The universe of group representation in Germany: Analysing formal and informal party rules and quotas in the process of candidate selection

Introduction

In the last several decades, there has been an intensive debate on the representation of social groups in parliaments. The under-representation of women and minority ethnic groups, in particular, has been criticised both in public and in academia. In reaction, national legislatures and political parties in more than 100 countries have adopted group-based quotas for political candidates. These mechanisms are meant to ensure a minimal level of representation for these social groups and serve as a ‘fast track to equal representation’ (Dahlerup, 2006: 6). Quotas are, however, not a new phenomenon but have been used in many countries for decades (Dahlerup, 2006).

Within the intense debate on electoral gender quotas, three main types of policies are distinguished: reserved seats, legislative quotas and party quotas (Dahlerup, 2006: 78; Franceschet et al., 2012). While reserved seats and legal candidate quotas are mandated either by the constitution or by the electoral law, party quotas are adopted ‘voluntarily by individual parties that commit the party to aim for a certain proportion of women among its candidates to political offices’ (Krook, 2009: 7). Worldwide, voluntary party quotas are the most common measure for guaranteeing women’s political representation (Franceschet et al., 2012; Global Database for Quotas for Women, 2013) and the dominant type of quota in Western democracies (Dahlerup, 2006: 295).

Due to the crucial role of political parties as gatekeepers within the candidate selection process, party quotas can have a strong impact on the composition of parliament. Candidate selection is important since ‘the candidates [a party] nominates play an important role in defining what the party is’ (Katz, 2001: 278). Candidates are also the public face of the party for the voters. They characterise, as Hazan and Voerman (2006) argue, the demographic, geographic and ideological dimensions of the party even more than their organisation or their programme. At the same time, candidate selection is also a key arena of internal power struggles (Gallagher, 1988; Katz, 2001). Therefore, parties try to balance their lists in order to represent different groups and interests (Hazan and Rahat, 2010: 108).

Despite the broad range of research on quotas, existing research is lacking in three central aspects. First, research focuses almost exclusively on quotas for women and minority ethnic groups. In contrast, there is very little attention to quotas and rules for other groups. This neglect is striking since there is evidence of quotas for a wide range of different groups. For instance, Krook and O’Brien (2010: 255) refer to quotas for ‘sex, race, ethnicity, nationality, religion, caste, language, age, disability, profession, and location of residence’. Dahlerup (2007) points to the presence of geographical quotas in most countries, and Rahat (2007: 160) mentions ‘candidate-selection rights’, which are granted to interest groups affiliated with political parties, such as unions or employers’ associations. These examples refer to a broader universe of group representation that has not yet been the subject of extensive research.

Linked to this, the second limitation is that most empirical research focuses exclusively on one quota at a time. In particular, there is a lack of studies that analyse quotas comparatively and that focus on the interaction effects and trade-offs between them. Recent research on intersectionality and comparative quota policies discusses interaction effects and trade-offs between claims for quotas on behalf of women and minorities (Htun, 2004; Hughes, 2011; Krook and O’Brien, 2010; Phillips, 1995; see also Celis e al., this issue). Various scholars stress that quotas for group representation are not necessarily complementary (for an overview, see Krook and O’Brien, 2010), being viewed as a hierarchy such that groups compete with each other for recognition (Borchert and Reiser, 2010; Celis et al., this issue; Kymlicka, 1995; Mansbridge, 1999). Understandably, not all social groups can receive quotas on the candidate lists of political parties. The literature mainly addresses these questions regarding hierarchy and competition between groups with respect to the introduction of specific quotas. However, even after their introduction, quotas are deeply controversial within the parties and their legitimacy is constantly questioned (Krook et al., 2009). Therefore, it can be assumed that hierarchy and competition between claims for group representation are also particularly relevant in practice, for instance, when quotas and rules conflict in the process of list-building. In these instances, the debates and decisions can shed light on the hierarchy and perceived importance of the representation of specific groups, as well as on the effectiveness of certain quota regimes.

The third limitation in research on quotas is the almost exclusive focus on formal mechanisms. Reserved seats and legal candidate quotas are stipulated in the constitution or by law and, hence, are formal. In the literature, party quotas are mostly defined as those quotas that ‘are written into the internal rules of a political party’ (Dahlerup, 2006), thus excluding ‘recommendations and targets’ (Global Database for Quotas for Women, 2013). This definition of quotas restricts the analysis to formal party quotas while informal party quotas and rules are excluded. Informal party rules and practices, however, might have an important impact on group representation, even if they have not been formalised in party statutes (Siavelis and Morgenstern, 2008). Therefore, informal rules and quotas that guarantee and influence the representation of certain groups should not be excluded ex ante from the analysis of the universe of group representation.

This article seeks to develop a broader perspective on the universe of group representation by analysing and comparing the practices of German political parties when they selected candidates to the state party lists for the federal elections in 2009. In Germany, as in many other democracies, voters cannot choose among the candidates of a single party, meaning that internal party nominations are decisive in shaping the composition of parliament (Rahat, 2007; Reiser, 2013). Political parties thus have a de facto monopoly on candidate selection and are the gatekeepers to parliament (Hazan and Rahat, 2010; Kirchheimer, 1965; Manow, 2007). The analysis focuses on the types of quotas and rules that German political parties adopted when composing the state party lists for the federal elections in 2009, examining both differences among the parties and how different quotas and rules interact with each other. The article employs an inductive approach to analyse and compare the formal and informal party quotas and rules adopted by six German political parties for the nomination of state party lists in order to theorise which types and forms of quotas and rules can be distinguished and how they interact. Before analysing and comparing the quotas and rules, the central characteristics of the institutional framework, in particular, the electoral system and the party system, in Germany will be described and the multiple-method design will be presented.

The German case: Party quotas, party system and electoral system

Germany presents an interesting case for analysing the universe of group representation measures and comparing political parties. The German party system is characterised by high stability and continuity. Six parties were represented in the German parliament after the federal elections of 2009: the major parties, the Social Democratic Party (SPD) and Christian Democratic Union (CDU), with its Bavarian sister party the Christian Social Union (CSU); and the minor parties, the Liberal Democratic Party (FDP), the Green Party and the Left Party. ¹ The literature assumes that quotas and rules differ systematically between different types of parties (Hazan and Rahat, 2010). Catch-all parties like the SPD, CDU and CSU are expected to have significantly more quotas to balance intra- and extra-party interests and demands than interest-based parties (FDP, Greens and Left Party). Comparative evidence also suggests that party ideology influences both the general evaluation of quotas and the specific quotas of the different parties. Generally, left-wing parties are more likely to introduce group-based quotas than conservative and liberal parties (Davidson-Schmich, 2006; Krook, 2009; Norris, 1997).

The German institutional framework, particularly the mixed-member electoral system (Klingemann and Wessels, 2001), gives parties multiple possibilities to guarantee group representation. Half of the 598 Members of Parliament (MPs) are elected in single-member constituencies, while the other half are elected on closed state party lists (proportional representation (PR) system). Accordingly, there are two types of candidacy and two different bodies that decide candidacies: the nominating convention on the district level for the single-member constituencies and the nominating convention on the state level for the party lists. Hence, formally, there are two independent forms of candidacies. Candidates may, however, run under both formulas simultaneously. Since the early years of the Federal Republic, the two formally independent forms of candidacy became more and more interlinked (Borchert and Reiser, 2010; Manow and Nistor, 2009).

Party quotas and rules are primarily relevant for the nomination of state party lists. ² Various studies show that descriptive representation is significantly higher in PR electoral systems than in single-member district systems, especially in combination with closed party lists and higher district magnitudes (Matland and Studlar, 1996). Parties can balance closed party lists more easily (Htun, 2004; Krook, 2010) and a larger district magnitude allows for a higher amount of quotas (Borchert and Reiser, 2010). Party lists for the federal elections are composed at the state level by the state party branches. The state branches are autonomous in the nomination processes, as well as in the adoption of party quotas. Thus, political parties can adopt different quotas for each of the 16 state party lists. In addition, district magnitude varies vastly between the states, from four seats in Bremen up to 65 seats in North Rhine-Westphalia. Therefore, there may be systematic differences not only between parties, but also within parties.

As in other European countries, in Germany, there are no legislative quotas or reserved seats (Krook, 2009). However, there is a long experience and tradition with party quotas and rules for the composition of state party lists for the federal elections. As early as the 1950s, parties adopted different quotas and rules of proportionality ‘to satisfy different interests and tendencies within the party and to present the most attractive list possible to the voters’ (Roberts, 1988). Various studies in the 1960s and 1970s revealed extensive intra-party rules for regions, intra-party factions, interest groups and social groups (Kaack, 1969; Mintzel, 1980; Zeuner, 1970). Since then, there has been little research on intra-party quotas in Germany, apart from a few studies on the introduction of gender quotas and the representation of migrants (Davidson-Schmich, 2006; Hennl and Kaiser, 2008; Kolinsky, 1991; Wüst, 2011; Xydias, 2008). In particular, there is a lack of studies that analyse the overall impact of these rules and quotas and the interaction effects and trade-offs between them.

Methods and data

To analyse the quotas and rules of proportionality in the German political parties, a combination of different methods was used. First, all national and state party statutes of the six parties – the SPD, the CDU, the CSU, the FDP, the Green Party and the Left Party – were examined. Second, intra-party debates on quotas and rules were analysed through a content analysis of the protocols of the party conventions and associated media coverage. Third, participant observation was used to observe the five state party nomination conventions of the SPD, the CDU, the FDP, the Greens and the Left Party in the run-up to the 2009 federal elections in the state of Hesse, with the aim of gaining further insights into intra-party debates and decisions on the nomination process. Fourth, 124 face-to-face guided interviews were undertaken with candidates and local and state party officials. The interviews focused on the informal rules and quotas, as well as on the interaction effects between them. This multi-method research design permits the analysis of both formal and informal quotas and rules.

The universe of group representation in Germany: Analysis of quotas and rules

Which quotas and rules did the German political parties adopt when composing the state party lists for the 2009 federal elections? How did these various rules interact with each other? Are there systematic differences between the parties? The analysis reveals that German parties adopted an impressively wide range of rules and quotas.

The first rule adopted by all six parties when composing state party lists is the priority rule for district candidates. It is usually only after accommodating all district candidates who want a place on the list that the remaining slots are then filled by list-only candidates in (unwinnable) slots. ³ Even long-term MPs who have important functions in the Bundestag cannot gain a winnable place on the state party list if they have not managed to be nominated again in their home districts. Only 13 deputies (2.1% of the MPs) were elected in 2009 without having a district candidacy. Although this rule is not written down in any party statute, it has existed for decades. The two forms of candidacy have become more and more interlinked since the early years of the Federal Republic, and this principle has eventually also found its way to the newer parties (Borchert and Reiser, 2010; Manow and Nistor, 2009).

The incumbency rule is another important rule. Incumbents get the safest places on the party lists, while newcomers normally follow only after all incumbents have been listed. Similar to the priority rule for district candidates, this is an informal but highly institutionalised rule that was established in the 1950s by all parties, explaining the strong incumbency effect in the German system. While this principle guarantees the representation of experienced incumbents, it poses a special obstacle for young candidates who are running for the first time. Participant observation, however, revealed that the incumbency rule can sometimes be sacrificed in the face of other quotas and rules. In the left-wing parties – the SPD, the Greens and the Left Party – gender quotas have had priority over the incumbency rule. Nevertheless, exceptions from the ‘incumbents first’ rule are otherwise rare.

The Green Party, however, has recently established a newcomer quota as a correction mechanism to the incumbency rule. In some state organisations (e.g. Lower Saxony and Berlin), this newcomer quota has already been formally adopted in the party statutes; in other states (e.g. Brandenburg), it is currently being debated. According to this rule, one of every three consecutive slots has to be guaranteed for a candidate who has not yet been a member of a professionalised parliament at the state, federal or European level. The main goal is to increase the representation of young people in parliament. The Left Party has made steps in this direction as well: informally, the party guarantees two winnable spots on the state party list to candidates under the age of 35 in some state party organisations (e.g. in Brandenburg).

Another very important – and still heavily discussed – regulation is the gender quota. In 1985, formal gender quotas were first introduced by the Greens. Since then, there have been controversial discussions in all other political parties and a contagion effect (Matland and Studlar, 1996) has been observed. Today, the Greens and the Left Party have a formal 50% quota with a ‘zipper system’ (alternating between women and men). The quota applied by the SPD varied between 40% and 50% between states, but in December 2011, was formally changed to 50%. Since the 1990s, the CDU has employed a ‘soft’ one-third quota for electoral lists if there are enough qualified female candidates. The CSU and FDP have neither a formal nor an informal gender quota, although there are intra-party debates regarding the nomination of women. The main argument in the FDP against formal gender quotas is that they violate free competition and individual equality of opportunity.

Traditionally, regional quotas – the allocation of party list positions to territorial party organisations – have played a very important role for the composition of party lists. In most West German states, a very strict and fixed system for this allocation was established in the early years of the Federal Republic (Kitzinger, 1960; Mintzel, 1980) – although it was never formally written down in the party statutes. In practice, this means that in most states, the two large parties do not debate the composition of the entire list at the nominating convention. Instead, slots on the party list are allocated to regions. In each region, a separate selection process for ‘their’ slots takes place. The state party board then simply merges the different regional lists, generally without altering the positions.

The allocation of the slots for each region is mostly based on the share of party members in the different regions. For instance, in 1949, the SPD introduced a strict regional quota system in North Rhine-Westphalia for the four regional party branches in the state by dividing the list into blocks of 10 places (Kitzinger, 1960: 33). Based on the number of party members in the different regions, today, there is a 4:3:2:1 formula, giving four slots to the region of Western Westphalia, three places to Lower Rhine, two places to Middle Rhine and one to the smallest district of Eastern Westphalia in each block of 10 places. The SPD, the CDU and the CSU use these fixed regional quotas in all western states, even in states with a small district magnitude (e.g. in 2009 in Rhineland-Palatinate: 16 list mandates).

Interestingly, these informal regional quotas have not appeared in the eastern states yet. However, the interviews revealed that the smaller regions feel disadvantaged and are regularly under-represented on the state party lists since their candidates lose against candidates of the other regions. As a consequence, at least in some eastern state party organisations of the CDU, the SPD and Left Party, informal rules are becoming institutionalised: ‘We have now agreed that each region gets at least one safe slot on the state party list in order to guarantee the representation of all regions in the parliament’ (interview with a state party official of the Left Party in the state of Saxony-Anhalt). For the FDP, informal rules of proportionality for the different regions are the most important rules during the process of candidate selection for the state party lists. In contrast, regional quotas are less institutionalised and less important for the Greens.

Other quotas and rules for composing the state party lists concern intra-party associations and factions. Traditionally, intra-party associations played an important role in nomination processes, predominantly within the catch-all parties of the SPD, the CDU and the CSU. Although their influence has decreased, in states with a larger district magnitude, the youth organisations and representatives of the right and left wings of the parties are guaranteed safe spots on the state party lists. The informal quotas for the women’s associations, in contrast, have receded with the rise of formal gender quotas.

Within the Green Party, quotas have not been established for the two dominant factions, but in the course of time and after heavy debate, different informal rules have been established in some state organisations. For instance, for the female spots on the state party list of Baden-Wurttemberg, the two factions have agreed that each faction gets half of the winnable spots: ‘“Four for four” is the solution: four candidates, two of each faction, run for the first two spots, and the losers then will get the next two female spots uncontested’ (interview with a state party official of the Green Party in the state of Baden-Wurttemberg).

These informal rules are, however, always subject to change and are generally heavily discussed ahead of the nomination conferences. In contrast, the Left Party has not established informal rules regarding different factions and intra-party associations. Instead, one faction typically dominates candidate selection, depending on the state. Within the Liberal Party, intra-party associations and factions have neither formal nor informal quotas.

In the early years of the Federal Republic, interest groups and associations had guaranteed slots on state party lists, a legacy of the Weimar Republic. However, the direct influence of these associations on the composition of the state party lists diminished very quickly. Today, parties do not have formal or informal quotas for interest groups. Nevertheless, they still try to nominate members of the most important groups and associations on their party lists. The strongest link can be observed between the SPD and the trade unions. The CDU and CSU pay attention to the representation of the farmers’ association and business associations; the Liberals to business associations and occupational groups like the self-employed; and the Greens to migrants and environmental and peace associations.

Overall, the analysis shows that German parties adopt a broad range of rules and quotas, which strongly interact with each other. From the party’s perspective, this affects the process of list-building and intra-party competition:

There is a dense net of factors such as regional quotas, local quotas, gender quotas, the influence of intra-party associations. And these factors are all interlinked with each other. Therefore, I use at least five columns with listings due to the quotas and then I try to integrate them. The whole process takes more than one year. (Interview with person in charge of preparing the list for the SPD of North Rhine-Westphalia)

Due to this complex and difficult process, small boards make proposals to the nominating conventions within the large parties (Zeuner, 1970: 158). Generally, no or only minor modifications are made in order not to jeopardise the highly fragile end product. Due to fewer quotas and rules, the three smaller parties experience more competition for specific spots at the nomination conferences.

Considering the large number of quotas and rules, together with the limited number of safe spots on the party lists, it is not surprising that different regulations often come into conflict. In these cases, the party has to decide which rule or quota has priority over the others. For instance, at the state list nomination convention of the Left Party in the state of Hesse for the 2009 federal elections, only male incumbents ran again. There was an intensive debate about which quota had priority: the incumbency rule, which meant that only male candidates would be ranked on the top slots, or the 50% gender quota, which would place some female newcomers above male incumbents. The gender quota was given priority, indicating that in cases of conflict, formal quotas tend to prevail over informal rules.

Typology of party quotas and rules

The analysis reveals that political parties in Germany adopt a broad range of voluntary party rules and regulations to guarantee group representation. The question arises, in turn, as to how these different types and forms of quotas and rules can be theorised and explained. This article argues that the universe of group representation can be categorised along three central dimensions: (1) the type of group; (2) the mechanism of preferential treatment; and (3) the type of party quota.

Type of party quota

Party quotas are – next to reserved seats and legislative quotas – one of the three main types of gender quota policies (Dahlerup, 2006; Krook, 2009). Yet, existing research defines party quotas as being written into party statutes. This definition restricts empirical analysis to formal quotas by excluding informal party quotas and rules, which the German case indicates are central to list-building. Despite not being formally fixed in the party statutes, regional quotas, for example, have been applied for decades and are governed by very detailed regulations. Therefore, any definition of party quotas must be expanded to include informal quotas and rules as well.

These regulations can be compared along two dimensions: the degree of formalisation and the degree of institutionalisation (see Siavelis and Morgenstern, 2008). Formalisation is distinguished with regard to the fact of being written down – or not – in party statutes. Institutionalisation varies to the degree that rules and quotas are routinely accepted and applied by a political party. Whereas highly institutionalised quotas and rules have been in place for a long time, those that are less institutionalised are regularly challenged and heavily debated within parties.

These two dimensions generate three ideal types of party quotas (see Table 1): formal, highly institutionalised party quotas; formal, little institutionalised party quotas; and informal, highly institutionalised party quotas. The fourth combination – informal and non-institutionalised – is not properly conceptualised as a party quota, but rather as more ad hoc arrangements. An example is the arrangement discussed earlier between the two factions of the Greens to share the female spots on the state party list in Baden-Wurttemberg. Such regulations may, however, become institutionalised and/or formalised quotas in the future.

OPEN IN VIEWER

Table 1.

Typology of party quotas.

Degree of Formalisation	Degree of Institutionalisation
	High	Low
Formal	Type 1: Formal, highly institutionalised quotas	Type 2: Formal, lowly institutionalised quotas
Informal	Type 3: Informal, highly institutionalised quotas	Type 4: No quota, but ad hoc arrangement

Patterns in the German universe of group representation

Using this scheme, it becomes clear that informal, highly institutionalised quotas and rules predominate within German political parties. Among the most important are the two informal priority rules for incumbents and district candidates and the regional quotas in the western states. Formal – but, so far, only partially institutionalised – quotas have only recently been adopted by some parties with the introduction of gender quotas and the newcomer quota. ⁴

The analysis also reveals clear differences between the six political parties. On the one hand, there are differences by party type. The catch-all parties of the SPD, the CDU and the CSU have established informal rules for different intra-party groups and factions – and, to a lesser extent, extra-party interest groups – instead of social groups. However, as discussed earlier, intra-party groups and factions aim to represent certain social groups and interests and thus are often functionally equivalent to quotas for social groups. The major parties also have significantly more quotas and rules than the smaller parties. This can be explained by the fact that the smaller parties have too few safe positions on their lists to accommodate more complex quotas. The same effect holds for the major parties in the states with lower district magnitudes.

On the other hand, there are clear differences by party ideology in terms of the groups that parties want to represent. As in other countries, left-wing parties (the Greens, the SPD and the Left Party) have introduced formal gender quotas, while conservative parties are against fixed quotas for women. The FDP, in particular, argues against formal (gender) quotas on the grounds that they violate free competition and individual equality of opportunity.

Integrating other groups, as well as informal quotas and rules, into the analysis further focuses attention on interactions between the different types of regulations. Conflicting cases, in particular, are highly relevant for gaining insight into the ranking and importance of specific party quotas and rules. In this respect, it is also worth raising the question of which types of groups are receiving formal quotas and which are not – as well as the role the degree of formalisation and institutionalisation has on party priorities in cases of interference and conflict. Two hypotheses appear plausible. The first is that the strongest and most privileged groups gain formal quotas in the party statutes to guarantee their group interests. The alternative is that parties adopt formal quotas for the weakest groups since their interests have to be safeguarded the most.

The evidence in this study suggests that the German case lends greater credibility to the second hypothesis. Although gender quotas have been formally adopted by four of the German parties, they are regularly questioned in the process of list-building. This occurred mostly when the formal gender quota and a highly institutionalised informal quota (the incumbency rule and regional quota, in particular) interfered with one other. In these cases, the gender quota was given priority only because it is formally adopted in the party statutes. Similarly, one of the central goals of the formal adoption of the newcomer rule was to change the incumbency rule, one of the strongest traditional informal rules.

Hence, the study of the interactions between these informal and formal regulations reveals which groups are privileged by political parties. But it also points to the important questions of whether and how some individuals are privileged over others. In doing so, it focuses attention on the implications of these multiple quotas for the individual aspirants and their chances of success in the process of candidate selection. Referring to the research on intersectionality (Hughes, 2011), it can be assumed that those who belong to more than one sought-after group are more likely to be selected as a candidate than those who fulfil only one or no quota.

Conclusion

The main aim of this article has been to develop a broader perspective on the universe of measures for group representation. At the outset, the article argues that current research on electoral quotas is lacking in three central aspects. First, the focus is predominantly on quotas for women and minority ethnic groups while other social groups are hardly researched. Second, there are few studies that analyse quotas comparatively and focus on interaction effects and conflicting cases to learn more about the hierarchy and competition between different quotas. The third limitation is that research is largely confined to formal quotas while informal rules and quotas are excluded.

To expand the research agenda, the current study maps and compares the informal and formal rules and quotas applied by the German political parties in candidate selection processes. The results reveal that broadening the perspective advances research in three ways. First, parties in Germany adopt a broad range of party rules to guarantee group representation for social groups, as well as intra-party groups and groups of candidates, which differ systematically by party type and ideology. Second, different quotas and rules regularly conflict in the process of list-building, requiring parties to decide which rule or quota has priority over the others – in turn, providing important insights into quota implementation, as well as the relative ranking and importance of specific quotas and rules. Third, the German case clearly shows a predominance of informal quotas and rules, emphasising the importance of having a neo-institutionalist perspective and integrating informal institutionalised quotas into the analysis. In contrast, formal party quotas are a rather new phenomenon adopted mainly by the left-wing parties. The resulting typology of groups and quotas, while developed in relation to the German case, may thus serve as a starting point for future comparative research.