Beating Social Democracy on its own turf: Issue convergence as winning formula for the centre-right in universal welfare states



Yüklə 66,78 Kb.
tarix09.02.2017
ölçüsü66,78 Kb.

Beating Social Democracy on its own turf: Issue convergence as winning formula for the centre-right in universal welfare states







Christoph Arndt, PhD

PostDoc, Adjunkt

Institut for Statskundskab

Aarhus Universitet

Bartholins Allé 7

8000 Aarhus C

Arndt@ps.au.dk

Paper to be presented at the Annual Meeting of the Danish Political Science Association 2012, October 25-26, Hotel Vejlefjord. Work in progress, please do not cite with the author’s permission.



  1. Introduction

Electoral change in Denmark and Sweden since the turn of the millennium has yielded a novel political landscape since the social democratic parties lost their traditionally dominant position whereas the secular centre-right now competes at eye-level. The Danish Liberals (Venstre) have become the largest party since 2001 despite the fact that the party has governed the country for more than one decade in a (legislative) coalition with the Conservatives and the Danish People’s Party. In contrast, the Danish Social Democracy has lost votes in each election since 1998. Not only from the issue voting perspective, but also from the traditional welfare state and class voting literature provides this pattern an anomaly as social democracy ought to have a clear advantage in large welfare states given its traditional issue ownership over welfare issues (Budge & Farlie 1983; Petrocik 1996; but see Blomqvist & Green-Pedersen 2004). Moreover, as welfare has increasingly turned into a valence issue and larger parts of the population are protected through the welfare state, welfare issues can be expected to have “natural” salience which ought to be beneficial for social democrats (Budge & Farlie 1983; Petrocik 1996; Aardal & van Wijnen 2005; Giger 2011).

How it is then possible that centre-right majorities can be formed and last for more than one decade in the Danish case on a hostile terrain from the perspective of issue-based explanations of political behaviour? This puzzle has not been tackled systematically in the newer English speaking literature on electoral behaviour and issue voting in Scandinavia and there is also limited research in the native languages (Goul Andersen 2003; van der Brugge & Voss 2007; Mortensen 2008; Oscarsson & Holmberg 2008).

Recent analysis of issue-based vote choice has – with few exceptions – typically focused on countries with majoritarian electoral systems and liberal welfare states (Petrocik et al. 2003; Bélanger & Meguid 2008; Green & Hobolt 2008; Clarke et al. 2009; see van der Brug 2004 or Aardal & van Wijnen 2005 for exceptions). Here, valence competition on economic issues has become the dominant mode of issue competition and performance voting the predominant model of issue voting. Contextual factors at the macro-level such as the welfare state’s importance and its effect on the positioning of parties under issue competition have rarely been incorporated in analysis of issue-based vote choice (see Green & Hobolt 2008 for an exception).

Incorporating the macro-perspective (welfare state importance and party positions) in a model of issue-based vote choice, this paper seeks to analyse issue voting in Denmark in the period (1998-2007) in which a pure centre-right majority won three consecutive elections after 1998. I argue that the repositioning of the Liberal Party has changed the mode of issue voting in Denmark. The party’s moderation allowed outperforming the social democrats on traditional welfare issues and winning voters on those issues for the Liberals and the centre-right bloc in general.

The paper proceeds as follows. The next section briefly reviews the literature and develops the argument on the different logic of issue competition and voting in universal welfare states. The third section describes the data and operationalisation. This is followed by the results. The paper final section discusses the findings from a theoretical and comparative perspective.


  1. Issue ownership, issue voting and the welfare state context

    1. Issue ownership and issue-based vote choice

Not least since Budge & Farlie’s (1983) seminal work, the scholarship of party competition has turned towards issue reputation, issue ownership and issue salience as main explanations of political competition and political behaviour in electorates that vote less on the basis of sociological determinants (Petrocik 1996; Green-Pedersen 2007; Bélanger & Meguid 2008; Clarke et al. 2009). According to this perspective, parties own particular issues and voters can to a sufficient degree associate issues and issue competence with parties. Classical examples for issue ownership are the reputation of centre-right parties in handling issues such as crime, defence, or taxes, whereas centre-left parties own issues such as health care, unemployment, and the welfare state (Budge & Farlie 1983; Petrocik 1996). Parties have therefore a strong incentive to get their issues on the agenda and to circumvent those issues where they have a negative reputation on or opponents are too competitive. Since Budge & Farlie’s and Petrocik’s work on the logic of issue competition, scholars of party competition and agenda-setting have elaborated upon and developed further arguments on the interplay between issue ownership, issue salience, the parties’ issue-positioning, and issue-based political behaviour (e.g. Green 2007; Bélanger & Meguid 2008; Green & Hobolt 2008; Stubager & Slothuus 2012). Other scholars also discussed conditions under which simple accounts of issue ownership do not hold, e.g. a popular candidate stealing the opponent’s issue (Holian 2004) or reframing a non-owned issue (Damore 2005).

Recently, the study of issue-based vote choice on the individual level has been expanded by issue salience as moderator (Bélanger & Meguid 2008; Green & Hobolt 2008). Parties or candidates can most effectively draw on their reputation on and ownership of issues that are salient during an electoral contest. Accordingly, issue ownership alone is not sufficient for voters to discriminate between political alternatives. Other recent scholarship has shown that in electoral contexts, where the main ideological competitors have converged, the valence model of issue competition has become more important in explaining party choice than ideology (Green 2007; Green & Hobolt 2008; Clarke et al. 2009). Labour is in Green & Hobolt’s example the traditional owner of the National Health Service (NHS) and benefits from salience of this issue in a context of ideological depolarization of Labour and the Tories since 1992. Apart from Green & Hobolt’s analysis of the British case though, the literature has not investigated the interplay between the parties’ issue positions on the macro level and issue-based vote choice on the micro-level and the problem whether and to which extent parties benefit from shifting towards the main opponents issue position. Furthermore, Damore (2005) argued that U.S. presidential candidates converged on issues which have a high salience among the electorate irrespective of the issue owning party.

These arguments imply that first the salience of issues is an important moderator for the parties’ ability to persuade voters on the basis of their issue ownership, and second that ideological convergence affects the voters’ reasoning in that way as they vote parties due to their perceived competence but not ideological position. Nevertheless an open question remains: What does happen if a party converges on issues which it does not own and which are often salient? Moreover, will such a move beneficial in terms of vote gains against the traditional issue owner? This puzzle is the point of departure for the argument in the next section.


    1. Issue ownership, issue voting and the welfare state context

What typically was lacking is to account for other structural factors on the macro-level shaping the logic of party positioning on issues in a first step and issue voting in a next step. Even most recent research on issue-based vote choices has used countries from the liberal welfare state regime (Esping-Andersen 1990). These countries, however, have typically majoritarian electoral systems with two party or two plus party systems, where the arguments from the early issue voting literature have its origin and seem to fit best. Here, party competition is between a pro-welfare centre-left and a welfare-sceptical centre-right with the former being keen to have welfare issues on the agenda and the latter being keen to circumvent those issues and talk about lower taxes or crime to increase the parties’ electoral prospects, respectively.1 Beyond liberal welfare states and majoritarian electoral systems, however, the game is different. This goes even more so in universal welfare states, where some welfare schemes such as health are not solely position issues but increasingly also valence issues (Goul Andersen 2003; van der Brugge & Voss 2007: 134f; see Green & Hobolt 2008 for the NHS as universal scheme in the British case). The reason is that every voter is automatically enrolled in the sickness insurance and pension system and thus a stake-holder interested in maintaining the service (Goul Andersen 2003; Giger 2011). This also implies that middle class voters are protected by the welfare state and not by market-based arrangements as common in the liberal regime, which most studies of issue voting use a universe for their case selection.2

This is especially true under conditions of issue competition. Voter groups are no longer tied to bourgeois parties beforehand due to other cleavages such as religion or and an urban-rural divide outbalancing the effect of class and, accordingly, the effects of welfare issues. Ideological pureness is therefore no longer feasible for the centre-right (or the major party of the centre-right) if the goal is to seek broader electoral support and to regain and keep office after long(er) opposition spells.3 Here, a different logic is at work for the centre-right in mature universal welfare states compared to arguments from traditional issue voting literature where the centre-right benefits from setting an agenda around smaller public sectors and lower taxes.

Hence welfare issues have a sort of natural salience in mature and encompassing welfare states (Aardal & van Wijnen 2005; Giger 2011) and may therefore be difficult to circumvent for centre-right parties. The centre-right parties have a structural disadvantage given the high saliency level of non-owned issues and cannot hope – as Bélanger & Meguid’s (2008) model suggests – that their own issues will dominate the agenda and offset the salience of non-owned issues. This latter option is not a realistic option in the long run.

Moreover, there is a considerable risk that the salience of welfare issues increase if cutbacks and reforms of welfare benefits are announced or proposed by centre-right parties since social democrats would use this against the centre-right. These features provide first an open flank for the centre-right in universal welfare states and favourable context for left having ‘its’ issues almost automatically on the agenda or at least a non-negligible level of salience at almost each election.

In sum, universal welfare states exhibit a special context for secular right-wing parties under conditions of issue competition and we would typically assume social democrats to have a competitive advantage given the ‘natural’ salience of their issues. From the perspective of issue ownership-based vote choice, we would also not expect that social democrats will lose those issues on their own turf and be outperformed by secular and traditionally market-oriented centre-right parties. In other words, centre-right parties have to do something to beat social democracy on its own turf and to cope with its structural disadvantage under conditions of issue competition.


    1. Bourgeois parties’ issue ownership strategies

If we accept the premise that welfare issues have a high natural salience in universal welfare states, it follows that the arguments and insights from Bélanger & Meguid (2008) and Green & Hobolt (2008) need to be refined. To compensate for their structural disadvantage in encompassing welfare states, the centre-right may at some point decide to compete with its social democratic opponent on some welfare schemes to limit the social democrats’ vigour on issues such as health or pensions. However, this does not only imply that the centre-right signals an increased competence on welfare issues in general, but also to converge on welfare issues to show that the welfare state is in save hands even under bourgeois reign. Convergence also means that some – originally positional –welfare issues have turned into predominantly valence issues (Green & Hobolt 2008). This is in line with Damore’s (2005) argument that presidential candidates converge on issues which have a non-negligible salience among the voters regardless of the original issue owner. A similar logic can be applied for vote-seeking centre-right parties in electoral contexts where the largest opponent owns high salience issues. In office, the centre-right also needs to show that the convergence to more welfarist positions was not just a campaign manoeuvre, but that there has been a sufficient policy change of the party (see Goul Andersen 2003: 168f for a brief discussion). Consequently, issue ownership or issue competence should work best among those voters who believe that the centre-right has converged to the social democrats on welfare issues. Issue ownership effects of health or pensions are expected to work most effectively among voter perceiving a small or even no ideological distance between the centre-right and its social democrats challenger. In other words, the centre-right performs best if has effectively signalled the voters that the welfare state or at least major universal schemes are in save hands even under bourgeois reign.

Another virtue of converging on welfare issues is that it opens the possibility to compete on other non-welfare issues such as crime or immigration on which the centre-right has a comparative advantage [to be included in analysis in later step if feasible]. This allows coming closer to the attitudes of working class constituencies on non-economic issues. These may gain more relative importance for vote decisions if the centre-right is no longer regarded as a dangerous anti-welfare opponent and allows gaining votes on issues of crime or immigration which are still positional issues.



    1. The Danish context

The case of Denmark allows analysing the effects of ideological convergence on issue voting. Denmark is a universal welfare state with broad coverage of the population in welfare schemes such as the People’s Pension. The party system consists of a social democratic party that until recently has been the largest party since the 1924 election and the traditional issue owner over welfare, pensions, and health (Goul Andersen 2003) and a secular market-liberal party Venstre. The Liberals have despite its market-liberal legacy become the largest party with the 2001 election. This was often attributed to the party’s moderation on welfare issues and on the left-right dimension in general (see Mortensen 2008), but rarely examined from the perspective of issue voting and issue convergence as argued above (see Goul Andersen 2003; van der Brugge & Voss 2007 for native language approaches). Moreover, the PR electoral system means that there is a real multiparty competition throughout the country and not only two party contests in the whole country as in the USA or multiparty systems with a two party competition in most regions or constituencies (as inCanada or the UK).

Accordingly, this case resembles the anomaly where a centre-right party has outperformed its social democratic opponent on hostile terrain; that is where social democrats are the traditional owner of the issues with the highest salience. Moreover, this is not only a one-shot development since the Liberals have remained the largest party in four consecutive elections since the turn of the millennium making the country a illustrative and critical case to study the effect of issue convergence on issue voting.



  1. Data and Methods

    1. Data and Variables

To analyse the effect of ideological convergence on issue voting, I used a pooled data set covering the Danish Election Surveys 1998, 2001, 2005, and 2007.4 Unfortunately, it is only possible to use the period 1998-2007 as items on issue ownership and salience had only been included since the 1990 study and item batteries with similar wordings for the most important issues were not included before 1998. Admittedly, a longer time span would be desirable to substantiate the findings over a longer period. On the other hand, this period allows testing the arguments on issue convergence as the Liberal Party adapted a new policy platform after the 1998 election and outperformed the social democrats in the elections thereafter (Goul Andersen 2003; van der Brugge & Voss 2007). Another virtue of the data and period used is that the recent Danish elections studies are – together with the Canadian studies used by Bélanger & Meguid (2008) – one of the few election studies that have included item batteries on both issue salience and issue ownership.

The dependent variable is vote choice in the elections 1998-2007 where the same eight parties and non-voting were distinguished.5 However, given the problem whether issue convergence affected issue voting and changed the nature of issue competition between the Liberals and the Social Democrats, I mainly focus on these two parties, but also look at the second largest centre-right party, the Danish People’s Party given its centrist stance on welfare. Another reason to focus on the two largest parties is that they form the main alternatives for government formation since they are the leading parties of the two bloc (centre-right vs. centre-left) around which party competition in Denmark is structured.

The independent variables of the main theoretical relevance are issue salience, issue ownership, and the distance between the Liberals and the Social Democrats on the left-right continuum. The issues examined are the five most salient issues in the period under review: “economy”, “labour market/unemployment”, “health”, “pensions”, and “immigration” with all other issues as reference category (see Table 1).

Issue salience in Danish election studies is measured by the item: “We have currently a parliamentary election, and therefore I would like to ask you about the most important problem politicians need to deal with” [Vi har jo lige haft folketingsvalg, og derfor vil jeg gerne spørge dig, hvilke problemer du mener, er de vigtigste, som politikerne skulle tage sig af i dag?]. Since respondents were asked about the most important issue, I constructed dummy variables for a each issue distinguishing between 0 ‘is not the most important’ and 1 ‘issue is most important’. This is the common measurement of issue salience in Danish election study and close to the wording used in British election studies (Green & Bobolt 2008) even though it does not allow replicate the practice of Bélanger & Meguid (2008) employing a trichotomous measure of issue salience.



Table 1: Most important issues in the 1998-2007 Danish elections [some figures are preliminary calculations]




1998

2001

2005

2007

Economy

10

4

5

13

Labour market/unemployment

7

3

10

3

Health

15

17

8

10

Pensions

13

17

13

9

Immigration

15

22

17

18

Source: van der Brugge & Voss (2007); own calculations based on 2007 Election Study.

Issue ownership is measured as the government alternative that is best at solving a given problem, for instance the economy. Given the bloc nature of the Danish party system, the Danish election studies yield three alternatives ‘a social democratic-led government (coded -1), no difference (0), and a bourgeois [that means a Liberal-led] government (1). [Jeg vil nu læse en række problemer op for dig, og jeg vil gerne høre, hvem du mener, er bedst til at løse problemet; en socialdemokratisk ledet regering, eller en borgerlig regering?] In contrast to Bélanger & Meguid’s specification, it neither possible nor feasible to recode issue (non-)ownership for each party for each issue as the item clearly discriminates between the alternatives centre-left and centre-right government plus a category for respondents unwilling to identify an alternative. The Danish constellation and issue ownership item marks also difference to their Canadian case with a tradition for single party governments rather than coalition governments that are either led by the main centre-left or centre-right party. Methodically it is also not necessary since I employ a multinomial logit rather than a binary specification as in their example.

The distance between the Liberals and Social Democrats was calculated as the absolute difference between the perceived positions of the Liberals and the Social Democrats on the left-right scale (0-10) using the item battery: ‘where would you place the individual parties on the left-right scale shown below’ [I politik taler man ofte om venstre og højre. Nedenfor er vist en skala? Hvor vil du placere dig selv og de enkelte partier på denne skala?]. This variable measures the convergence effect where higher values on the distance scale indicate a larger perceived ideological difference between the two main rivals for government formation, whereas a value of ‘0’ means no perceived ideological difference.


    1. Method

I conducted a multinomial logistic regression model with vote choice as dependent variable. However, I focus mainly on the contrast between Liberals and Social Democrats (the reference category) and will look at the role of the Danish People’s Party as argued above. Following the setup of Bélanger & Meguid (2008) and Green & Hobolt (2008) and the arguments about ideological distance made above, the main independent variables are issue salience, issue ownership and the absolute distance Liberals-Social Democrats. Moreover, interactions between issue salience and ownership are included to capture the contingent effect of issue salience on issue ownership; that is issue ownership are expected to be strongest if the issue is also salient.

I also included class and labour market status, gender, age, union membership, place of residence, party identification, and survey year as control variables. These controls reveal the standard demographic controls of electoral behaviour in Scandinavia and also rule out that the effects of issue convergence and ownership on party choice are mainly driven by partisanship.



  1. Results

Table 2 shows the results of the multinomial logistic regression analysis for the contrast Venstre and Social Democrats as well as the contrast Danish People’s Party (second largest centre-right party in the period) and Social Democrats. As regards the interpretation of the coefficients for issue salience and ownership and the presence of their interaction one has to be aware of that the individual coefficients are conditional upon the values of other variables (Brambor et al. 2006; Bélanger & Meguid 2008: 484f). To give an example, the coefficient for the Liberal and issue salience of the economy is only 0.65 if issue ownership is zero and the other variables are held at their reference category or minimum value. We should also consider that all coefficients capture the effects when the perceived distance between the Liberals and the Social Democrats is zero.

Table 2: Logistic Regression Model of party choice in Danish National Election, 1998-2007




Venstre

DF




B

SE

B

SE

Constant

0.62

0.38

-1.15*

0.46

1998 (Reference)













2001

0.16

0.21

-0.12

0.25

2005

0.36!

0.19

0.46*

0.22

2007

0.40!

0.21

0.01

0.26
















Ideological distance S-Venstre

-0.11***

0.03

-0.02

0.04

Party ID Venstre

6.22***

1.02

2.32*

1.08

Party ID DF

-39.1***

0.89

2.10!

1.26

Issue Salience (Reference: all other issues named)

Economy

0.30

0.34

0.26

0.41

Labour Market/Unem-ployment

0.65*

0.29

0.91**

0.33

Pensions

0.12

0.24

0.52!

0.27

Health

0.75***

0.22

0.40

0.29

Immigration

0.03

0.21

1.33***

0.29

Issue Ownership (Reference: all other issues named)

Economy

1.57***

0.11

1.09***

0.12

Labour Market/Unem-ployment

0.83***

0.09

0.83***

0.11

Pensions

0.62***

0.11

0.89***

0.12

Health

1.10***

0.10

0.66***

0.12

Immigration

0.93***

0.09

1.77***

0.18

Interactions

IS X IO: Economy

0.61

0.39

-0.16

0.46

IS X IO: Labour Market

1.20***

0.33

0.70!

0.36

IS X IO: Pensions

0.53!

0.29

0.34

0.30

IS X IO: Health

0.06

0.25

0.34

0.31

IS X IO: Immigration

0.32

0.22

-0.08

0.30




McFadden R2

0.36

Nagelkerke R2

0.75

-2LL

-6687.65

N

5690

Notes: Entries are logit coefficients from a model with robust standard errors. Model also consists of demographic control variables [not shown]. ! p≤=.10, * p≤=.05, ** p≤=.01, *** p≤=.001. IO: Issue ownership, IS: Issue salience.

Issue salience alone is not always a significant factor for the likelihood to vote the Liberals (or the DF) in contrast to vote the social democrats as only half of the 10 coefficients reach significance at the conventional 5 per cent level.6 In contrast, all coefficients for issue ownership are significant for both parties. The strongest effects are economy and health for the Liberals and economy and immigration for the Danish People’s Party.

While the stark coefficients for economy for both centre-right parties and immigration for the Danish People’s Party constitute non-surprising effects, the strong effect of issue ownership over health for the Liberal Party constitutes a somewhat surprising effect. In turn, it may indicate that the efforts of the Liberals to show competence and stripping off the market-liberal image on that issue were a successful strategy in the period under review. This is especially true when we keep in mind that this effect reveals issue ownership for respondents perceiving no (longer an) ideological distance between the Liberals and the Social Democrats.

This is also corroborated by the sign and the significance of the distance measure. The negative sign and the high significance (p≤ 0.001) reveals that the likelihood to prefer the Liberals to the Social Democrats decreases with increasing perceived distance between the parties.7 This is not the case for the Danish People’s Party, where the coefficient is weaker and far from statistical significance. The Liberals obviously benefit from having moved towards the centre since 1998 (at least on some issues). There is also a larger share of respondents who believe a smaller ideological distance between Venstre and the Social Democrats at least in the 2001 and 2005 elections compared to 1998 (not shown).



The next step was to calculate predicted probabilities based on the model shown in Table 2 to see whether the ideological convergence has affected to likelihood to vote the Liberals on the issues of health and pensions, the most universal schemes, and the two schemes where the Liberals and also the DF moderated themselves markedly. Table 3 yields the predicted probabilities for the effects of issue ownership over health on party choice given the issue is salient. The figures yield the probabilities across the perceived ideological distance keeping the other variables at their means or typical values. A value of zero indicates no ideological distance between the two major parties and 10 indicates the largest perceived distance.

Table 3: Probabilities of vote choice contingent on ideological distance between 2 major parties for health issue




0

1

2

3

4

5

6

7

8

9

10

Social Democrats

0.12

0.13

0.14

0.15

0.16

0.17

0.18

0.19

0.19

0.20

0.21

Venstre

0.36

0.34

0.33

0.31

0.30

0.28

0.27

0.26

0.24

0.23

0.22

DF

0.09

0.09

0.10

0.10

0.11

0.11

0.11

0.12

0.12

0.13

0.13

Probabilities of party choice in the 2001 election calculated for a 50 year old male skilled worker who is not union member, residing in a medium-sized to large town (50.001-500.000 inhabitants), when health is the most salient issue and a Liberal government the issue owner.

Beginning with health, we can observe that the Liberals translate issue ownership over health most effectively into vote shares among voters perceiving no or only a small distance to the social democrats. Here, Venstre clearly outperforms the social democrats. If we move to larger perceived distances between the rivals for government formation, this issue ownership effect vanishes and the Liberals do no longer outperform the social democrats even though they are regarded as the issue owner. This indicates that Venstre’s ambition to signal competence on welfare, in this case health, has only turned into an efficient strategy among voters who regard the centrist move as credible and effective. There is also a notable effect for the Danish People’s Party as the party gains vote shares with increasing ideological distance between the Liberals and the Social Democrats in case the Liberal Party respectively a Liberal-led government is regarded as issue owner.



Table 4: Probabilities of vote choice contingent on ideological distance between 2 major parties for pension issue




0

1

2

3

4

5

6

7

8

9

10

Social Democrats

0.18

0.19

0.20

0.21

0.22

0.23

0.23

0.24

0.25

0.26

0.27

Venstre

0.17

0.16

0.15

0.14

0.14

0.13

0.12

0.11

0.10

0.10

0.09

DF

0.18

0.19

0.20

0.20

0.21

0.22

0.22

0.23

0.23

0.24

0.24

Probabilities of party choice in the 2001 election calculated for a 50 year old male skilled worker who is not union member, residing in a medium large town (50.001-500.000), when pensions is the most salient issue and a Liberal government the issue owner.

Similar albeit weaker patterns occur for the pension issue (Table 4). As for the issue of health, Venstre’s issue ownership works most effectively for respondents perceiving no or only a minor ideological distance between the two major parties.8 As the distance increases, the likelihood to prefer the social democrats increases while it decreases for the Liberals even though the latter is the issue owner. Venstre is at level with the social democrats in case there is no perceived ideological distance, whereas the shares for the Social Democrats triple the Liberals’ share if the ideological distance is at maximum. Again, the Danish People’s Party benefits from an increased distance between the main opponents and capitalises on the issue ownership of the centre-right over the pension issue and is at level with the social democratic party, traditionally the owner of pension issue.

The effects found for the DF may indicate a watchdog function for the party among voters that preferred a centre-right government, but did not believe that the Liberal’s centrist turn was authentic (enough) and that there still was a remarkable difference between Venstre and the Social Democrats in protecting the welfare state. These voters may have voted for the DF to save the welfare state against too market-liberal positions of a right-wing government. This corresponds to arguments from the literature on directional voting, where some voters are expect to discount parties’ positions in campaigns and potential (legislative) coalitions (Adams et al. 2005).

In sum, as more voters felt a smaller ideological distance between and the main rivals for government formation, the Liberal Party was successful in translating its more welfare-friendly stance into votes. Venstre has become more competitive on some welfare issues since thanks to the party’s credible convergence and thus improved image. This helps us to understand how the party has managed to become the largest party under conditions of issue voting on a hostile terrain. Another supportive feature was the ability of the Danish People’s Party to attract voters that regarded a centre-right government as better alternative in handling welfare issues, but (still) regarded the Liberals as to distant alternative to the original social democratic issue owner.



  1. Conclusion

The paper’s point of departure was the argument that some contextual conditions such as universal welfare state (schemes) change the rules of the game of issue voting. Since many voters are enrolled in major welfare schemes such as health or pensions, the simple logic of issue voting, i.e. a pro-welfare centre-left and a welfare-sceptical centre-right trying to set the agenda around their own issues can no longer applied for countries where many welfare schemes have a universal character. In universal welfare states do welfare issues have a high level of issue salience and avoidance is seldom feasible (Goul Andersen 2003; Aardal & van Wijnen 2005; Giger 2011). This marks a structural advantage for social democrats and disadvantage for the market-oriented centre-right. Such scope conditions have often been neglected in recent studies of issue voting and we lack more systematic investigations why some countries saw centre-right parties or coalitions winning against the hitherto dominant social democrats on hostile terrain.

I argued that centre-right parties that seek to increase their electoral prospects under conditions of issue competition in a universal welfare need to adapt to the unfavourable circumstances by signalling competence; that is issue ownership, on some crucial welfare issues. This is, however, not sufficient since the centre-right also needs to demonstrate that it has effectively stripped off its market-liberal stance and converged to the traditional issue owner, the social democrats, in a credible manner. In other words, voters ought to belief the welfare state in save hands under centre-right governments.

Using the case of Denmark in the period 1998-2007, where the main party of the centre-right Venstre has made such a centrist move on welfare, I analysed whether issue ownership effects are moderated by the ideological distance between the major parties. The results demonstrated that issue ownership over issues of health and pension worked most effectively for the Liberal Party, if voters perceived no or only a minor ideological distance between the Liberals and its social democratic contender. Larger ideological distance benefited the social democrats even if the Liberals or more precisely a Liberal-led government was regarded as issue owner over these welfare schemes. Given its social protectionist stance, the second largest centre-right party, the Danish People’s Party, benefits if the ideological distance between the two major parties is still perceived as large.

The findings have several implications for the issue voting literature. First in view of the phenomenon that a traditional market-liberal party has become the largest party in a universal welfare state (Denmark) and similar developments occurred in other countries, especially Sweden9, we need to expand our theoretical framework on issue voting by taking into account the role of the welfare state for issue voting at the micro-level. Not in all electoral contexts is issue voting a simple competition around getting the own issues on the agenda and winning electoral support, but parties need to adapt to different contextual conditions of issue voting.

Another implication is that issue ownership effects are moderated by ideological proximity. Attempts to win issue ownership over non-owned issues are most effective among voters that perceive the ideological convergence as a credible move. Furthermore, issue convergence may have an effect on performance or competence evaluations as competence against the traditional issue owner is only credible if a party also signals that it has moved towards the traditional issue owner in this policy area.

The study also analysed issue voting outside the typical universe of cases where first-past-the-post electoral systems and two party, candidate-centred systems, are common and the issue voting framework may work best. Outside this universe, we may have to refine our arguments on issue ownership voting and to take further scope conditions such as large welfare states, multiparty systems, and a two or three dimensionality of the party system into account. A natural next step would be to investigate whether convergence on the welfare-economic dimension allows competing on other issues on the value dimension of politics, e.g. crime or immigration, and to see to which extent bereaving the social democrats of its main weapon makes value competition more effective.



References:

Aardal, Bernt & van Wijnen, Pieter (2005): Issue Voting. In: Thomassen, Jacques (ed.): The European Voter. Oxford, Oxford University Press. Pp. 192-212.

Adams, James F., Merrill III, Samuel & Grofman, Bernard (2005): A Unified Theory of Party Competition. A Cross-National Analysis Integrating Spatial and Behavioral Factors. New York, Cambridge University Press

Bélanger, Éric & Meguid, Bonnie M. (2008): Issue Salience, Issue Ownership, and Issue-based Vote Choice. Electoral Studies 27(3), pp. 477-491.

Blomqvist, Paula & Green-Pedersen, Christoffer (2004): Defeat at Home? Issue-Ownership and Social Democratic Support in Scandinavia, in: Government and Opposition 39(4), pp. 587-613.

Brambor, Thomas, Clark, William R. & Golder, Matt (2006): Understanding Interaction Models: Improving Empirical Analyses. Political Analysis 14(1), pp. 63-82.

Budge, Ian & Farlie, Dennis (1983): Party Competition – Selective Emphasis or Direct Confrontation? In: Daalder, Hans & Mair, Peter (eds.): Western European Party Systems. London, Sage. Pp. 267-305.

Clark, Harold D., Sanders, David, Stewart, Marianne C. & Whiteley, Paul F. (2009): Performance Politics and the British Voter. Cambridge, Cambridge University Press.

Damore, David F. (2005): Issue Convergence in Presidential Campaigns. Political Behavior 27(1), pp. 71-97.

Giger, Nathalie (2011): The Risk of Social Policy? The Electoral Consequences of Welfare State Retrenchment and Social Policy Performance in OECD Countries. London & New York, Routledge.

Goul Andersen, Jørgen (2003): Partiernes Image: De Borgerlige Bedst til at Sikre Velfærden. In: Goul Andersen, Jørgen & Borre, Ole (eds.): Politisk Forandring. Værdipolitik og Nye Skillelinjer ved Folketingsvalget 2001. Aarhus, Systime. Pp. 151-170.

Green, Jane (2007): When Voters and Parties Agree: Valence Issues and Party Competition. Political Studies 55(3), pp. 629-655.

Green, Jane & Hobolt, Sara (2008): Owning the Issue Agenda: Party Strategies and Vote Choices in British Elections. Electoral Studies 27(3), pp. 477-491.

Green-Pedersen, Christoffer (2007): The Growing Importance of Issue Competition: The Changing Nature of Party Competition in Western Europe. Political Studies 55(3), pp. 607-628.

Holian, David B. (2004): He’s Stealing my Issues! Clinton’s Crime Rhetoric and the Dynamics of Issue Ownership. Political Behavior 26(2), pp. 95-124.

Jæger, Mads Meier (2006): Welfare Regimes and Attitudes Towards Redistribution: The Regime Hypothesis Revisited. European Sociological Review 22(2), pp 157-170.

Mortensen, Hans (2008): Tid til Forvandling. Venstres Vej til Magten. Copenhagen, Gyldendal.

Oscarsson, Henrik & Holmberg, Sören (2008): Regeringsskifte. Väljarna och Valet 2006. Stockholm, Norstedts Juridik.

Petrocik, John R. (1996): Issue Ownership in Presidential Elections, with a 1980 Case Study. American Journal of Political Science 40(3), 825-850.

Petrocik, John R., Benoit, William L. & Hansen, Glenn J. (2003): Issue Ownership and Presidential Campaigning, 1952-2000. Political Science Quarterly 118(4), pp. 599-626.

Rothstein, Bo (1998): Just Institutions Matter. The Moral and Political Logic of the Universal Welfare State. Cambridge, Cambridge University Press.

Stubager, Rune & Slothuus, Rune (2012): What Are the Sources of Political Parties’ Issue Ownership? Testing Four Explanations at the Individual Level. Political Behavior DOI 10.1007/s11109-012-9204-2.

Svallfors, Stefan (2006): The Moral Economy of Class. Class and Attitudes in Comparative Perspective. Stanford, Stanford University Press.

Van der Brug, Wouter (2004): Issue Ownership and Party Choice. Electoral Studies 23(2), pp. 209-233.



Van der Brugge, Jimmy & Voss, Henning (2007): Årsager til de Socialistiske Partiers Tilbagegang i Perioden 1990-2005. In: Goul Andersen, Jørgen, Andersen, Johannes, Borre, Ole, Hansen, Kasper Møller & Nielsen, Hans Jørgen (eds.): Det Nye Politiske Landskab. Folketingsvalget 2005 i Perspektiv. Aarhus, Academica. Pp. 127-151.

1 This corresponds with the literature’s typical focus on the UK case, where the electoral system reduces the number of parties to two (and a half), and valence voting on the economy as predominant form of issue voting.

2 In this respect, research on social policy attitudes has shown that universal welfare states have attracted considerable middle class support as middle class voters are included in the major welfare schemes in contrast to the liberal model where the middle class has been said to prefer a more lenient welfare and private solutions or provisions (Rothstein 1998; Jæger 2006; Svallfors 2006).

3 Typically, centre-right governments in Denmark and Sweden had been short-lasting until the turn of the millennium or were characterized by frequent removal or withdrawal of coalition parties and government reshuffles as in the case of the Swedish non-socialists government (1976-1982) or the various centre-right minority cabinets in Denmark since 1982.

4 The data comes from a pooled data set for the Danish Election studies 1971-2005 compiled at the Department of Political Science, Aarhus University. Since not all variables of the 2007 election are included in this dataset, the 2007 survey has been re-coded after same principles and merged with the other dataset.

5 In the period under review, the Social Democrats, Venstre, the Danish People’s Party, the Conservatives, the Socialist People’s Party, the Radical Liberals, the Unity List, and since 2007 the New Alliance contested elections and gained parliamentary representation. These parties and non-voting plus independents and minor parties were included in the analysis, while smaller parties (EL, LA, independents) form a residual category to avoid non-occupied cells.

6 This corresponds to the findings of Bélanger & Meguid (2008).

7 This is also largest effect for the distance measure for all parties and non-voting vis-à-vis the social democrats, only the coefficient for the small radical liberal party comes close to that

8 If immigration is modelled as salient issue together with pensions and it is assumed that a Liberal government owns the immigration issue, there are similar patterns. The Liberals perform best if there is perceived ideological convergence, whereas the social democrats’ shares increase with larger ideological distances. Not surprisingly, the Danish People’s Party outperforms both major parties if both immigration and pensions are a salient, but the party’s share increase with increasing ideological distance as in the simulation where immigration is assumed to be non-salient.

9 In Sweden, the conservative Moderates were able to compete at levels with the hitherto dominant social democratic party SAP in the two recent elections and able to form pure bourgeois governments after a centrist move on welfare. In New Zealand, the Conservative Party (National) recently regained power with a more moderate program after a long Labour incumbency that saw the expansion of the welfare state since the introduction of PR.


Yüklə 66,78 Kb.

Dostları ilə paylaş:




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2020
rəhbərliyinə müraciət

    Ana səhifə