Women’s Underrepresentation in German Politics—Reforms and Reform Needs Based on the Local Level ()
1. Introduction
The resignation of Chancellor Angela Merkel (CDU) marks the end of an era that has also changed Germany. Significantly programmatically modernized, women were no longer exclusively granted the role of housewives or the role as additional earners, but through a strong expansion of childcare, the design of parental allowance and women’s quotas in parts of the economy, the career and advancement opportunities of women were somewhat improved. Particularly, the CDU has tried to improve its image vis-à-vis its voters, who were once fonder of it than the SPD (Wiliarty, 2010: p. 21). This descriptive and partly substantial representation of women’s interests by Merkel, von der Leyen and Kramp-Karrenbauer has demonstrated that power in Germany does not always have to be male (Akyün, 2019).
Nonetheless, Angela Merkel has paid little attention to increasing or at least adhering to the quorum (soft quota) for women in the CDU. In the longest term of office of a German Chancellor, the CDU parliamentary group in the House of Representatives did not even reach the already very low target of 33% women, and Angela Merkel has long remained silent on this (Ahrens & Lang, 2021: p. 16). Even after the 2021 House of Representatives elections, the CDU again remain well below its own target (23.8% proportion of women) and central leadership positions were also exclusively awarded to men. However, this backlash did not negatively impact women’s representation in the House of Representatives generally, as the mostly male CDU deputies had to free up their seats for Green female candidates who repeatedly exceeded their quota of 50% (cf. German House of Representatives 2021). Still, the representation of women in the House of Representatives, with 34.7%, remains only one third, which has been true for the parliaments in the federal, state, and large cities for several decades (Holtkamp et al., 2011).
This constancy at least speaks against a general backlash thesis, after a phase of progressive dominance and post-materialist change, among other things through reactions to increasing immigration, a stronger turn to authoritarian and traditional images of women can be observed (Inglehart & Norris, 2016). For some American feminists, Germany compared to the USA, under the Merkel government is already regarded as a utopia, a utopia that has become reality in Germany (Wiliarty, 2010: p. 33). Within the party, the CDU was proud that the first woman was made chancellor and thus legitimized the party’s underrepresentation of women (Holtkamp & Schnittke, 2010). These times are now over and the desire for a correct quota is growing among party women after the election fiasco of the CDU in 2021 (Birnbaum & Fiedler, 2021). For example, there is an increasing demand for the rapid introduction of a 40% quota for women, whereby the top jobs in the party and parliamentary group have already been divided among men, so that it is likely to be more difficult than before to enforce a binding quota for women. Generally, resistance to binding quotas also seems to be great among younger women in the CDU, while more experienced women are more open to the quota (Ahrens et al., 2020: p. 42). It can be seen, especially in eastern Germany that the proportion of women in the CDU has fallen extremely in some cases. In Saxony-Anhalt, the proportion of women in the CDU parliamentary group fell from 32.1% in 2000 to less than 7% in 2020 (CDU, 2021: p. 26). In Thuringia, too, the proportion of women has almost halved in the same period (2000: 18.4%; 2020: 9.5%). Only in Brandenburg and Bremen have the CDU parliamentary groups reached the quorum of 33% in 2020.
The reasons for the under-representation of women are relatively easy to name: Very few parties in Germany have a quota of 50%, which they also comply with, making sure that when drawing up direct candidacies in personalized voting rights, not only men secure the best constituencies of their party in order to get a direct mandate past any quota regulations.1 In this regard, the one-third women representation in Germany and some other European countries is not a mysterious glass ceiling, rather actually said, is the inadequate institutional solutions (quotas and suffrage) from the men and incumbents regularly in the majority in parties and parliamentary positions that stands to benefit from such changes.
The question now remains how science should react to this constant under-representation? Should it resign, re-legitimize the inequality, or should it make a clear case for institutional reforms to reduce this social inequality. Regarding feminist political science, the point of view is clear: it should be followed by other countries that have overtaken Germany after the standstill since the 1990s and have achieved at least a proportion of women of over 40% through legal candidacy quotas with consistent sanctions (Dahlerup, 2018; Ahrens et al., 2020). For many years, the proportion of states with quotas has been increasing and after they have been circumvented more frequently, the sanctions have also become stronger and more effective (Norris & Dahlerup, 2015). Today, 28 states already belong to the club of parliaments with at least 40% women. Germany is currently still in 42nd place in this international ranking (cf. IPU, 2021), after it was still in 9th place in terms of women’s representation in the 1990s and 20th in 2008 (Holtkamp & Schnittke, 2010).
2. Causes of Under-Representation from a Theoretical Perspective
To examine parliamentary representation of women from a focused perspective, explanatory approaches of international comparative nomination research and rational choice institutionalism (Norris, 2004), which until 2010 were only slightly applied to German parliaments (Holtkamp & Schnittke, 2010), are particularly suitable. meanwhile, this explanatory approach is also standard or “secured knowledge” in German-language gender research, and institutional solutions are preferred to increase women’s representation (Bieber, 2021; Ahrens et al., 2020).2
From the perspective of rational choice institutionalism, which is subsequently combined with a new critical rational choice approach (Valdini, 2019), the distribution of mandates in parliaments can be understood as the result of a market process in which voters select from the candidate offer provided to them by parties. However, the decision-making activities of voters, parties and candidates are subject to special institutions (suffrage, quotas, etc.), which can vary considerably according to federal states and parties.
The behaviour of individuals and collective participants are dependent on Institutions. They provide incentives for citizens to make themselves available as candidates for an election, influence the strategic orientation of parties in the preparation of lists and finally, also shape the voting behaviour of the population. Central to the declaration of women’s representation is the behaviour of the parties as the essential gatekeepers.
From the perspective of rational choice institutionalism, the participants basically adapt a rational behaviour with a benefit-maximizing and risk-minimizing orientation, institutional configurations can be found and actively developed in terms of design, which can lead to a parity parliamentary representation in the shortest possible time (“fast track”). With sufficient sanctions and incentives, legal candidate quotas can motivate the parties to draw up equal lists of candidates, which are then confirmed by the electorate in pure proportional representation with rigid lists (Norris, 2004: p. 258). On the assumption that it is an essential goal of the politician and parties to be re-elected, they can be motivated relatively easily to a parity representation if, for example, the lists are not admitted to the election if they fall below the binding proportion of women of 50%, as is the case, for example, for French municipal electoral law (with exceptions for small municipalities). In the case of proportional representation with rigid lists, this equal nomination behaviour of the parties then determines the composition of parliament.
With these quotas of female candidates anchored in the electoral laws, a parity representation has been achieved in the French municipal parliaments in a very short time, and quite a few countries have overtaken the German House of Representatives regarding women’s representation with these legal quotas of female candidates. Thus, the benefit of this explanatory model, which can be derived from rational choice institutionalism, lies in the fact that it sparingly explains political under-representation and leads to clear recommendations for political practice (Holtkamp, 2017: pp. 39-40). However, the model is to be understood more as a heuristic. In empirical analyses, the institutions in heuristics, such as actor-centred institutionalism (Scharpf, 2000), are often supplemented by other explanatory variables to be considered, and the influence of institutions on actor action is classified as contingent.3 Pure proportional representation alone, without statutory candidate quotas, provides on average a significantly higher representation of women. If, due to this right to vote, parties set up a so-called candidate’s package centrally, then they will pay much more attention to socially balanced offers (“ticket balancing”) than if only one or two candidacies are put forward in small constituencies for direct mandates (Norris, 2004: p. 189; Hennl & Kaiser, 2008: p. 328).
Additionally, women’s quota can be applied to all candidacies in this way and the predominantly male incumbents with direct mandates can no longer override internal party or legal quotas. The “dual situations” given by majority voting in a constituency make it more difficult for women to assert themselves, because the power centres of the parties are usually controlled by “old-boy networks” already at the level of local associations or local associations and district associations, in which the chairmen usually play the central role. For example, in the CDU in Baden-Württemberg, 7.3% of the district chairmen are women today, just as they were 20 years ago. In the case of the NRW CDU, only minor progress can be observed (5.6% in 2000 to 9.3% in 2020). Among the local association chairmen, this nomination voting behaviour has opened up somewhat more (BW 17.4% and NRW 15.1%), but the continuing strong under-representation of women in these leadership positions leads to the fact that mostly men structure the nomination behaviour and also bring themselves into play for positions at all levels of the political system or their constituency nomination as already elected members of the state parliament and House of Representatives as district chairmen. secure.
For municipalities in Baden-Württemberg, it has already been empirically proven that women, once they have arrived in these central positions, ensure a higher representation of women on the municipal election list, from which the understandable conclusion is drawn: “more local female gatekeepers are necessary” (Ruf, 2017: p. 24). In addition, it is easier for a female party leader to address and effectively recruit women as role models for politics (see for Belgian municipalities: Devroe et al., 2020: p. 187). These locally strongly anchored party elites (at least in the two people’s parties SPD and CDU) tend to homosocial co-optation on situations where only very few tickets are awarded. Social networks will then include those that are similar. The familiarity in dealing with one’s own gender reduces empirically proven uncertainty. This creates trust, which is the central functional prerequisite for exchange networks, which often bridge formal organizational rules for the purpose of exercising power and maximizing common benefits (Holtkamp & Schnittke, 2010).
A special focus should be placed in the analysis on the interests of men (the majority society) in the local party elite (cf. on this rational choice perspective: Valdini, 2019), because through their over-representation in small informal pre-decision groups, they can decisively determine the candidacies (Holtkamp & Schnittke, 2010) and in some cases also consciously promote women if they apply for bad positions, to enable them to concentrate on more interesting positions. That means also that they can allow a certain amount diversity or even just suggest it and continue to make a career as a man or at least secure the previous positions of power (Holtkamp et al., 2020; cf. in general: Valdini, 2019), as will be shown by the example of the mayoral candidacies.
3. Municipalities as Experimental Construction Sites —Personalized Elections
While discussions about reform backlog and excessively strong forces of perseverance in Germany dominated the national level in 1980, this obviously does not apply to local institutional policy. Of all the levels of government in Germany, the municipalities are exposed to the strongest reform efforts and would also be, for example, the level for which parity laws could most likely be enforced with less resistance. While local policy research in the 1980s propagated a reform of politics from below and classified the municipalities as “experimental construction sites”, the municipal reforms since the 1990s have largely come from the federal states, whereby these reforms were implemented throughout Germany through rapid diffusion (cf. Holtkamp, 2007). In the 1990s, for example, the direct election of Mayors prevailed nationwide, while in the city-states governing Mayors continues to be elected by parliament, because here the parliaments vote on their own competences in the upgrading of governing Mayors through the “consecrations” of direct elections (one-level game).
It is obviously easier for state politicians, through fundamental constitutional reforms to campaign for more participation at the municipal level, for example, and thus to limit the scope for action of the municipal parliamentarians (two-level game), than to implement such reforms in the state constitutions at the expense of their competences. In addition, municipal constitutional reforms, in contrast to constitutional reforms at higher levels, do not require qualified majorities in the state parliament. Overall, the decision-making rules of local institutional policy entail significantly lower hurdles than is normally expected in constitutional and electoral reforms. For example, new personal voting systems have been introduced (a significant exception for cumulation and variegating is North Rhine-Westphalia), the empirical results of which can be investigated in the municipal experimental construction sites.
It should not have been a coincidence that before the parity laws for the state parliaments, the municipal level was to be moved to parity in target regulations in Baden-Württemberg, Rhineland-Palatinate, and Hesse (cf. Land Brandenburg, 2019). Similarly, in France, this two-level game was used and first imposed the toughest sanctions on the municipal parties (non-admission of the election proposals), while the sanctions for the higher levels were highly moderate (low fines: cf. Valdini, 2019: p. 107) and were only sharpened when it became clear that the party leaders were not impressed by this and willingly accepted smaller financial penalties, to protect themselves and the predominantly male incumbents.
Although the parity laws at the state level have so far been overturned by courts in Germany, the extensive under-representation of the centre-right parties will likely lead a change in jurisprudence in the long term, especially since the CDU is no longer represented in the federal government. Due to the fact that the legal quota can be applied directly to all candidacies in pure proportional representation, which includes in particular cumulation and variegating, the greatest successes in a short time would also be expected at this level (Holtkamp & Schnittke, 2010), in contrast to, for example, the parity legislation in the Brandenburg state parliament (which ultimately failed in court), in which the direct mandates were hardly regulated; probably also to motivate the mostly male incumbents of the SPD, who have only moved in via direct mandates, to agree or to protect their interests. Also, regarding the protection of men in the majority factions of the state parliaments, a consistent parity legislation should initially be easier for the municipalities in the two-level game to enforce.4
There is a consensus in gender research that there is currently the greatest need for research and the greatest research gaps at the municipal level in Germany (Fuchs, 2020: p. 251; Najemnik, 2021). The problem, however, is that uniform statistics can only be used for large cities (Holtkamp et al., 2017), while for municipalities with less than 100,000 inhabitants, in which the largest proportion of Germans live (over 90%), there are no uniform statistics, so that the council members would have to be collected individually from the websites of the more than 10,000 municipalities should one want to arrive at reliable statements regarding women’s representation beyond the big cities.
3.1. Cumulate and Variegate
The personalized proportional representation system, as it is applied, for example, in the council elections in NRW, combines elements of majority and proportional representation. Here, the voter in NRW casts only one vote, but at the same time for the direct candidate in his Single-constituency and is counted for the rigid list of the party of his choice. The concrete personnel composition of the parliamentary groups is then initially based on the successful direct candidacies. If a party has won a share of the vote that entitles it to more council members than the direct mandates won, further council mandates are awarded in the order on the rigid reserve list.
Refers to the election of the city council. Separately, the mayor is directly elected by the people German gender research has traditionally assumed that a patriarchal political culture dominates and that citizens therefore discriminate against or do not vote for female politicians if they are given the opportunity to do so via the right to vote (Hoecker, 1998: p. 396; Hoecker, 2008). Cumulating and variegating has always been the prime example of this, while proportional representation with closed lists was recommended, where the voter cannot determine a person. If, on the other hand, one proceeds from the rational choice approach, such as political economy, then women are not disadvantaged by this right to vote, but women’s representation increases with the introduction of cumulation and variegating (Holtkamp et al., 2013).5
The misjudgements of proportional representation with preferential votes by German gender research are because the original placement of women on the party lists is not considered enough (cf. Holuscha, 1999: p. 307). The seats were later included, but it was still concluded that women were elected down by the voters (Magin, 2011: p. 49) and thus allegedly the representation of women was reduced by this right to vote (51). A recent study also concludes that under cumulation and variegating, the electorate systematically votes strongly downwards (Bieber & Wingerter, 2020: p. 18), and it should therefore be reformed. For example, in new overview works of gender research for Germany, it is already presented as the state of research that men are more likely than women to benefit from cumulation regarding their representation in municipalities and city states (Ahrens et al., 2020: p. 105, 110).
However, this positioning is not correct. Bieber and Wingerter (2020) claim with reference to the literature that this may be different in large cities and overlook several German studies that have long proven the positive effect for smaller municipalities (Holtkamp et al., 2013). In view of the empirical findings of international gender research, it is also unlikely that a strong patriarchal political culture in Germany has continued to be maintained and firmly established on the part of the electorate over the years. Accordingly, forms of open discrimination have been increasingly frowned upon since the 1990s and Germany’s political culture is considered particularly egalitarian regarding attitudes towards female politicians (Inglehart, Norris, & Welzel, 2002; Davidson-Schmich, 2006: p. 213).
This leads to the conclusion that the results of German gender research are more likely to be due to methodological problems or artifacts. Thus, the placement on the party lists was considered, but not that these placements generate very different attentions among the voters. For example, the last place is a relatively good placement and more often leads to the fact that many voters vote for this candidacy, who then often move from the last seat into the municipal parliament. The same applies to the first three places. The visibility of the places means that they are chosen by many voters significantly more than if you start from ninth or tenth place.
This is best explained from the rational choice perspective. The voters are trying to reduce information costs, also because they do not even know many of the candidates. Then it is quite rational to concentrate first on the first places and then look at the last places for correction, while saving the information costs for all other candidates, especially since local elections are more often considered less important from the point of view of the civil society, as shown by the extremely declining municipal voter turnout (compared to the federal government) in Germany compared to the 1970s. It is, from the point of view of political economy, a very similar effect to the placement of products on the shelves of supermarkets. The cheap products are usually not found at viewing height, but usually at the bottom, because it is known that customers on average first look at the products at viewing height and often find what they are looking for, to then move on to the next purchase wishes quickly. This saves the customer’s time with a barely manageable offer, as is also the case with the candidates list with cumulation, only that the voters often start at the top of the ballot paper (as seen on this argument: Jankowski, 2017).
If these effects are taken into account and kept constant in regression analyses, all studies from the rational choice perspective for Germany have found on average no negative discriminatory voting behaviour against female candidates (Mechtel, 2014 ; Friedhoff, Holtkamp, & Wiechmann, 2016; cf. Jankowski, 2017 for the city-states). On the contrary, Mario Mechtel shows that in the Baden-Württemberg municipal elections, women have a better chance of being elected to the top than men.
Even if there may be discrimination in smaller German cities, the effect is very weak (Friedhoff, Holtkamp, & Wiechmann, 2016), which is certainly interesting for political cultural research, but can only marginally influence women’s representation. Even in countries such as Poland, where this discrimination is more pronounced, the parties can neutralize this effect if only one woman is placed in first place (Jankowski, 2017), as the Greens in Germany consistently practice in accordance with their statutes. The expected positive effects of proportional representation on party offerings are greater. These provide incentives for ticket balancing and rather enable the application of quotas, as has already been shown in the theoretical explanations. Thus, even in a direct comparison of the NRW cities with personalized proportional representation with the medium-sized and large municipalities of other federal states with proportional representation, it turns out that to a strong and highly significant extent, even with third-party variable control, the representation of women in the city council in NRW is lower (Holtkamp et al., 2013; see very similar: Fortin-Rittberger et al., 2017: p. 62f).
Proportional representation with preferential votes is also suitable for state elections and could, as in Hamburg and Bremen, be enforced by referendum or their threat, both of which are probably not coincidentally at the same time by far the leader in the proportion of women in the state parliaments (36.9%; 43.9%). The abolition of direct mandates under proportional representation would ultimately significantly increase the proportion of women in parliaments. The existing party quotas would then be related to all mandates, just as with a legal candidate quota, and could no longer be undermined by the fact that mostly men prevail in the district associations. In the 2013 House of Representatives election, for example, only 63 out of 299 constituencies were won by women—a quota of 21 percent (Hellmann, 2016; see detailed Bieber, 2022). In the current House of Representatives election, women also won only 26.1 percent of all constituencies directly, while they have a share of 40.7 percent of the successful state list places (BPB, 2021). So you can see that the 40% limit could be reached with proportional representation, even if a parity law at the higher levels were waived for the time being. For the municipalities, on the other hand, the parity law is probably rather indispensable if one wants to achieve gender parity in the municipal council beyond the big cities, because the proportion of women here is significantly lower (Holtkamp & Schnittke, 2010).
3.2. Direct Elections of Mayors
Especially in central political leadership positions, women are even more underrepresented in Germany than in parliaments. This applies particularly to the level of municipalities and federal states. Although it is assumed in the mainstream of German political science that this is now more history even with the longest reigning female chancellor in Germany, up to and including statements that “there can be no question of higher positions being taken less often by women than by men” (Höhne, 2019: p. 15). But these statements are already irritating because they deviate seriously from most empirical research results. In Prime ministers in Germany, for example, the under-representation of women is very pronounced. In the long history of the 16 federal states, there were only six women among the Prime ministers and (indirectly elected) governing Mayors until November 2021.
Women are similarly underrepresented in the municipal mayor’s office, so that direct elections have hardly contributed to an acceleration of women’s representation compared to the indirectly elected prime ministers or governing mayors.6 In a representative survey, mayors were interviewed in 2020. The result was that in the municipalities with more than 20,000 inhabitants, only 6% of the mayors were women (Mahler Walther & Lukoschat, 2020). Compared to previous surveys, the proportion of women has not grown. Only in the smallest municipalities, where this office is often exercised only on a voluntary basis, is the proportion of women significantly higher. Almost every second female mayor works on a voluntary basis, while only a quarter of men hold this position on a voluntary basis (Mahler Walther & Lukoschat, 2020: p. 7; Najemnik, 2021: p. 152).
Across Europe, the proportion of women among Mayors according to a study is around 17% (see Table 1), with Germany continuously below this average (cf. Riethmüller, 2021). According to the evaluation of current data for the year 2021, Germany remains below 10% with an EU average of 17.2%. This puts Germany in 21st place out of 27 countries in the EU comparison.
Based on the few Mayor Interior Election Studies that deal with the causes of under-representation, the state of research is even more deficient than in local parliaments. The German mainstream of local political research usually explains the under-representation of women in the Mayor’s Office by the fact that not enough women are interested in the office, or that the electorate female candidacies is reserved (e.B. Wehling, 2000: p. 205). In addition, recent studies have also pointed out that women hardly have the time to exercise the Mayor’s Office because of the unequal division of labour in the family (Heinelt, 2018: p. 29). A daring thesis, especially since foreigners can also compete in Mayoral elections and often do so successfully, as for example in Baden-Württemberg (Holtkamp, 2008).
![]()
Table 1. Proportion of women among mayors 1.
Source: Own research and calculations 2021.
But the role of parties as the central gatekeepers is mostly ignored, although it has already been explained for council elections how central the nomination by the parties is (Holtkamp & Schnittke, 2010). Because the candidates offer is unlikely to limit the parties, because in larger municipalities there should always be a qualified candidate. In addition, the well-rewarded full-time office is also likely to be more attractive for women than the voluntary council mandate, which must also be agreed with work and family (Eder et al., 2016: p. 381). The Electorate is also unlikely to discriminate strongly against women during the change in values.
But nomination processes of Mayors by parties and their subsequent election have hardly been studied by the mainstream of local political research. The central result of these studies is usually that there is a pronounced incumbent bonus (Gehne, 2008; Holzwarth, 2016), whereby the gender dimension was often ignored here. In the few studies on female mayors from German gender research, which are also very rare throughout Europe compared to the USA (cf. e.B. Devroe et al., 2020: p. 171), the role of the parties is considered much more strongly, and the selection processes seem to proceed as the rational-choice approach to majority voting would suggest.
In her qualitative study of female mayors in Germany, Anja Scholz (2004) first makes it clear that the major parties mainly only field women if a successful candidacy is unlikely. If the party has to run against an incumbent, or the party traditionally achieves poor election results (diaspora), then the male competition is more willing to give priority to women (ibid.: 156; cf. similar Kletzing, 2017). Thus, in these “embarrassment candidacies”, there are often no inner-party opposition candidates (Scholz, 2004: p. 241; cf. also Kletzing & Lukoschat, 2010: p. 42ff; Lukoschat & Belschner, 2014: p. 24f). The consequence of these initial constellations is that female mayoral candidates have significantly worse election chances than male candidates.
The importance of the Mayor’s Office is outstanding for many men and women in the parties and the “main prize” of a long voluntary, usually modestly financed, commitment: The Mayor’s Office, on the other hand, is on average well paid (if it is full-time), offers many creative possibilities and enjoys a high reputation in the party and partly also in the local media, for the Mayors or their Speakers are often the main source for local political research. With the size of the municipality, the attractiveness of the office increases even further, so that the people’s parties, unlike in council elections, do not put up more women in percentage terms with increasing community size. These hypotheses, mainly obtained in qualitative case studies, were quantitatively tested in a contribution to the mayoral elections in Germany using multiple regression models, with the following results (cf. Holtkamp et al., 2020): The major parties that make up the most elected Mayors nominate significantly fewer women than small parties. In addition, women are significantly more likely to be placed in these parties in constellations in which electoral success is less likely (diaspora and candidacy against incumbents). If these adverse constellations are considered by third-party variable control, there was no evidence of discriminatory behaviour on the part of the electorate.
Also, significantly no fewer female incumbents stand for election than their male counterparts, and their election chances ultimately did not differ either. Seen in this way, there is no gender difference in the strong incumbent bonus regarding the tendency to stand for re-election. However, if the absolute number of incumbent candidates is taken into account, it becomes clear that male applicants from the office strongly dominate (male = 180; female: 17) and, because of their high re-election rates, continue the previously existing over-representation of men in the mayor’s office in a path-dependent manner (cf. on this argument already Kjaer, 2010: p. 8).
Particularly noteworthy is the finding that in the two people’s parties (SPD and CDU) the proportion of female candidates does not increase with increasing community size but decreases significantly (unlike in the municipal parliaments). If one takes the market model of the rational choice perspective as a basis here, it becomes clear that this is largely due to the selectors in the party organization. Because it is hardly to be expected that the Electorate discriminates more strongly, especially in large cities (Friedhoff, Holtkamp, & Wiechmann, 2016) or that it is more difficult to recruit a mayoral candidate under 100,000 citizens than in a small town. This indicates that because of the increasing attractiveness of the office with increasing community size, the mostly male party elite secure these outstanding candidacies.
In a recently completed project, in which all mayoral candidacies in North Rhine-Westphalia were surveyed, it could be proven, especially for the CDU, that they put women up significantly more if they must run against the incumbent or the party is in a diaspora situation. Deviating from the usual nomination patterns in NRW, these candidates are addressed to a large extent externally outside the village by the party and thus hardly have their own house power. This differs significantly from the recruitment patterns for male CDU candidacies, who have usually been involved in the party for a very long time at the place of candidacy (Ochsentour) and have already held important leadership positions, from which they have successfully prevailed against intra-party competition, to then compete on more attractive tickets with a better starting position for the local elections (Holtkamp et al., 2022).
This selective effect was even more pronounced in the case of candidacies with a migration background in the CDU and SPD, which were largely only established in a difficult starting position of the party. In addition, these parties that won a large part of the mayoral elections, produced the fewest candidacies with a migration background in percentage terms. This is at least one reason why more than 20% of all mayoral candidacies had a migration background, but only 5% were elected afterwards (Holtkamp, Garske, & Müller, 2022).
While it is easy to develop institutional solutions for a higher representation of women in council elections (combination of legal candidate quota and cumulation and variegating), a quota does not work in pure majority voting. However, as shown, one reason for the strong underrepresentation of women is that so many male incumbents are running and then are usually re-elected by the office bonus until they reach the legal maximum age. As a reform perspective, it is therefore appropriate to limit the possibility of re-election of Mayors to two terms of office, as was introduced many years ago in Italian and Portuguese municipalities, for example. These so-called term limits could make the Mayoral elections more open again, which are no longer (real) elections in many smaller municipalities, as often only the incumbents are running here, since no opposition candidates can be found in the expectedly high re-election rates (Klein & Lüddecke, 2018).
Term limits can increase the representation of women somewhat, because the nomination behaviour of some parties has modernized somewhat in recent years (e.B. the SPD in NRW has already nominated more women over time, while the CDU has put up as many women as in the first mayoral elections in 1999; cf. Holtkamp et al., 2022) and so not (anymore) over many years of incumbent in fact the nomination behaviour of the last century becomes a dependent path. In the case of majority voting, incumbent effect is one of the main explanatory reasons for women’s underrepresentation, as international studies have shown, and therefore the introduction of term limits is also recommended in gender research (Christensen & Bardall, 2014: p. 23). Due to the more open Mayoral elections at term limits, voter turnout is also likely to increase, because the individual vote has more weight (open election outcome) than with a pronounced office bonus. In addition, increased competition and limited tenure could also contribute to less patronage, clientelism and corruption. Term limits can, for example, restrict a very liberal granting of building permits to supporters of one’s own party as a barter for election campaign support. In addition, mayors will not be able to maintain their networks for decades and thus maximize their influence. They are thus easier to control by the local parliaments.7
4. Research Findings
Since gender research is still rather a small niche in male-dominated political science (Wilde & Bomert, 2019: p. 668), such a focus on institutional solutions to increase women’s representation makes sense from our point of view. The mainstream of political science then already represents the other positions and partly aggressively legitimizes this gender gap (Höhne, 2019: p. 15). In the following, the theoretical discussion on the effect of institutions on women’s representation will be outlined first. Then the empirical experiences with these institutions in Germany are touched upon. Particular attention is paid to the municipal level and the personal elections taking place there—i.e. direct elections of mayors and cumulation and variegating in council elections. Here, an overview of the empirical state of research is given, considering current own research results on the descriptive representation of women. Finally, the central results of the empirical analyses are summarized and the developed institutional reform proposals for the promotion of descriptive women’s representation are summarized, which could be applied at all federal levels in Germany.
5. Conclusion and Policy Reform Proposals
Since the 1990s, women’s representation in German politics has stagnated and so Germany has been overtaken by quite a few countries. In the case of parliaments, this is institutionally due to the personalised proportional representation with low female representation in the majority voting component of direct mandates, which is still dominant at many levels, and to the inconsistent voluntary party quotas, which leave the mostly male party elite relatively large scope for action to assert their interests or those of their colleagues (homosocial co-optation) in the nomination process. This is even more pronounced in the mayoral elections than pure majority elections, in which no quota applies. Here, the male party elite without a migration background in the CDU and SPD usually cherry-picks, while candidacies with a poor starting position are more often left to women or migrants to strive for more attractive tickets, especially at the district level, in which the office bonus plays a subordinate role. In order not to be “burned” by (expected) bad mayoral election results in the party, one then applies for the nomination as a member of the state parliament or House of Representatives, in which the district level is dominant in the nomination process, also because there is hardly any office bonus among the voters. Thus, if you wish, you can present yourself as diverse or gender-equitable and continue to make a career as a man of the majority society (cf. fundamental to this Rational Choice research program Valdini, 2019).
So far, too little research has investigated why men in the majority society in the best starting position refrain from applying for attractive political offices or even actively promote women or migrants. The rational choice perspective focuses on the still dominant male party elite and on members of the majority society and, in our view, could further enrich migration and gender research in many fields, because underrepresented groups hardly enforce their concerns in the political institutions past this majority. Here, however, there would certainly also be a qualitative need for research to examine the distribution of positions and motives of candidacies in larger parties, especially at the district level, in a longitudinal manner, which functions as a kind of hub for horizontal and vertical careers in professional politics (cf. in Approaches already Holtkamp & Schnittke, 2010).8
Looking to the future, we have presented three possible institutional reform proposals to increase the descriptive representation of women in German politics: First, the introduction of legal quotas for female candidates initially for the local parliaments with cumulation and variegating (open list election) in the two-level game (members of the state parliament vote on the male career opportunities in the municipalities but are less affected by it themselves). As a result, the number of female candidates is also increasing for higher levels. In the medium term, public pressure on the state and federal parties will increase, which could also ensure a more balanced representation at these levels, such as in France.
Second, a reform of electoral systems towards pure proportional representation with options for preferential votes for individual candidacies, which increases the descriptive representation of women and offers women more opportunities to represent the interests and preferences of women than the personalized proportional representation that has so far dominated at the higher levels. This can be achieved, at least at the state level, for example through referendums or the threat of referendums, as the Bremen and Hamburg state parliaments show. This would provide good conditions for the introduction of effective parity laws at higher levels. An equally conceivable introduction of proportional representation with rigid lists is hardly an option for pragmatic reasons, because the voters do not want to have no influence on the election of the individual candidates but want to see this influence significantly expanded and have already successfully articulated this in referendums in federal states.9
Third, a limit on the term of office of mayors to reduce the path-dependent extreme under-representation of women (and migrants) and at the same time to benefit from the increased number of female candidates during the introduction of a parity law for the municipal parliaments. At the same time, this step would revive local democracy, increase voter turnout and candidate selection, and limit the trend toward local “Sun Kings” through decades of network maintenance.
If these reforms take effect, then the path-dependent women’s under-representation in Germany for decades may soon be history.
Acknowledgements
We would like to thank Dr. Elke Wiechmann particularly for many hints from the Distance learning institute Hagen.
NOTES
1Greens 50%, Left 50%, SPD 40%, CDU Quorum (Soft Quota) 33%; FDP and AFD no proportion.
2See, in addition, the previously dominant German model of the “magic triangle”. Hoecker 2008; see critical Holtkamp, 2017: p. 39).
3At least it corresponds to the rules of attention of the actor-centred Institutionalism first to address the institutions and the standard interests of the actors, which can then be supplemented by other theoretical explanations if the behaviour of the actors cannot be explained by these two variables (cf. for Women’s Representation: Holtkamp & Schnittke, 2010; cf. for Migrant Representation: Hossain et al., 2016).
4”On the SPD state list were 33 women and 39 men. However, the 25 seats to which the SPD is entitled after the election were only filled with direct mandates. Seven of them are women” (cf. RBB, 2019).
5Practical means Cumulating and variegating for the electorate, for example, in a Baden-Württemberg city with 48 council members, that each voter can award 48 votes. First, the ballot paper for a party can be cast unchanged and thus each candidacy on the party list receives one vote. In addition, there is also the possibility to “heap” the votes by cumulating. Up to three votes can be awarded to a single candidate to change the order of candidacies on the party list in the election result. Due to the possibility of variegating, the electorate is not limited to the candidates of a party list, but it is can also individual candidates of other parties are elected. With this right to vote, therefore, the greatest personnel decision-making powers of the electorate exist.
6This is the name of the heads of government in the three city-states of Germany (Berlin, Hamburg, and Bremen).
7Cf. empirically convincing recently proven for French mayors (Leveque, 2020: p. 138, 153).
8See e.B. on the substantial representation höhmann 2021, according to which male deputies are particularly committed to women's issues in small inquiries in the German House of Representatives if their constituency is uncertain to open new voter potential. See, on intersectional representation and the recruitment of young women with a migrant background, who at the same time represent three minorities in the party ('three birds with one stone', if you want to print it out coarsely) and secure the older incumbents more free places under proportional representation without endangering their retention of power in the short term (due to the lack of experience of the new candidates) (Stockemer & Sundström, 2019: p. 382; Celis & Erzeel, 2017: p. 16f.).
9In 2004, a referendum was successfully held in the city-state of Hamburg against the proportional representation in force until then initiated with rigid lists and then later replaced by cumulating and variegating. Finally, proportional representation with rigid lists in Germany was still in the city state of Bremen and was also after referendums 2006 successively through cumulate and panic exchanged.