

Source: Data from World Values Survey (Inglehart et al., 2014) and European Values Study (1990; 1999; 2008)

(23)

4. Analysis

The analysis is structured into three main parts. The first part, 4.1, looks at the relationship between voters’ preferences on economic issues and their self-placement on the Left-Right scale. It deals with research question (1), which is the question of to what degree voters’ preferences about political issues can explain their Left-Right position. Furthermore, the selection of cases will be explained in this part, as the cases are based on the results of the first section. This part of the analysis is done through the quantitative model, whereas the two other parts uses case studies. The second part, 4.2, deals with the question of whether changes in national discourse change the relationship between voters’ preferences on a particular political issue and their self-placement on the Left-Right scale. Netherlands and France will be used as cases in this section, where both represent most-likely cases. This answers research question (2). The third and final part, 4.3, deals with the question of whether the Left-Right scale can facilitate in the matching of voters and parties of similar ideological conviction, which is research question (3). Sweden and Japan are used as cases to answer the question, where the latter is a least-likely case, and the former is a most-likely case.

4.1 Economic Preferences and the Left-Right Scale

In this section, I have studied the relationship between voters’ preferences regarding economic related issues and self-placement on the Left-Right scale. The data that has been utilized is extracted from the World value survey and the European value survey. A more detailed explanation of the questions that the variables are comprised of can be found in the method section, where the model is explained in greater detail as well. The purpose of this model is to see how well the respondents’ views and positions on economic related issues can predict where on the Left-Right scale the respondents’ self-placement is

The full result can be found in the tables in appendix 1. However, most of the results are presented in the graphs below. The x-axis of the graphs measure time while the Y-axis presents r-square of the model in percentage. A low percentage means that the model is a bad predictor of the respondents’ self-placement, while a high percentage means that the model can predict the self-placement quite well. For example, the model works well in Sweden, where the r-square is about 30%, while Japan has an insignificant score statistically speaking.

(24)

21

preferences on economic issues and their self-placement on the Left-Right scale is remarkably stable around 10%. Another picture is given by France, who drastically

changes between 1990 and 1999, then 1999 and 2008. Germany has similarly changed to a less strong relationship, while Austria slowly climb from almost insignificant numbers to an r-square of 10%. Over to figure 4.2, the Nordic countries, they show quite some stability as well, showing a strong relationship between economic preferences and the Left-Right scale, where Sweden has seemingly the strongest relationship and Iceland the relative weakest. In figure 4.3, Netherlands show just as strong relationship as the Nordic countries, but declines throughout the year. The rest of the countries show a similar relationship as mainland Europe, albeit Ireland stands out showing almost insignificant numbers. Finally, figure 4.4 presents data from the World Values Study, and suffers from the fact that few countries has a consistent participation. Therefore I have looked at trend lines only, as some countries only have data for two years. Basically, data shows that Sweden has a strong relationship, scoring around 35% in r-square, while Japan’s relationship is insignificant. Australia and Germany have quite stable relationship, while USA shows quite a rapid increase.

At this stage, one might ask what important points can be extracted from the data? The analysis of this thesis is that countries show remarkable stability throughout the years, but also the capability to change. Although this data is not enough to explain changes, it

0% 5% 10% 15% 20% 25% 1990 1999 2008 R-square in percentage Survey year

Mainland Europe

Austria France Germany Italy

Source: European Values Study (1990; 1999; 2008), Note 1: The circles represent data points, and are replaced with trend lines when data is missing

Figure 4.1: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in Mainland Europe

(25)

is still an important fact to acknowledge. The data does not seem to be random, or

sporadic. The Nordic countries show a clear trend where respondents’ preferences on economic issues show a significant relationship, over a period of about 20 years, while Japan and Ireland show an insignificant relationship during a similar time period. There must be some underlying factors that can explain these outcomes, and not simply a result

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 1990 1999 2008 R-square in percentage Survey year

Nordic Countries

Sweden Denmark Finland

Norway Iceland Linjär (Sweden) Linjär (Denmark) Linjär (Norway)

Figure 4.2: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in Nordic countries

e

Source: European Value Study (1990; 1999; 2008), Note 1: The circles represent data points, and are replaced with trend lines when data is missing, Note 2: “Linjär” means linear trend line

0% 5% 10% 15% 20% 25% 1990 1999 2008 R-square in percentage Survey year

British Isles and Benelux countries

Great Britain Ireland Belgium Luxembourg Netherlands Linjär (Belgium)

Figure 4.3: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in British Isles and Benelux countries

Source: European Values Study (1990; 1999; 2008), Note 1: The circles represent data points, and are replaced with trend lines when data is missing, Note 2: Luxembourg misses data from 1990, Note 3: “Linjär” means linear trend line

(26)

23

economic preferences can explain Left-Right self-placement. As has been described above, it varies depending on region, showing both significant and insignificant relationships, just as theory, outlined in the theory section, tells us.

To analyze the questions of this thesis further, I have chosen Japan and Sweden to study how a strong relationship between economic preferences and Left-Right self-placement can help with the ideological linkage, or the matching of parties and voters of similar ideological conviction in other words. Sweden represent the most likely case, as it has the strongest relationship, while Japan constitutes the least-likely case, seeing that the relationship is low. To analyze what factors might cause changes in the relationship, France and Netherlands will serve as case studies, where France offers to points in time where a drastic change has taken place, and Netherland gives one with a dramatic change. The hypothesis is that changes in national political discord, in the form of shifts in saliency of political issues, can explain why the relationship between economic preferences and Left-Right self-placement changes.

Figure 4.4: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in World Values Survey countries

Source: Data from World Values Survey (Inglehart et al., 2014), Note 1: The circles represent data points, and are replaced with trend lines when data is missing

(27)

4.2 France and Netherlands

This sections is devoted to the case studies of France and Netherlands, to deal with the question of whether changes in national political discourse changes the correlation between voters’ preferences on economic issues and self-placement on the Left-Right scale.

4.2.1 France

As elaborated on previously, France is an interesting case to study, as the correlation between economic issues and left-right position decreases between 1990 and 1999, just to rebound between 1999 and 2008. If the ideological linkage functions properly, we can expect economic issues to decrease in saliency in the year of 1999, and to have a greater importance in the years of 1990 and 2008. To get an answer, this thesis has looked at voters and their opinions of political issues over time. Data from years where a presidential election has been held, which are 1995, 2002 and 2007, have been gathered. Although the years are not optimal, considering they are not the same as when the European Values Survey was carried out, it is the second-best option, and provides a wealth of useful data.

The most obvious starting point in this puzzle is of course French voters, and how

20,30% 10,50% 21,30% 0% 5% 10% 15% 20% 25% 1990 1999 2008 R-square in percentage Survet year

France

France

Figure 4.5: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in France

(28)

25

(2012, pp.87-88), one can indeed observe a shift in the saliency of issues between the presidential elections, which follows the suggested pattern of other issues than economic ones being salient around 1999. In table 4.1, the issue ranking of three presidential elections, according to French voters,

Issue 1995 Issue ranking 2002 2007 Unemployment 1 1 1 Education 2 7 4 Social protection 3 Homelessness 4 Inflation 5 3 AIDS 6 Corruption 7 10 Personal security 8 2 7 Environment 9 6 6

(29)

Source: Inspired by Lewis-Beck, Nadeau and Bélanger, E. (2012, pp.87-88) . Note 1: This thesis added section about

the relationship between economic issues and other issues

are displayed. The most interesting aspect of this table, at least for this thesis, is to what degree French voters prioritize economic issues in relation to other issues. Even in this table, we can see the suggested pattern of economic issues being more dominant around 1999. Although economic issues are prevalent through all three elections, which is consistent with figure 4.6 that shows that at least about 10% of the variation in one’s Left-Right position can be explained by economic issues every survey year, the presidential election of 2002 has two issues in the top four that pertains to cultural issues rather than economic ones. These two are immigration and personal security. Both issues were ranked further back during the elections of 1995 and 2007, coming in at eight and eleven respectively in the former election, and seven and ten respectively in the latter election. Consequently, it makes sense if economic issues have a lesser impact on the political discourse, and thus what policies and ideas people associate with the left and the right.

The matter of how one determines if an issue is of economic or cultural nature is obviously of great importance when drawing conclusions. To address this, I have explained the reasoning in table 4.2. There an explain of why each issue has been coded

Work sharing Immigration France in the world Europe Inequality Retirement Taxes Terrorism Sovereignty Deficits Housing Suburbs 10 11 12 13 4 11 3 5 8 9 12 10 12 2 8 5 9 11 Total economic issues in top four

(30)

27

as it has, based on the theoretical background that is presented in this thesis, for each election year will be presented. There are two points that must be underscored, however. Firstly, one could arrive at a different codification should one refer to other theories about the nature of the Left-Right dimension. Secondly, the issues that are coded to the GAL-TAN dimension could justifiably be coded to the Left-Right dimension as well, seeing that cultural issues can be divided along that dimension as well. However, the point is to separate cultural and economic issues from one another, thus I need an alternative way to sort out cultural from economic issues, which is where the GAL-TAN dimension steps in. The Left-Right dimension remains the chief focus of this thesis, and GAL-TAN serves as an analytical tool rather.

Table 4.2: Coding of issues

Issue Linkage to theory Political

dimension

Education Expansion of welfare/Free

enterprise Left-Right Unemployment Regulated capitalism/Economic incentives Left-Right

Social Protection Expansion of

welfare/Economic incentives Left-Rigjt Homelessness Expansion of welfare/Economic incentives Left-Right

Personal security Defense of national

community

GAL-TAN

Inequality Regulated capitalism/Free

enterprise/Economic incentives Left-Right Immigration Inflation Cultural diversity/Immigration Regulated Capitalism GAL-TAN Left-Right

Source: Issues are extracted from the top four issues each of the presidential elections in table 4.1, while the theoretical linkage and the GAL-TAN and Left-Right dimension is based on the theory section in this thesis, Note 1: The coding of whether an issue is a traditional, economic, Left-Right dimension or GAL-TAN dimensions, is created by linking each issue to the theoretical framework of this thesis.

(31)

In summation, according to data presented in this thesis, changes in national political discourse could impact the relationship between voters’ economic preferences and their Left-Right self-placement. In other words, when voters prioritize immigration and crime for example, they changed their voting behavior in a manner that might not tally with their regular Left-Right identity, which could structurally change the substance of Left-Right scale to adapt to new political realities. This notion is shared with Lewis-Beck, M., Nadeau, R. and Bélanger, E. (2012, pp. 100-102) who remark that changes of saliency, from economic issues to crime and immigration, can explain the success of the far-right presidential candidate, Le Pen 2002, and then consequently the success of Sarkozy 2007 as voters realigned to traditional Left-Right issues again.

(32)

29

4.2.2 Netherlands

Netherland is another interesting case, as the result of 1990 and 1999 are quite similar, but with a drastic shift downwards occurring 2008. Thus, what we can expect to find is that similar issues should remain equally important between 1990 and 1999, while economic issues should decrease in saliency around 2008.

Naturally, a similar strategy has been employed to analyze the shift in Netherlands as the one used in the analysis of France. In other words, voters’ preferences and ranking of issues will be the core focus. Consequently, Similar problems have arrived when analyzing Netherlands, as when analyzing France. Thus, there is a wealth of data around election dates, but not so much around the year when the EVS was conducted, which my analysis in figure 4.6 is based on. This means that the analysis looks at the election years in closest proximity to the EVS, which are 1989, 1998 and 2006.

Taking election survey data at a glance, the prospect of issues driving what Left-Right is associated with looks bleak. Seeing that the correlation between people’s preferences of economic issues and Left-Right identification is highest 1990, one would expect economic issues to be dominant during the previous election year. However, the election of 1989 was dominated by environmental issues, as can be seen in table 4.3, that is more related to post-material values, that would rather be associated by the GAL-TAN

21,60% 19,30% 9,80% 0% 5% 10% 15% 20% 25% 1990 1999 2008 R-square in percentage Survey year

Netherlands

Netherlands Source: European Values Study (1990; 1999; 2008)

Figure 4.6: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in Netherlands

(33)

dimension. The reverse is true for the election of 2006, where healthcare and the economy top the list, while immigration comes in as number third.

Table 4.3: Top three most important issues in parliamentary elections

Election Year First Second Third 1989 Environment (42.3%) Unemployment (14.14%) Crime (3.59%) 1998 Crime (15.59%) Public health (11.59%) Refugees (11.44%) 2006 Healthcare (12.9%) Economy (9%) Immigration (6%)

Source: Data extracted from Dutch Parliamentary Study (Thomassen et al. 1989; Kamp et al. 1998; Kolk et al. 2006)

Naturally, looking at the top issues is only scraping the surface. One could find several explanations for this discrepancy, of which I will present three. Firstly, it could be that the EVS and the election surveys were conducted different years, and that the electorate fluctuated between those occasions. Secondly, this puzzle might also be more complicated than it ought to be, as voters might relate to the Left-Right scale through economic preferences during normal circumstances, but are prepared to deviate from this in favor of an issue that they deem more urgent, like the environment for instance. Finally, it might be the case that issues that quickly rise and descend in saliency do not divide the electorate along new dimensions, at least not initially. Both voters that position themselves as left and those that self-describe as right might agree that something should be done, and it is only after some time, if the issue prevail, that we see new divisions that changes what issues constitute that Left-Right dimension. For example, a trend of right-wing parties holding back on climate action might attract voters who agree with that sentiment, while the opposite would be true for left-wing parties if they instead support climate action. This in turn could make those who have a skeptical view of climate action, but are self-described as left, change their self-positioning to the right instead, even though their views of economic matters remain similar as before.

However, an answer could lie in the election of 2002 and its aftermath. During the parliamentary election of 2002, Pim Fortuyn brought up issues such as immigration and ethnicity to the top of the national political discourse. Although, he was assassinated shortly before the election, which garnered enough sympathy to propel his party to roughly 17%, that consequently brought his party and its anti-immigration agenda into a coalition government. As one might suspect by such chaotic events, the coalition government did not last, which led to two more irregular elections. The latter, which is

(34)

31

the election of 2006, did not center around immigration issues. However, there was a scandal concerning a member of parliament lying about her citizen status, which sparked the election in the first place, which continues the theme of immigration and ethnicity (Holsteyn, 2007).

In summation, just looking at the top issues each election in approximation to the EVS do not tally with theory. Where the correlation between voters’ economic preferences and Left-Right self-placement is high, the saliency of economic issues is low, and vice versa. However, when considering previous elections and how the national political discourse developed after 2002, the decline of the relationship between 1999 and 2008 makes sense. The conclusion that is drawn from this is that top issues during an election cycle does not necessarily influence the meaning of the Left-Right scale in the long run, but that developments such as what happened during the election of 2002 can have a long term impact.

(35)

4.3 Japan and Sweden

Figure 4.7: The relationship between respondents’ preferences regarding economic issues and self-placement on Left-Right scale, in Sweden and Japan

Source: Data from World Values Survey (Inglehart et al., 2014), Note 1: Circles represent data points, Note 2: “Linjär”

means linear trend line

The final question on the outlined agenda is if the Left-Right scale facilitates parties’ ability to attract voters of the same ideological conviction. As figure 4.7 explains, based on data from WVS, Sweden and Japan represent each side of the extreme. The figure also tells us that the trend is very stable in both cases. In Sweden it hovers about 35% of r-square in percentage, while Japan moves from a low level at 3.1% of s-square in percentage, to the statistical insignificant number of 0.2% in r-square in percentage. In other words, roughly 35% of one’s Left-Right position can be explained with one’s preferences on economic issues, in Sweden. As for Japan, almost a statistical insignificant portion of one’s Left-Right position can be explained by economic preferences. If the hypothesis of this thesis holds true, this means that the dominant left and right leaning parties in Sweden will have a greater success in garnering likeminded voters on economic issues, while the dominant left and right leaning parties in Japan will experience a greater difficulty in this task. One should note, however, that the Japanese parties might be better at gathering voters based on other kinds of dimensions, like something akin to GAL-TAN.

31,40% 39,50% 33,20% 3,10% 3,20% 1,50% 0,20% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 1995-1998 1999-2004 2005-2009 2010-2014 R-square in percentage Survey year

Correlation between economic preferences and Left-Right

self-description

(36)

33

To get an idea of how well parties have succeeded in attracting likeminded voters, this thesis looks at data from WVS. The two main contenders in each country are in focus: one for each side on the Left-Right scale. This choice is based on the fact that the two main parties have enough supporters to get interesting results statistically, because even when using the two main contenders the number of respondents are limited to a couple of hundred respondents, and also the fact that in order to get a proper answer, the parties in the analysis need to be large and serious enough to contend for seat at government and not narrow issue parties.

The two main parties in Sweden are the Social Democratic party and the Moderate party, where the Social Democratic party is the main left leaning party, and the Moderate party is the main right leaning party. Japan is a more complicated story, however, where different parties have been main contenders throughout the years. The exception is of course the Japanese Liberal Democratic Party (LDP), which has been in possession of government for most of the time in modern history. LDP represent the center right in Japan. The center left candidate is harder to figure out, as was suggested above. The fact is that, relatively speaking, the opposition has always been in disarray. For the first survey period 1995-1998, both the New Frontier Party (NFP) and The Democratic Party of Japan (DPJ) performed well, although the NFP performed much better during the election in the 1995-1998 survey period, whereas DPJ would go on and be dominant the following elections as NFP came to dissolve (Hayes, 2009, p.311). There is also the issue of if the party qualifies as a center left party. DPJ qualifies for this title, which I will elaborate on further down, but the story for NFP is much more complicated, as it was a coalition of various oppositional parties with different interests (a.a, p.81). However, seeing DPJ was not an alternative during the 1995-1998 survey period, I have reluctantly chosen NFP as the second-best option, and the DPJ for the remainder of the analysis.

(37)

Table 4.4: The mean score of party supporters regarding economic issues and Left-Right self-placement, in Japan Survey Year: Question 1995-1998 1999-2004 2005-2009 2010-2014 LDP LDP LDP LDP Left-Right selfplacaement 7,11 (R) 6,92 (R) 6,35 (R) 6,66 (R) Owenership of Business 4,64 (R) 4,53 (R) 4,65 (R) 4,56 (R) Competition is good/harmful 4,45 (R) 3,86 (R) 4,27 (R) 4,03 (R) Income equality 5,88 (R) 6,14 (R) 6,08 (R) 5,38 (L) Responsibility to provide 4,45 (L) 4,90 (L) 4,32 (L) 4,04 (L) NFP DPJ DPJ DPJ Left-Right selfplacement 5,96 (R) 5,31 (L) 5,01 (L) 5,20 (L) Ownership of business 4,81 (R) 4,28 (R) 4,48 (R) 4,44 (R) Competition is good/harmful 4,26 (R) 3,93 (R) 4,13 (R) 4,11 (R) Income equality 5,73 (R) 5,81 (R) 6,33 (R) 5,18 (L) Responsibility to provide 3,83 (L) 4,37 (L) 4,17 (L) 3,79 (L)

Source: Data from World Values Survey (Inglehart et al., 2014)

Table 4.5: The mean score of party supporters regarding economic issues and Left-Right self-placement

Survey Year:

Question

1995-1998 1999-2004 2005-2009 2010-2014

SAP SAP SAP SAP

Left-Right self-placement 4,12 (L) 4,10 (L) 4,10 (L) 3,83 (L) Owenership of Business 4,95 (R) * 5,77 (L) 5,74 (L) Competition is good/harmful 3,66 (R) 3,65 (R) 3,74 (R) 3,84 (R) Income equality 5,26 (L) * 5,47 (L) 4,28 (L) Responsibility to provide 4,45 (L) 4,67 (L) 5,79 (R) 4,67 (L) M M M M Left-Right self-placement 7,95 (R) 7,72 (R) 7,76 (R) 7,77 (R) Ownership of business 3,17 (R) * 3,36 (R) 4,18 (R) Competition is good/harmful 2,40 (R) 2,54 (R) 2,67(R) 2,97 (R) Income equality 7,59 (R) * 7,67 (R) 6,31 (R) Responsibility to provide 8,26 (R) 3 (L) 7,74 (R) 6,99 (R)

(38)

35

The information given by table 4.4 and 4.5 can be quite confusing at first glance, which is why I will explain it briefly. The design of the questions is outlined in the method section, but in summary a respondent can choose an answer on a scale from 1-10. In the question regarding self-placement on Left-Right scale, 1 represent the left-most option, while 10 represents the right-most answer. When looking at averages, 5.5 represents the center, while everything below is often left-leaning and everything above is often right-leaning. However, in some questions, the Left-Right order is reversed. For example, the question of whether competition is good or harmful, 1 is the right-most answer instead. The number that is presented is the average score of each party’s supporters. After each number, there is a letter that indicates whether the average leans to the left or right, to create some clarity seeing that some questions have inverted Left-Right scale.

The tables paint quite a remarkable picture, where the difference in the result could not be more profound between the two countries. In Sweden, there is a clear distinction between what M and SAP supporters desire regarding economic preferences. M supporters vehemently place themselves on the right wing of the scale, on average, and mimics this among economic preferences. Translating this to numbers, one can see that in the year 1995-1998, M supporters scored 7,95 in their Left-Right placement, and 7,59 on average regarding income inequality. The two averages are in very close proximity. Naturally, the same is true for the questions with an inverted Left-Right scale, where a score of 7,95 on the Left-Right scale tallies with a score of 2,40 in the question about whether competition is good or harmful. However, it is important to know that there is an outlier that I cannot explain. The outlier is found in the survey year 1999-2004, where M supporters have quite a heavily leaning left score of 3, while the answers given the other years ranges from 6,99-8,26. The outlier is the only left leaning average in all of the survey years, and of all the questions, so I have chosen to not put much significance to it at this stage.

As of SAP supporters, they place themselves in the center, slightly to the left, just as they do with economic preferences. Their self-described Left-Right position average varies from 3,83 to 4,12, while their preference regarding equality goes from 4,28 to 5,26. The supporters are a bit more right-leaning regarding preferences on economic issues than what they score themselves on a Left-Right scale, but in total, the averages still leans to the left in most questions. Although, one important exception is that they consistently lean to the right regarding the issue of competition. In other words, their supporters’

References

Related documents

46 Konkreta exempel skulle kunna vara främjandeinsatser för affärsänglar/affärsängelnätverk, skapa arenor där aktörer från utbuds- och efterfrågesidan kan mötas eller

The increasing availability of data and attention to services has increased the understanding of the contribution of services to innovation and productivity in

Närmare 90 procent av de statliga medlen (intäkter och utgifter) för näringslivets klimatomställning går till generella styrmedel, det vill säga styrmedel som påverkar

• Utbildningsnivåerna i Sveriges FA-regioner varierar kraftigt. I Stockholm har 46 procent av de sysselsatta eftergymnasial utbildning, medan samma andel i Dorotea endast

Den förbättrade tillgängligheten berör framför allt boende i områden med en mycket hög eller hög tillgänglighet till tätorter, men även antalet personer med längre än

På många små orter i gles- och landsbygder, där varken några nya apotek eller försälj- ningsställen för receptfria läkemedel har tillkommit, är nätet av

Det har inte varit möjligt att skapa en tydlig överblick över hur FoI-verksamheten på Energimyndigheten bidrar till målet, det vill säga hur målen påverkar resursprioriteringar

Industrial Emissions Directive, supplemented by horizontal legislation (e.g., Framework Directives on Waste and Water, Emissions Trading System, etc) and guidance on operating