Political parties and styles of representation

Abstract

This paper focuses on how characteristics of parties and party voters explain the styles of representation emphasised within parties. Styles of representation are defined at the party level as the proportion of representatives within parties who are partisans, delegates or trustees. Each style manifests due to different incentives related to the characteristics of their party and/or their party voters. The findings show that the main explanatory factors for the proportion of partisans are parties’ leadership control over nomination and party socialisation. The main determinant for the proportion of trustees is how often parties have been represented in government. For delegates the results are mixed, but it is indicated that a high proportion of party identifiers among party voters is related to a high proportion of delegates within parties.

Keywords

candidates empirical research political parties representative democracy styles of representation

Introduction

In recent years, there has been a renewed interest in studying politicians’ styles of representation and legislative roles (e.g. Blomgren and Rozenberg, 2012b). There is a growing consensus that the Responsible Party Model (American Political Science Association, 1950) (RPM) that has heavily influenced studies of representation is too simplistic (e.g. Bengtsson and Wass, 2011; Valen and Narud, 2007; Wessels and Giebler, 2011). One of the requirements of the RPM to be effective is that representatives should follow their party’s policies; that is to say, they should ideally all be partisans for representation to work. This contradicts the classic notion originally used to explain representation in the US that divides representatives into either trustees or delegates, where trustees use their own judgment to make decisions while delegates supposedly follow voters’ opinion regardless of their own stand on particular issues (e.g. Eulau et al., 1959). Examining representation in France, Converse and Pierce (1986) show that the partisan style together with the trustee and delegate styles are closer to capturing the practice of political representation. Rozenberg and Blomgren (2012) argue that this same threefold classification is linked to both a normative and a philosophical debate on representation, with the major question being how elected representatives should make decisions in modern democracies. These three different roles reflect the different sources representatives use, or claim to use, to make decisions in their work as representatives – specifically when there are conflicting opinions.

The core of political representation is that there is a political relation between representatives and those they represent (Castiglione and Warren, 2006). Political representation is socially constructed, and it is important to understand the context within which it emerges and operates. In the early days of democracy, the concept of political representation was attributed to legislatures as a whole and only later did it become a feature of individual representatives and parties (Pitkin, 1967, in Castiglione and Warren, 2006). This shows that representation was first, and still is, a collective (e.g. a legislature or a party) representing a collective (e.g. the nation or party voters). The representation of those collectives is carried out by individuals operating within parties, and the electorate votes for those parties. This indicates that it is meaningful to analyse styles of representation at the party level, and how parties differ in the emphasis of their representatives (those who constitute a collective) on different styles of representation. In this paper, styles of representation are divided into the proportions of representatives within each party who are trustees, partisans or delegates. The four main explanatory factors for which style is prevalent within parties are: the proportion of candidates who are nominated by the leadership of their respective parties; the proportion of representatives who have a background as locally and/or regionally elected representatives; how often parties have been represented in government; and the proportion of party identifiers among their voters. In this paper, I use the term ‘representatives’ interchangeably with ‘candidates’, referring to both actual and potential representatives for their respective parties.

Styles of representation

Eulau et al. (1959) make the distinction between the focus and style of representation. Focus refers to the group represented, such as constituency voters, party voters or the nation as a whole, and style refers to the manner in which representatives approach their role, traditionally classified into trustees and delegates. In practice, the focus and those two styles are closely related since trustees are more prone to consider themselves to represent the nation as a whole and delegates to focus on specific group(s) (e.g. Bengtsson and Wass, 2011; Wessels and Giebler, 2011).

Due to the seemingly limited impact of styles of representation on the behaviour of individual representatives, studies on those and other role orientations became less popular in the 1980s but are now appearing again on the academic agenda (e.g. Bengtsson and Wass, 2011; Blomgren and Rozenberg, 2012a; Wessels and Giebler, 2011). It is apparent from both the early and recent literature that there is little consensus about how to define representatives’ role-orientation, what terms to use, what best explains them and their consequences. Many authors have proposed other terms and definitions than the trustee, delegate and partisan style, such as purposive roles (Wahlke et al., 1962), position and preference roles (Searing, 1994), and policy representation and interest representation (Thomassen and Esaiasson, 2006). Rozenberg and Blomgren’s (2012) distinction between legislative and representation roles clarifies the subject; legislative roles are concerned with how representatives organize their work in the legislative, while representational roles are about how to represent as the trustee, delegate and partisan styles reflect.

Another important distinction exists between the output and input of styles of representation; the former about the consequences of different styles and the latter referring to what explains them. Only if styles of representation explain a difference in their output is it meaningful to study what explains their input. While the analysis in this paper concerns the input, I first discuss research about the consequences of different styles of representation in order to establish the importance of examining what explains them.

The output of styles of representation

Most studies on the output of representation focus on whether different styles of representations have different behavioural or attitudinal consequences at the individual level (e.g. Andeweg, 2012; Gauja, 2013). It is generally not presumed that if there are behavioural consequences the same or similar behaviour can be expected under all circumstances (e.g. Andeweg, 1997). There are numerous other factors that could potentially explain how decisions are made such as party discipline (Kristinsson, 2011) and the limited number of issues on which representatives can be experts (Andeweg, 2012). Even if role orientation lacks explanatory power when studying legislative behaviour, commonly operationalised as representatives roll-call vote (e.g. Converse and Pierce, 1986; Kuklinski and Elling, 1977), it could very well explain other behaviours of MPs such as how they approach their work in parliamentary committees and, as pointed out by Andeweg (2012), their interaction and attitudes towards voters.

Using the terms trustees and delegates, adding the role of politicos, which is between the trustee and the delegate role (sometimes follow voters and sometimes use own judgment), Andeweg (2012) finds that delegates in the Netherlands are in less contact with voters and are more cynical about them. These findings are contrary to what might be expected, if it is to be assumed that delegates should have a more favourable view of voters compared to trustees and politicos. However, Andeweg’s findings do show that there is systematic difference related to how representatives think about their role, even if it is contrary to what is expected. In the same study, Andeweg shows that representatives that emphasise that the policy preferences of voters should be incorporated into their party’s policy (bottom-up representation) are in more contact with voters while representatives who favour elite-driven party politics (top-down representation) have more frequent contact with ministers and officials to discuss the problems of individual citizens.

Önnudóttir (2014) establishes that styles of representation do explain policy congruence as the outcome of representation between parties and their voters. She shows that parties with a high number of partisans have lower policy congruence with party voters compared to parties with a high number of trustees who have higher policy congruence with their voters. Her findings repudiate the RPM, which assumes that adhering to the prevailing party policy is the optimal way for representatives to represent the voters of their party; who should have voted for the party that is closest to their own policy preferences. These findings do call for an explanation of why parties differ in their emphasis on styles of representation, which is the subject of this paper.

The input of styles of representation

An important feature of the trustee, partisan and delegate styles is that those are about how representatives think about their representational role and what source(s) they use, or claim to use, when making decisions. Under the trustee style, the source for decision making is the representative themselves, under the partisan style it is the party policy and under the delegate style the source is the voters. Considering this main difference, it is reasonable to hypothesise that there are different factors linked to each source that encourage different styles within parties. When the source for decision making is the party, as in the partisan style, characteristics of that source should explain whether the partisan style is prevailing within parties. The very same goes for the promotion of the delegate style; in that case, the characteristics of the voters of each party (party voters) may explain a high or a low proportion of delegates within parties. When those factors are absent, parties have more flexibility to promote the trustee style which is the only one where the source for decision making is not a collective. However, as elaborated on below, I argue that parties’ representation in government promotes the trustee style within parties along with the absence of the abovementioned factors.

The effect of the institutional setting of the political and electoral system on political representation has received considerable attention (e.g. Golder and Stramski, 2010; Wessels, 1999; Wessels and Giebler, 2011). Those studies that focus exclusively on representatives’ role-orientation commonly include one country or a handful of countries (e.g. Costa et al., 2012; Ilonski, 2012; Saalfeld, 2007). The low number of countries means that the effect of the political and electoral system can only be speculated on. To this there are a few exceptions. Wessels (1999) finds that the smaller the district magnitude the more both European and national MPs focus on their constituency. Farrell and Scully (2010) show that as electoral systems become more open, with greater flexibility for voters to indicate a preference for certain candidates, the more prone elected representatives in the European Parliament (EP) are to focus on their constituency. More recently, Wessels and Giebler (2011) find that the higher the chance candidates for the EP consider themselves to have of getting elected the less likely they are to be partisans, indicating that the more secure the candidates’ election is, the less of a stronghold their parties hold over them.

The main question in this paper is whether there is a difference in the emphasis of styles of representation on the level of parties and party voters. As such, my approach differs from many earlier studies where the focus is on individual representatives or candidates operating within different parties and under different electoral settings. My argument is that representation is, at its core, about a collective representing a collective. Thus it is meaningful to analyse whether and how different styles of representation are promoted on the level of those collectives. Due to the past focus in the literature on what explains individuals’ role-orientation and the fact that political parties are collectives made up of individuals, my hypotheses are formulated based on expectations about how different political context affects representatives within parties, but focusing on the party level.

Strøm (2012) casts styles of representation as strategies representatives use to attain their political goals, whether that is for example (re-)selection on the party list or (re-) election. These strategies are conditioned by contextual factors such as the electoral system and the parties’ characteristics. One obvious contextual factor is the control the party leaderships’ have over the nomination of candidates. The greater their control the more likely it would be for representatives within those parties to adopt the partisan style because it increases their chances of advancement for and within the party, or:

H1: The higher the proportion of party candidates who are nominated by the parties’ leaderships, the higher the proportion of partisans within parties.

Zittel (2012) explains styles and focus of representation in Germany with the main contextual factors as party competition, party socialisation and the mode of the election (elected via party list or in a single seat district). He finds that representatives who consider themselves to have a fair chance of winning and are elected in single member districts are more inclined to consider themselves to represent their district and those who consider it unlikely to win are more likely to be partisans. Zittel operationalises party socialisation on the individual level as years of party membership, party employment in years and for how many years representatives have held a local or regional party office. He finds no support for the effect of party socialisation through these measurements, but finds that younger representatives lean more towards the partisan role, perhaps signalling increasing professionalisation of politics, with young politicians subscribing to the partisan role as the best strategic choice for them to advance their careers.

Years of party membership or holding a party office are not the only means to enhance a political career. Parties’ socialisation effect might manifest itself through the support needed when competing for public office at the lower levels of the political system, i.e. the local and/or regional level. Political careers in public office generally take off at those lower levels. It is thus rational for representatives on those levels to gain and maintain the support of party elites within their districts to enhance their political career (Zittel, 2012). For this reason, I argue and test whether the effect of parties’ socialisation on the partisan style manifests itself through a high number of representatives who have been elected on the local and/or regional level:

H2: The higher the proportion of party candidates who have been elected on the local/and or regional level, the higher the proportion of partisans within parties.

Strøm (2012) argues that the more a representative is dependent on his constituents for (re)-election the more likely s/he is to adopt a role that conforms to the expectations of her/his constituents. Which style s/he would adapt to under those circumstances depends on features of the electoral system and party competition. In addition, it could also depend on the homogeneity or heterogeneity of the constituents. The more heterogeneous the policy preferences of the constituents, the harder it is to advocate the delegate style because the constituency is highly diverse. Thus when faced with a group of homogenous voters, the delegate style is more easily advocated as it does not run the danger of alienating a group of voters who have different preferences.

Rohrschneider and Whitefield (2012) make the point that parties today are faced with the challenge of competing for and representing two diverse but roughly equally sized groups of voters, independent voters and voters who hold a partisan attachment. This they call the ‘strain of representation’. They find that party identifiers are closer to their party on an ideological left-right space and more polarised in their issue positions compared to independent voters. Given that party identifiers are more likely to agree with the issue positions of their party and to take cues from their party for evaluating policy issues (e.g. Dalton, 2008), it can be reasoned that those parties who have a high number of party identifiers among their constituents can more easily promote the delegate style. I do recognise the tentative character of my argument and would like to highlight that my hypothesis below (H3) is explorative:

H3: The higher the proportion of party voters who identify with a party, the higher the proportion of delegates within parties.

Rehfeld (2009) argues that historically the trustee style has three components: aim of decision making; source for decision making; and responsiveness to sanctions, most obviously electoral sanctions. Trustees aim for the greater good in their work, use their own judgement as a source and are less responsive to sanctions compared to delegates, and instead act on some form of civic virtue. If it is true that trustees aim for the greater good, it is plausible to argue that those parties who have a greater experience in working for the greater good promote the trustee style. Parties who are frequently represented in government where they have to implement collectively binding decisions might be more used to justifying their decisions as being for the greater good. Parties’ representation in government can thus encourage a nation-wide focus of representation, a focus that is inherent in the trustee style.

A different but related argument is that government-seeking parties aim to secure enough votes to make them viable candidates for government. For that, they both have to secure the votes of their own partisan voters (party identifiers) and of independent voters. Thus it could be strategically prudent for those parties to downplay the partisan style in order not to alienate independent voters and downplay the delegate style because they want to secure the votes of both groups. Instead, it is strategically advantageous to promote the trustee style to signal that the party and its representatives are competent enough to make informed decisions based on their own judgement, representing both partisan and independent voters. This is not meant to predict how elected representatives will behave once their party is in government. There are many reasons to believe that individual representatives of government parties will stick to their party’s positions in their work as legislators, whether that be factors such as the importance of securing their government’s majority in parliament (Kristinsson, 2011) or aspirations to hold onto an executive position (Müller, 2000). It is also worth noting that in this paper I make use of candidate data for measures on styles of representation and not only elected representatives, who might reply differently. Instead of limiting my analysis to elected representatives, it is justifiable to incorporate all candidates based on the idea that if parties do systematically promote certain styles of representation that should be found among all candidates including the elected ones.

Whether it is a socialisation effect as a consequence of parties’ representation in government or strategically wise for parties to emphasise the trustee style in order to maximise their electoral success is an open question. However, it can be tested whether those two constructs, government-seeking parties and the proportion of trustees within parties, go together:

H4: The more often parties have been represented in government the higher the proportion of trustees within parties.

Research design

Data and selection of countries

Data on styles of representation, the proportion of candidates nominated by parties’ leaderships and the proportion who have backgrounds as elected representatives on the regional/local level comes from the Comparative Candidates Survey (CCS; www.comparativecandidates.org). The countries and national elections in the CCS data that include the variables of interest are the following 10: Belgium 2007, Estonia 2011, Germany 2009, Greece 2007, Hungary 2010, Iceland 2009, Ireland 2007, Netherlands 2006, Portugal 2009 and Sweden 2010, covering a total of 62 parties. In countries where legislatures are bicameral, only representatives running for the lower house are included in the CCS and the research is therefore limited to those.¹

Information about parties’ representation in government comes from the Parliament and Government Composition Database (ParlGov; www.parlgov.org). Four data sources are used for the proportion of party identifiers for each party: the Comparative Study of Electoral Systems (CSES; http://cses.org/datacenter/module3/module3.htm), the European Social Survey (ESS; www.europeansocialsurvey.org), European Election Studies (EES; www.ees-homepage.net) and the Irish National Election Study (INES; www.tcd.ie/ines).²

Response variables: Styles of representation

In the CCS data there are three items that reflect representatives’ styles of representation. Those three are questions are about how an MP should vote in parliament if there are different opinions between:

The party position and his/her party voters’ opinion.

The constituency voters’ opinion and the MP’s opinion.

The party position and the MP’s opinion.

‘Constituency voters’ and ‘his/her party voters’ are both referred to as voters in this paper.³ To classify representatives’ style of representation, I use two out of the three questions for each style. Those who say that the MP should vote according to the party when contrasted with own opinion and voters in items 1 and 3 are partisans, those say that the MP should follow the voters’ view in items 1 and 2 are delegates and those who name the MP in items 2 and 3 are trustees. The representatives who cannot be categorised according to this rule are coded as ‘non-classifiable’.

Table 1 lists the proportion of representatives classified under each style for the 62 parties. It is notable that in 55 of the parties less than 15% of the candidates are non-classifiable and in 57 of them less than 20% are. The low proportion of non-classifiable candidates indicates that there is a systematic component in their replies that can be used to categorise them as partisans, delegates or trustees. At the bottom of the table, a comparison between Eastern and Western Europe is shown, as well as the total proportions for each style; firstly for the pooled un-weighted data, secondly weighted by country and thirdly by party. In the analysis, representatives’ replies are not weighted since the unit of analysis is at the party level.

Table 1.

Proportion of partisans, delegates, trustees and non-classifiable within parties.

	Partisans	Delegates	Trustees	Non-classifiable	N
Belgium
Socialist Party	37.2%	14.0%	34.9%	14.0%	43
Reformist Movement	28.3%	13.0%	32.6%	26.1%	46
Humanist Democratic Center	44.4%	8.3%	33.3%	13.9%	36
Ecolo	47.7%	2.3%	27.3%	22.7%	44
National Front	25.0%	50.0%	20.8%	4.2%	24
Christian Democratic and Flemish	41.7%	14.6%	33.3%	10.4%	48
New-Flemish Alliance	21.4%	35.7%	35.7%	7.1%	14
Socialist Party. Different	57.1%	14.3%	28.6%	0.0%	35
Spirit	62.5%	12.5%	25.0%	0.0%	8
Open Flemish Liberals and Democrats	30.0%	30.0%	30.0%	10.0%	50
Flemist Interest	41.2%	20.6%	29.4%	8.8%	34
Green!	32.1%	12.5%	48.2%	7.1%	56
List-Dedecker	7.5%	52.5%	30.0%	10.0%	40
Estonia
Estonian Center Party	33.3%	11.1%	44.4%	11.1%	9
Estonian Reform Party	32.4%	24.3%	32.4%	10.8%	37
Union of Pro Patria and Res Publica	25.0%	20.0%	45.0%	10.0%	40
Social Democratic Party	14.3%	26.5%	49.0%	10.2%	49
Germany
Social Democratic Party	19.9%	9.3%	56.3%	14.6%	151
Christian Democratic Union	11.5%	9.4%	72.7%	6.5%	139
Christian Social Union	0.0%	4.2%	87.5%	8.3%	24
Free Democratic Party	6.3%	5.6%	77.6%	10.5%	143
Alliance 90 / Greens	4.6%	7.9%	79.5%	7.9%	151
Left Party	13.1%	23.4%	51.8%	11.7%	137
Greece
New Democracy	47.4%	20.5%	23.1%	9.0%	78
Panhellenic Socialist Movement	27.1%	24.3%	40.2%	8.4%	107
Hungary
Fidesz – Hungarian Civic Alliance	27.2%	40.8%	14.3%	17.7%	147
Hungarian Socialist Party	46.2%	26.9%	15.4%	11.5%	52
Jobbik	9.1%	73.6%	14.5%	2.7%	110
Politics Can be Different	7.5%	41.8%	38.8%	11.9%	67
Iceland
Social Democratic Alliance	10.8%	16.2%	58.1%	14.9%	74
Progressive Party	16.9%	7.7%	61.5%	13.8%	65
Independence Party	8.9%	7.1%	73.2%	10.7%	56
Left Green Movement	13.2%	17.6%	52.9%	16.2%	68
Civic Movement	6.3%	25.4%	63.5%	4.8%	63
Ireland
Fianna Fáil	57.1%	7.1%	14.3%	21.4%	42
Fine Gael	44.4%	8.3%	30.6%	16.7%	36
Labour	47.1%	23.5%	17.6%	11.8%	17
Green Party	53.6%	14.3%	28.6%	3.6%	28
Progressive Democrats	44.4%	11.1%	11.1%	33.3%	9
Sinn Féin	66.7%	16.7%	0.0%	16.7%	12
Netherlands
Christian Democratic Appeal	56.7%	0.0%	30.0%	13.3%	30
Labour Party	33.3%	9.5%	47.6%	9.5%	21
Peoples Party for Freedom and Democracy	30.8%	15.4%	46.2%	7.7%	26
Green Left	27.3%	0.0%	72.7%	0.0%	11
Socialist Party	64.0%	0.0%	28.0%	8.0%	25
Democrats 66	22.2%	0.0%	72.2%	5.6%	18
Christian Union	55.6%	0.0%	33.3%	11.1%	9
Political Reformed Party	63.6%	9.1%	27.3%	0.0%	11
Party for the Animals	36.4%	9.1%	27.3%	27.3%	11
Portugal
Left Bloc	12.2%	39.0%	39.0%	9.8%	41
People’s Party	18.5%	37.0%	33.3%	11.1%	54
Democratic Unity Coalition	52.0%	28.0%	20.0%	0.0%	25
Social Democratic Party	26.5%	22.4%	42.9%	8.2%	49
Socialist	24.0%	16.0%	52.0%	8.0%	25
Sweden
Center Party	38.8%	16.3%	37.8%	7.1%	196
Liberal Party	26.5%	21.8%	38.8%	12.9%	147
Christian Democrats	36.5%	15.1%	38.5%	9.9%	192
Green Party	21.8%	27.4%	42.1%	8.6%	197
Moderate party	46.9%	17.6%	24.6%	10.9%	256
Social Democrats	53.1%	12.5%	24.6%	9.8%	256
Sweden Democrats	60.0%	12.0%	24.0%	4.0%	25
Left Party	39.0%	16.6%	34.2%	10.2%	187
Comparison between Eastern and Western Europe, weighted by country*
East (Estonia and Hungary)	22.4%	31.2%	35.5%	10.9%
West (Belgium, Germany, Greece, Iceland, Ireland, the Netherlands, Portugal and Sweden)	32.0%	40.6%	16.5%	10.8%
Total	29.3%	19.6%	40.6%	10.5%	4200
Weighted by country*	30.0%	20.6%	38.6%	10.8%
Weighted by party**	32.7%	18.5%	38.2%	10.6%

Note: Entries are proportion of candidates assigned to each style of representation. *In the country weight respondents in each country have equal weight (for example respondents’ replies in Estonia 2011 have the same weight as respondents’ replies in Ireland 2007). **In the party weight respondents in each party have equal weight (for example respondents’ replies for the Social Democratic Party in Germany 2009 have the same weight as respondents’ replies for the Left Green Movement in Iceland 2009). Country and party weights are calculated based on the total number of replies to questions on style divided by the number of countries or number of parties in each scenario.

Even if the focus in this paper is on party and party voters’ attributes, it cannot be ignored that there seems to be a partial difference in main emphasis on styles between countries. The strong emphasis on the trustee style in Iceland and Germany and on the partisan style in Ireland might be due to the decentralisation of candidate selection in the two former countries (e.g. Rahat, 2007) and the strong party discipline in Irish politics (Marsh, 2000). The trustee style is more common in the two Eastern European countries compared to the West. However, in Hungary the delegate role is the one emphasised by most candidates in three out of four parties, most notably by 73.6% of candidates of Jobbik, a right-wing populist party. Analysing differences between countries is beyond the scope of this paper but is a clear avenue for future research.

For party measures on styles of representation, I use the proportion of representatives within parties who are either: 1) partisans, 2) delegates or 3) trustees. Representatives that are non-classifiable are included in the calculation for the proportion of each style within parties, but not analysed any further. An example of party measures for the Socialist Party in Belgium is 37.2 for the proportion of partisans, 14.0 for the proportion of delegates and 34.9 for the proportion of trustees.

Explanatory variables: Parties and party voters

The control of party leadership over who is nominated is operationalised as the proportion of candidates of a given party who are nominated by their party’s leadership using the CCS data and party socialisation from the same data is the proportion of the parties’ candidates who have been elected for local and/or regional office (parliament and/or government). Data from ParlGov is used to construct a measure on parties’ representation in government. This measure is the proportion of days the parties have been represented in government in the past four electoral terms preceding the election under study in each country. Information about the proportion of party identifiers among the voters of each party is from the third wave of the CSES for Estonia, Iceland, Germany, Netherlands and Portugal; from the EES 2009 for Belgium and Greece; from the ESS 2010 for Hungary and Sweden; and from the INES 2007 for Ireland. For this measure, I use the proportion of party identifiers among voters (based on vote-recall) for each party.

Control variables

Previous research has established that representatives of traditional left-wing parties are more likely to emphasise the partisan style and that representatives of parties from the centre to right are more likely to emphasise the trustee style (e.g. Damgaard, 1997; Gauja, 2013). To control for those possible effects, I include two dummies indicating whether the parties are social democratic/communist/left parties capturing the traditional left or conservative/Christian democratic parties who are usually placed from centre to right. The party codes are retrieved from the Comparative Manifesto Project (CMP; https://manifestoproject.wzb.eu).

Earlier research has also shown that candidate-centred politics as opposed to party-centred politics encourage representatives to focus more on their constituency (e.g. Farrell and Scully, 2010; Norris, 2002). This is quite often operationalised as a distinction between the different ballot structures, contrasting an open ballot structure with a closed one. Under a closed ballot structure, voters can only vote one party ticket and that has been shown to provide incentives for parties to promote the partisan style (Norris, 2002). In open ballot structures, voters can either vote directly for a certain candidate or have the opportunity to rank or indicate their preferred candidate, resulting in a greater focus on their constituency (Farrell and Scully, 2010; Norris, 2002). As such, I include ballot structure as a control measure in my models contrasting Iceland and Portugal, which use a party ballot, with the remaining eight countries that make use of various versions of ballots where a candidate preference can be or is expressed.⁴ However it is noted that the number of countries in this study is too low to any draw firm conclusions about the effect of the electoral system on styles of representation.

The last control variable introduced here is a dummy for Estonia and Hungary as former members of the Eastern European communist regime. The modern party systems in these countries are younger compared to the other countries included in this study, and it is still a matter of debate whether they have stabilised (e.g. Birch, 2001; Dahlmann, 2005; Enyedi, 2006; Róbert and Papp, 2012). For that reason, it is possible that the party indicators used in this study have a different impact on styles of representation in Estonia and Hungary compared to other countries included here. A table with an overview of all variables used in this study can be found in Appendix I.

Data analysis

In order to examine how the parties leaderships’ control over nomination, party socialisation, how often the parties have been represented in government and the proportion of party voters who identify with the party, explains styles of representation, I use Ordinary Least Square (OLS) regressions. The response variables are the proportion of representatives within parties who are trustees, partisans or delegates. The three proportions for style of representation are not independent of each other; a higher proportion in one style group is followed by lower proportions in the other two groups. For that reason, I run separate regressions for each of the three style groups. The regressions are done stepwise adding in the first four steps one explanatory variable at a time. In the fifth step I add the control variables for the type of parties and the ballot structure, and in the sixth I control for the Eastern European countries. For each style, I first enter the variables that relate directly to the subject of my hypotheses, and my discussion focuses on significant parameters. The parties included are not chosen randomly and for that reason the significance levels are only meaningful as indicators about the strength of the relationship between the explanatory variables and the response variables among the 62 parties included.

My first hypothesis (H1), that the higher the proportion of candidates who are nominated by the parties’ leadership the higher the proportion of partisans, is supported in the full model (Table 2). It is notable that the effect of party leadership control over nomination is positive for both the proportion of partisans and delegates once all explanatory variables have been entered into the calculation, while it is negative for the proportion of trustees. The relationship between party socialisation and the proportion of partisans is as expected (H2); the higher the proportion of representatives who have a background in local and/or regional politics, the higher the proportion of partisans. Examining this for the other two style groups, the direction is the opposite, but non-significant in the case of the trustee style.

Table 2.

Determinants of styles of representation.

	% of partisans within parties						% of delegates within parties						% of trustees within parties
	Standardised betas						Standardised betas						Standardized betas
Step:	1	2	3	4	5	6	1	2	3	4	5	6	1	2	3	4	5	6
% of party representatives who:
Are nominated by the party’s leadership	0.20	0.27*	0.28*	0.30*	0.26*	0.22+		0.16	0.08	0.11	0.11	0.24*		–0.33**	–0.37**	–0.36**	–0.34*	–0.37**
Have been represented in local and or/ regional politics		0.44***	0.47***	0.49***	0.39**	0.38**			–0.39**	–0.32*	–0.28+	–0.29*			–0.19	–0.19	–0.12	–0.12
Parties and party voters:
Representation in government			–0.12	–0.12	–0.14	–0.12				–0.23	–0.20	–0.26+	0.16	0.18	0.23+	0.22+	0.21	0.23+
Proportion of party identifiers among party voters				–0.08	–0.09	–0.08	0.14	0.11	0.21+	0.20	0.21+	0.17+				–0.01	–0.01	0.00
Control variables
Socialist democratic / socialist / communist parties					0.16	0.13					–0.20	–0.03					–0.05	–0.02
Conservative / christian democratic parties					0.11	0.10					–0.06	–0.09					0.04	–0.04
Party ballot					–0.24+	–0.27+					0.06	0.15					0.19	0.16
Hungary/Estonia						–0.17						0.54***						–0.15
Intercept, p–value	0.000	0.112	0.096	0.082	0.028	0.018	0.008	0.015	0.000	0.003	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
R^2:	0.04	0.22	0.24	0.24	0.30	0.32	0.02	0.04	0.18	0.23	0.26	0.52	0.03	0.14	0.17	0.17	0.21	0.22
Change in R2 significant, p value:	0.123	0.000	0.329	0.516	0.224	0.164	0.289	0.209	0.003	0.061	0.499	0.000	0.201	0.008	0.134	0.907	0.504	0.264

Note: Response variables are three; the proportion of candidates within each party who are 1) partisans, 2) delegates and 3) trustees. Significance levels: +p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001. N = 62 parties. Variance inflation factor (VIF) in the full model (step 6) is in all cases lower than 1.4 and tolerance from 0.73 (% have been represented in local and/or reginal politics) to 0.90 (% nominated by the party’s leadership).

The third hypothesis (H3), that the higher the proportion of party identifiers among party voters, the higher the proportion of delegates, is supported in the full model, but the relation is not very strong. The fourth and final hypothesis (H4) is supported in the full model; the more often a party has been represented in government, the higher the proportion of trustees within parties. Representation in government has an opposite effect on the proportion of delegates; the more often parties have been represented in government the lower the proportion of delegates.

Looking at the other control variables, types of parties and party ballot structure, the type of parties as operationalised here do not have an impact on the proportions of any of the three styles of representation, but the party ballot structure is negatively related with the proportion of partisans. This is contrary to the expectation that a party ballot structure promotes the partisan style. Due to the low number of countries in my models, it is meaningless to draw any firm conclusions here about the effect of the party ballot structure on the partisan style. The control variable for Hungary and Estonia is statistically significant in the delegate model, indicating that parties in at least one of those countries are more likely to contain a high proportion of delegates.

In Figures 1 to 4, I examine graphically the relation between the main explanatory variables with each of the relevant predicted values (unstandardised) for styles of representation calculated from the final models (step 6) in Table 2. This I do both for the total number of parties and for each country. Figure 1 plots the proportions of candidates who were nominated by the parties’ leaderships as against the predicted values for the partisan style. The positive but modest relation (R² = 0.12) for the total number of parties is mainly driven by parties from Belgium, Greece, Iceland and Portugal. Only in two out of those four, Belgium and Portugal, does the proportion for candidates who are nominated by their parties’ leadership vary – while in the other two it does not. In five of the countries the parties’ leaderships do exercise very little or almost no control over nomination in all or almost all parties (Estonia, Germany, Iceland, Ireland and Sweden) indicating that for parties in those countries there are other factors that explain the partisan style. Parties’ socialisation effect has a much stronger relation to the proportion of partisans (Figure 2) compared to parties’ leaderships control over nomination. The proportions of candidates who have backgrounds as local and/or regional representatives have a positive relation with predicted values for the partisan style in all countries except Estonia and a strong positive relation among the total number of parties (R² = 0.47). This effect is weak in Greece, a country that together with Estonia makes use of a strong preferential vote and that could possibly explain the negligent effect found in those two countries. It could be that under a candidate centred ballot structure, candidates are prone to downplay the partisan style in order to distinguish themselves from their competitors. However, the overall positive trends give ground to the claim that this effect is independent of other factors such as the electoral system or different types of parties.

Figure 1.

Predicted values for the partisan style and parties’ leaderships’ control over nomination.

Figure 2.

Predicted values for the partisan style and party socialisation.

Figure 3.

Predicted values for the delegate style and the proportion of party identifiers in the parties’ electorates.

Figure 4.

Predicted values for trustee style and the proportion of parties’ representation in government.

When plotting the proportions of party identifiers against the predicted values for the delegate style, the results for each country are mixed, and the explained variance among the total number of parties is almost non-existent (R² = 0.04) (Figure 3). Only Belgium and Estonia show a positive trend, while in the other countries it is either negative or there is no trend. This indicates that party identification is neither a strong mover of the delegate style nor a universal trend. Going back to the regression model in Table 2, this does not come as a surprise as the proportion of party identifiers in the parties’ electorates has a weak relation with the proportion of delegates within parties.

The proportions of parties’ representation in government have a positive relation with the predicted values for the trustee style in all 10 countries, and a modest positive trend among the total number of parties (R² = 0.12) (Figure 4). This gives reason to believe that the effect of parties’ representation in government on whether the trustee style is emphasised within parties is independent of other factors of the political system. It can also be gleaned from Figure 4 that among those parties who have never been represented in government there is a high variance in their predicted values for the trustee style and this deflates the explained variance among the total number of parties. This shows that parties’ representation in government is clearly not the only factor that encourages the trustee style, but it is indeed important.

Discussion

In this paper, I have established that the context of political parties, and to some extent the attributes of party voters, does indeed matter for the styles of representation emphasised within parties. I show that there are partly different incentives for the emphasis within parties on each of the three styles – partisans, delegates and trustees. The wider control parties’ leaderships have over nomination and the greater the party socialisation, the more the partisan style is promoted in parties, with party socialisation weighing more heavily than party leadership control over nomination. This means that the success of the highly debated model for representation, the Responsible Party Model, depends on the parties’ control over their candidates’ career paths.

A high proportion of party identifiers in the parties’ electorates is modestly but positively correlated to the proportion of delegates within parties. This could be so because party identifiers of a given party are more homogeneous and in greater agreement with the parties’ policies compared to independent voters. In those cases, parties could promote the delegate style because it is ‘easier’ to advocate the idea that the party voters’ opinions should prevail when party voters are homogeneous. This may even apply more to parties who run on a single-issue platform or a populist right-wing platform. In the case of single-issue parties the ‘will of the voters’ is more clearly signalled compared to in parties who run on a wider platform. Populist parties might be more prone to emphasise the delegate style because the message of that style, that decisions should be based on the ‘will of voters,’ fits the populist platform; referring to the will of the people and justifying their actions by appealing to and identifying with the people (e.g. Jagers and Walgrave, 2007).

The trustee style is more common among government-seeking parties. This may be because representatives of those parties have been socialised into the trustee style as through their work they should aim for the greater good. It is also possible that government-seeking parties promote the trustee style as a response to the ‘strain of representation’ outlined by Rohrschneider and Whitefield (2012). Parties that are faced with the challenge of representing both independent voters and those who identify with the party might be more prone to emphasise the trustee style as a strategy to maximise their electoral success. On a related note, party leadership control over nomination reduces the likelihood of the trustee style. It can be argued that a decentralised nomination process, where candidates’ nominations are dependent on selectors other than their parties’ leaderships, not only creates greater flexibility to adopt the trustee style but also encourages it. Under such a process, a strategic way for competitors to secure their nomination is to signal that they are more competent than their competitors to represent their party and can be trusted to make informed decisions and are thus socialised into the trustee style. This could of course depend on the electoral system, with candidate-centred systems more likely to encourage the trustee style.

It is clear from my results that the way representatives of different parties think about how MPs should treat their mandate is related to and conditioned by their parties’ attributes and to some extent the attributes of their voters. My results have several implications for the understanding of how representation works in modern democracies. Considering that parties offer different policies and that voters have heterogeneous policy preferences, I have shown that it is reasonable to frame the representational link as the link between parties and their voters. What criterion parties’ representatives claim to use to make decisions once in office revolves around their understanding of how they should treat their mandate; whether they are elected to carry out their party’s policy, follow the will of their voters or use their own judgement when making decisions. There is, however, more to this story. Given that a politician’s ultimate goal is either to be a representative or to contribute to their party’s electoral success, specific contextual factors that condition their or their party’s election can encourage different styles of representation. This can be the case whether these are their parties’ or party voters’ attributes as established here, or the setting of the electoral system.

To conclude, my results show that the context of political parties and party voters is important for the promotion of styles of representation within parties. A country comparison shows that the effects of party identifiers and party leadership control over nomination is different between countries, while the effects of party socialisation and party representation in government are more consistent. The mixed results found between countries underlines the importance of a more detailed theorising and analysis of how those factors differ both between political systems and different types of parties, taking into account for example the different degrees of an open ballot structure, decentralisation of the nomination process, competition within parties and party types. One of the contributions of this paper is that it provides the groundwork for such a detailed analysis. It also shows how the different characteristics of the representatives and of those they represent explain the difference in emphasis between parties on representational styles. Those styles are about how to represent which is a crucial element of representation.

Footnotes

Author note

Drafts of this work have been presented in the panel ‘Roles of representation – a comparative approach’ at the 7th ECPR General Conference in Bordeaux and at the CDSS workshop for doctoral students at Mannheim University in Spring 2013.

Acknowledgments

I would like to thank my two supervisors, Hermann Schmitt and Ólafur ñ Haräarson, for substantial comments on the work in process, as well as Åsa Bengtsson, Michael Marsh, my colleagues in the CDSS program at Mannheim University and many others.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This paper is part of my PhD project, paper track. In my PhD I have been funded by the CDSS and MZES at the University of Mannheim and by the Icelandic Research Fund.

Supplemental material

The online appendices are available at .

Notes

References

American Political Science Association (1950) Toward a more responsible two-party system: A report of the Committee on Political Parties. APSR 44(3): Part 2, Supplement 44(3).

Andeweg

(1997) Role specialisation or role switching? Dutch MPs between electorate and executive. The Journal of Legislative Studies 3(1): 110–127.

Andeweg

(2012) The consequences of representatives’ role orientations: Attitudes, behaviour, perceptions. In: Blomgren

Rozenberg

(eds) Parliamentary Roles in Modern Legislatures. Oxon: Routledge, pp. 66–84.

Bengtsson

Wass

(2011) The representative roles of MPs: A citizen perspective. Scandinavian Political Studies 34: 143–167.

Birch

(2001) Electoral systems and party system stability in post-Communist Europe. 97th annual meeting of the American Political Science Association, San Francisco, USA, 30 August – 2 September.

Blomgren

Rozenberg

(2012a) Introduction. In: Blomgren

Rozenberg

(eds) Parliamentary Roles in Modern Legislatures. Oxon: Routledge, pp. 1–7.

Blomgren

Rozenberg

(2012b) Parliamentary Roles in Modern Legislatures. Oxon: Routledge.

Castiglione

Warren

(2006) Rethinking democratic representation: Eight theoretical issues. Centre for the Study of Democratic Institutions, University of British Columbia, USA, 18–19 May.

Converse

Pierce

(1986) Political Representation in France. Cambridge: The Belknap Press of Harvard University.

10.

Costa

Freire

Pilet

J-B

(2012) Political representation in Belgium, France and Portugal: MPs and their constituents in very different political systems. Representation 48: 351–358.

11.

Dahlmann

(2005) Government stability in Estonia: Wishful thinking or reality? Potsdam University, Germany. Available at: http://www.uni-potsdam.de/prof_franzke/moe/moe_paper_03.pdf (accessed 15 February).

12.

Dalton

(2008) Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies (5th edn). Washington DC: CQ Press.

13.

Damgaard

(1997) The political roles of Danish MPs. The Journal of Legislative Studies 3(1): 79–90.

14.

Enyedi

(2006) Stability in the shadow of chaos: The Hungarian party system in 2006. Parliamentary Elections and Party Landscape in the Visegrád Group Countries, 27 October, Masaryk University, Czech Republic.

15.

Eulau

Wahlke

Buchanan

. (1959) The role of the representative: Some empirical observations on the theory of Edmund Burke. American Political Science Review 53: 742–756.

16.

Farrell

Scully

(2010) The European parliament: One parliament, several modes of political representation on the ground. Journal of European Public Policy 17: 36–54.

17.

Gauja

(2013) The Politics of Party Policy: From Members to Legislators. Basingstoke: Palgrave Macmillan.

18.

Golder

Stramski

(2010) Ideological congruence and electoral institutions. American Journal of Political Science 54: 90–106.

19.

Ilonski

(2012) Role stability in the context of institutional and positional change. In: Blomgren

Rozenberg

(eds) Parliamentary Roles in Modern Legislatures. Oxon: Routledge, pp. 162–183.

20.

Jagers

Walgrave

(2007) Populism as political communication style: An empirical study of political parties’ discourse in Belgium. European Journal of Political Research 46: 319–345.

21.

Kristinsson

(2011) Party cohesion in the Icelandic Althingi. Icelandic Review of Politics & Administration 7: 229–251.

22.

Kuklinski

Elling

(1977) Representational role, constituency opinion, and legislative roll-call behaviour. American Journal of Political Science 21: 135–147.

23.

Marsh

(2000) Candidate centred but party wrapped: Campaigning in Ireland under STV. In: Bowler

Grofman

(eds) Elections Australia, Ireland and Malta under the Single Transferable Vote. Ann Arbor, MI: Michigan University Press, pp. 114–130.

24.

Müller

(2000) Political parties in parliamentary democracies: Making delegation and accountability work. European Journal of Political Research 37: 309–333.

25.

Norris

(2002) Ballot

structure

& legislative behavior. Exporting Congress? The influence of the U.S. Congress on world legislatures. Jack D Gordon Institute for Public Policy and Citizenship Studies, Florida, USA, 6–7 December.

26.

Önnudóttir

(2014) Policy congruence and style of representation: Party voters and political parties. West European Politics 37(3): 538–563.

27.

Rahat

(2007) Candidate selection: The choice before the choice. Journal of Democracy 18(1): 157–170.

28.

Rehfeld

(2009) Representation rethought: On trustees, delegates, and gyroscopes in the study of political representation and democracy. American Political Science Review 103: 214–230.

29.

Róbert

Papp

(2012) Party identification and party preference at the 2010 Hungarian general elections. Elections, Public Opinion and Parties conference, University of Oxford, UK, 7 to 9 September.

30.

Rohrschneider

Whitefield

(2012) The Strain of Representation: How Parties Represent Diverse Voters in Western and Eastern Europe. Oxford: Oxford University Press.

31.

Rozenberg

Blomgren

(2012) Bringing parliamentary roles back in. In: Blomgren

Rozenberg

(eds) Parliamentary Roles in Modern Legislatures. Oxon: Routledge, pp. 211–230.

32.

Saalfeld

(2007) Professionalisation of parliamentary roles in Germany: An aggregate-level analysis, 1949–94. The Journal of Legislative Studies 3(1): 32–54.

33.

Searing

(1994) Westminster’s World: Understanding Political Roles. Cambridge: Harvard University Press.

34.

Strøm

(2012) Roles as strategies: Towards a logic of legislative behavior. In: Blomgren

Rozenberg

(eds) Parliamentary Roles in Modern Legislatures. Oxon: Routledge, pp. 85–100.

35.

Thomassen

Esaiasson

(2006) Role orientations of members of parliament. Acta Politica 41: 217–231.

36.

Valen

Narud

(2007) The conditional party mandate: A model for the study of mass and elite opinion patterns. European Journal of Political Research 46: 296–216.

37.

Wahlke

Eulau

Buchanon

. (1962) The Legislative System: Explorations in Legislative Behavior. New York: John Wiley and Sons.

38.

Wessels

(1999) Whom to represent? Role-orientation of legislators in Europe. In: Schmitt

Thomassen

(eds) Political Representation and Legitimacy in the European Union. Oxford: Oxford University Press, pp. 209–234.

39.

Wessels

Giebler

(2011) Choosing a style of representation: The role of institutional and organizational incentives. The 6th ECPR General Conference, 25–27 August, University of Iceland.

40.

Zittel

(2012) Legislators and their representational roles. In: Blomgren

Rozenberg

(eds) Parliamentary Roles in Modern Legislatures. Oxon: Routledge, pp. 101–120.