Skip to main content


Front. Polit. Sci., 29 August 2023
Sec. Elections and Representation
Volume 5 - 2023 |

Power struggles in the German Christian Democrats. The dynamics of three consecutive leadership contests

Uwe Jun* Marius Minas*
  • Trier Institute for Democracy and Party Research (TIDuP), University of Trier, Trier, Germany

In 2021, the German Christian Democrats held three leadership contests. First, the Christian Democratic Union (CDU) had to find a new party leader. While the members tended toward the more conservative candidate, Friedrich Merz, the party elite increasingly spoke out in favor of the more moderate Armin Laschet, who prevailed at a digital party conference. Just a few months later, he was challenged by his counterpart in the CDU's Bavarian sister party (CSU), Markus Söder, over who would be the joint chancellor candidate in the upcoming federal election. Söder was clearly favored by voters as well as the party on the ground, but Laschet found powerful and decisive support in the CDU's party elite. Yet the Christian Democrats surprisingly lost the federal election, mainly because of the unpopularity of its frontrunner. After Laschet announced his retirement, the party prepared the third leadership contest. This time, the party headquarter declared a membership ballot as the mode of decision-making, in which Merz triumphed. As this study clearly points out, each of the three selection modes developed its own dynamic in the matter of influencing factors, qualitatively as well as quantitatively. Process-tracing shows that the various selection processes are interrelated in terms of their dynamics and decision modes.


One of the most important events of 2021 in German politics was unquestionably the drama around the various selection processes for top positions in the Christian Democratic Union (CDU) and Christian Social Union (CSU). These “union parties” form a joint parliamentary group in the Bundestag, and for more than 50 of the last 74 years, they have provided the head of government, which makes the Christian Democrats the main actors in the German party competition since 1945. The selection of a CDU party leader at the beginning of the year, and then of the union parties' electoral leader in the spring, significantly influenced the Bundestag election in September. The example of the CDU offers impressive support for at least three propositions. First, the selection of the party leader can have a considerable impact on the outcome of a subsequent parliamentary election. Second, external shocks provide impetus for changes in the party organization.1 The union parties' defeat in the parliamentary election of September 2021 induced a third internal selection process that year, in which a new system was employed. Third, parties conduct their selection processes with only limited rationality regarding their own objectives. The CDU, which is usually seen as the German party that is most focused on maximizing votes, chose a party leader and electoral leader in an election year who was already very unpopular in the electorate at the time of his election (Niedermayer, 2021b) and who proved to be a “burden” (Schmitt-Beck, 2021) for his party in the election campaign. This study analyses how this could happen and what factors can explain this unusual process.

Our study follows the qualitative concept of Aylott and Bolin (2021), who identified in their conclusion of their edited volume that a research gap with respect to intra-party electoral processes. The authors discussed “the possibility that change in the method of selecting a leader might shift the balance of power within a party, through, for example, enhancing the power of the party on the ground.” Moreover, they reflect, that “[o]ne area that the studies in this volume came only fleetingly near to exploring are the mechanisms of […] process management” and “several observations of the same party over time, […] will take as a long way toward identifying a party's prevailing power structure more generally” (Aylott and Bolin, 2021, p. 239–240). Finally, through case analysis, this study aimed to contribute to a deeper understanding of intra-party selection processes and thus power structures.

Theory, case selection and methods

In our study, we contribute to research into intra-party democratization in general (see Rahat and Hazan, 2001; Cross and Katz, 2013; Cordero and Coller, 2018; Detterbeck, 2018; Ignazi, 2020) and leader selection processes in particular (see, i.e., Cross and Blais, 2012; Cross, 2013; Cross and Pilet, 2014; Pilet and Cross, 2014; Ding, 2015; especially for the German case: Detterbeck and Rohlfing, 2014; Reiser, 2018; Jun and Jakobs, 2021).2 Previous studies suggest that inclusive procedures more often produce inexperienced party leaders (Sandri et al., 2015, p. 105) and, with regard to political elites in general, allow political outsiders and more extreme candidates rather than exclusive selectorates do (Buisseret and Van Weelden, 2020; Seddone and Sandri, 2021, p. 234f). At the same time, “[i]nclusiveness does not lead to higher female representation nor to the selection of younger nominees” (Seddone and Sandri, 2021, p. 236).

We still know too little about the informal dynamics that shape leader selections. Our article seeks to shed light on the “secret garden of politics” (Gallagher, 1988) by describing the real story (Katz and Mair, 1992) behind the scenes of leadership selection processes and examining power dynamics as well as the importance of intra-party networks (Bolin and Aylott, 2021).

The German Christian Democrats' intra-party power struggles in 2020–2021 are an attractive case study in this regard, as we can observe different degrees of inclusiveness in the three separate selection processes within a short time.3 The inclusiveness of the selectorate is a key indicator in the study of intra-party organization (Panebianco, 1988; Cross, 2008, 2013; Kenig, 2009; Cross and Blais, 2012; Von dem Berge and Poguntke, 2017). Consequently, our case study addresses the composition of the selectorate, the distribution and dynamics of power within the party, and the party's electoral performance.

In Germany, party leaders4 are powerful actors in politics, meaning that

[t]hey preside over the party executives and are key to senior appointments in the party central office and parliamentary party. With respect to party policies, party chairmen are involved in all important internal decision-making processes at the federal level. Most often, they are leading figures in the party in public office. It is also quite common for party chairmen to head electoral lists (electoral leader). However, there is no automatism here and chairmen sometimes have to accept the choice of another electoral leader with stronger voting appeal (Detterbeck, 2013, p. 273).

We therefore differentiate between the party leader, who is the head of the party's organizational and power structure, and the electoral leader, who is the face of the party in the electoral campaign and aspires to the highest national office in government (Kenig, 2009, p. 241; Cross and Blais, 2012, p. 145; Costa Lobo, 2014). The clear delineation of these individual terms is central to obtaining clear insights into the power dynamics in parties. Both roles involve leadership claims, but they derive their legitimacy and influence from different sources.

What can be clearly stated is that “[l]eaders matter” (Bittner, 2011). As a consequence of the personalization of politics (Karvonen, 2010, p. 4; Rahat and Kenig, 2018), a greater accountability of the leadership with regard to the electoral performance of their party can be generally observed (cf. Bean and Mughan, 1989; Von dem Berge and Poguntke, 2017; Pruysers et al., 2018; Pedersen and Rahat, 2021; Garzia et al., 2022). This may be especially true in an election year like 2021 and in vote-maximizing parties like the German Christian Democrats. Therefore, it is reasonable to consider the selection procedures for both party leaders and electoral leaders alongside each other (see also Astudillo and Detterbeck, 2020).

If we accept that, “since there is little doubt that party leaders are important, their recruitment merits attention” (Marsh, 1993, p. 229), it is worth taking a closer look at these selection processes. In all three, we seek to identify factors that influenced the respective outcomes and which actors played decisive roles. Three processes are presented in their entirety, as they can only be understood both historically and structurally if the respective connections between the three within one year are shown and broken down. It is only in the overall context that it becomes clear why, for the first time in its history, the CDU determined its leader by polling all its members. In this respect, this contribution can trace and explain how greater inclusion of the members came about in the party. As a case study, it shows how the modes of the selection process have influence on its result.

The analysis is largely in line with the concept of party leader selection by Aylott and Bolin (2017, 2021), who convincingly shows that this process can be divided into three phases: gatekeeping, preparation, and decision. Whereas, the gatekeeping phase tells us the official story of the selection process, which means the eligibility and nomination requirements for aspirants and candidates, as well as the composition of the selectorate, the preparation phase and the decision phase shift the focus away from party statutes toward the real story of leader selection. Before determining the nature of the final decision-making, the process of negotiation—in other words, the internal party election campaign that prepares the decision—will be examined in more detail (Aylott and Bolin, 2017, 2021, p. 221ff). As the authors rightly point out, a comprehensive look behind the scenes is necessary to reconstruct events that relate to the party leader selection, which in turn requires a qualitative and rather narrative approach; as such, a reconstruction can hardly be done on the basis of statistical data. To identify and reason about causal relations, we thus employ an inductive style of “causal process tracing” (Blatter and Haverland, 2012, p. 111–118; for a detailed description of this method, see the joint volume of Bennett and Checkel, 2015) to identify these explanatory factors or events. We combine plausible explanatory events, similar in nature, and call them potential factors, as their significance is stronger in combination. Based on our observations and analyses of the process dynamics through media reports and interviews, a high plausibility of the explanatory power of various events can be stated. Our aim, however, is to find empirical evidence for causality if possible. In our understanding, the explanatory power increases when factors are more necessary and sufficient for the final outcome (García-Montoya and Mahoney, 2020). If—and only if—we found empirical evidence through our qualitative approach that a potential factor is critical in its explanation of an unexpected outcome (García-Montoya and Mahoney, 2020, p. 10–16) with spatiotemporal contact (Mahoney, 2021, p. 97), we would speak of a causal mechanism (Blatter and Haverland, 2012; see also Bennett and Checkel, 2015, p. 12). Otherwise, we only state a high plausibility instead of causality for the influence of identified factors on the decision.

Our data consist of a comprehensive media analysis and three expert interviews with high-ranked party members.5 The first interview was with a former CDU party leader, Annegret Kramp-Karrenbauer, whose succession is examined in our contribution. The second was with a long-standing member of the CDU's National Executive Committee, who had deep insights into the selection processes. This person preferred to remain anonymous. The third was with a former CDU vice party chair, Volker Bouffier, who is seen as an “éminence grise” of German Christian Democrats.6 As many details of the processes took place behind the official scenes, only initiated leaders could reveal them. In our opinion, other approaches would not have offered this possibility to reconstruct the “real story.”

Party leader of the CDU


On 10 February 2020, when Kramp-Karrenbauer announced her intention to give up her position as CDU chair, the question of her successor arose.7 Two names were quickly on everyone's lips: Friedrich Merz, the former leader of the parliamentary party at the beginning of the century, who had lost narrowly in 2018 against Kramp-Karrenbauer (see for details Jun and Jakobs, 2021), and Armin Laschet, acting minister-president of North Rhine-Westphalia and chair of the CDU in that state. Ideologically, the contest was between the continuation of chancellor Merkel's path of programmatic modernization toward the political center (Kronenberg, 2020; Wiliarty, 2021), represented by Laschet, and the re-emergence of more traditional values together with the prioritization of a more market-liberal economic policy, represented by Merz.

Furthermore, the party's elites8 preferred a “team solution” (Interview 1). Ideally, there would be no intra-party competition for party leadership. Instead, both candidates should be brought under one roof while forming the new leadership team (SZ 17 February 2021). In this case, a “team solution” did not mean a dual leadership (as, e.g., in the German Green Party or currently in the Social Democratic Party), but rather that one candidate should become the party leader and the other his deputy. However, the attempt to prevent a real competition for the highest party office failed doubly.


Kramp-Karrenbauer, Annegret; former party chair of the Christian Democratic Union; 25 January 2022.

On 18 February, Norbert Röttgen, chair of the Foreign Affairs Committee of the federal parliament, surprisingly announced his intention to run (Interview 1; Alexander, 2021, p. 164f.). He claimed to stand for programmatic renewal and organizational change, with more intra-party democracy to forestall secret backroom deals for the highest party office (FAZ 19 February 2020). Thus, the party executive's plan to convince all applicants to support their intended team solution was scuppered. Laschet saw himself exclusively in the role of party chair. Röttgen continued to insist on his candidacy and looked forward to appearing at the party congress in front of the selectorate. Merz's claims were still alive because of a long-running personal feud with Chancellor Merkel (Interview 3; Alexander, 2021, p. 170ff).


Before the course of events is analyzed, the formal requirements for the selection must be briefly clarified (Jun and Jakobs, 2021).

German party law (§9) stipulates that only the party congress may select the party leader. Although it obliges the parties to guarantee their members the opportunity to participate in the internal decision-making process (§7, §15) the law does not specify how the democratic decision-making process in general, and the selection of party chairs in particular, should be structured.

In our case, however, the CDU party statute (§6), in combination with its rules of procedure, provides satisfactory answers. The primary requirement to run for office is formal party membership. The next hurdle to overcome is the transition from “applicant” to “candidate” status. This can be done in two ways. One is that, prior to the party congress, a written proposal to the federal headquarters can be submitted by an eligible local party branch. The other is that, at the party congress itself, delegates can propose another delegate, themselves or another member who is not a delegate to the party congress.

In accordance with these criteria, three candidates were finally selected from the 16 applicants (Spiegel 6 March 2020): Laschet, Merz, and Röttgen. All others failed to get the necessary support (Interview 1).


After the failed attempt to avoid intra-party disputes and present a joint team, party leader Kramp-Karrenbauer had shot her bolt as steering agent and had to let events take their course.9 Whereas, in 2018 the then party leader Angela Merkel had played an important role by supporting her protégé Kramp-Karrenbauer (Zeit 7 December 2018), Merkel was almost absent in this process. Indeed, it is impossible to identify a steering agent at all.

To understand the result at the party congress, potential factors in various forms can be identified from the ensuing intra-party election campaign, which lasted almost a year. The CDU's protracted internal disputes were partly a result of the COVID-19 pandemic. There was supposed to be barely 3 months between the announcement of Kramp-Karrenbauer's resignation and the election of her successor. Because of the pandemic, however, the party congress scheduled for April was postponed twice and finally scheduled to take place in January 2021—for the first time, in digital form.

The most important circumstance that has to be taken into account is the composition of the selectorate, because the final decision is up to the delegates. Consequently, the intuitive question to ask is as follows: What influences their opinion-forming process and ultimately their voting behavior?.10 A good starting point are the premises of the micro-sociological approach of electoral research, according to which people develop the attitudes that determine voting behavior through social interactions with other people (Schoen, 2014, p. 172). That means political preferences of the family, the circle of friends, and colleagues from work (see also Campbell and Cooper, 1956). This effect is additionally strengthened by social pressure (Lazarsfeld et al., 1960, p. 154f).

Therefore, the first potential factor stems directly from the concept of delegates. According to the party statutes (§28), the party congress

consists of 1,000 delegates from the state branches, [… plus] the delegates from the foreign branches […]. Of the 1,000 delegates of the state branches, 200 are sent in proportion to the number of second votes cast for the individual state lists […] in the last [federal] election […], 800 in proportion to the number of members of the individual state branches [own translation].

In this case, 1,001 delegates were sent to the party congress, representing the party at the regional level, and through the principle of proportional representation: Branches with relatively many members sent more delegates than branches with relatively few. Consequently, as the first potential factor, we took the events of declared support of state branches into account, especially those with high membership.11 Of course, all delegates decide individually and secretly, so that the decision-making process within every state branch has an informal character (Interviews 13). Commitments to one or the other candidate reflect the general tenor among the state branch's members and officials but are no guarantee for the vote of all its delegates (Interview 1). Furthermore, there are other influences on the delegates, which may either reinforce their support for the position of their state branch or lead them to swim against the tide. In some states, like Rhineland-Palatinate and Baden-Wuerttemberg, the regional leaders and the party on the ground stood on different sides—the leaders in favor of Laschet, the majority of party members in favor of Merz (Interview 1).

With 120,928 members, the North Rhine-Westphalian branch had by far the largest number of the CDU's 399,110 members (Niedermayer, 2021a). Being its chair and simultaneously the state's minister-president, Laschet benefitted from the support of this most influential state branch (Figure 1). This was a small disadvantage for his opponents (Interview 2) because they too came from the same state. Also important for Laschet was the overall support of two other large states, Lower Saxony and Hesse. Baden-Württemberg's branch was a little stronger in terms of membership. Like the eastern German ones, it favored Merz. No state branch declared support for Röttgen. The other states remained undecided, which meant that their delegates' decisions varied.


Figure 1. Support of CDU state12 branches13 in the question of party leadership.14


Anonymous; long-standing former member of National Executive Committee of the Christian Democratic Union; 2 February 2023.

In addition to the state branches, the CDU's special organizations may be relevant in the delegates' opinion-forming process. They represent demographic groups inside the party (women, youth, seniors), socioeconomic interests (business and economic), and geographical regions. As regional interests have already been covered through the state branches, we will only consider the influence of the other five special organizations.

The influence and power of these organizations are primarily based on their membership strength (Schönbohm, 1985, p. 229) and the position of their respective leaders within the party (D'Antonio and Werwath, 2012, p. 50ff). Based on membership, the demographic-based organizations clearly overtrump the socio-economic-based ones, as Table 1 illustrates. However, because this selection was also about whether the party will continue on a more centrist and welfare-oriented course (Laschet) or align itself (again) more economically market-liberal, the influence of the economic wing and the social wing of the CDU must be ranked as the highest among all special organizations (Interview 1). D'Antonio and Werwath (2012, p. 51) identify a close connection between the party's special organizations and its state branches, which is particularly evident in this case. The North Rhine-Westphalian branch, which orients itself toward trade union social policy, often opposes the southern branches, which tend to be business and entrepreneur dominated. It was therefore not surprising that the economic wing and the southern state branch of Baden-Wuerttemberg supported Merz (Spiegel 26 February 2020; Spiegel 2 March 2020), while the social wing and the North Rhine-Westphalia branch favored Laschet (t-online 27 February 2020; Hb 11 December 2020).


Table 1. Special organizations within the CDU.

With the generational change in the youth organization since the 1980s, there was also a shift away from the more socio-political positions toward an alliance with the economic wing, which has endured (D'Antonio and Werwath, 2012, p. 52). Therefore, their support for Merz was also no surprise (NZZ 3 November 2020) but should not be overestimated because young members are underrepresented among the delegates (Interview 1). While the Women's Union, which is important for the external perception of the party (Interview 1), tended toward Laschet, just ahead of Röttgen (Hb 9 January 2021), no clear tendencies can be found on the part of the seniors (Watson 14 January 2021). The declared support of a special organization can be seen as the second potential factor in this contest.

Additionally, the support of the CDU's central opinion leaders in the form of high-ranking committees or individual personalities, as the third potential factor, comes into question. While Laschet is seen as an integrative team player, Merz stands for an authoritarian, strong leadership style. The central opinion leaders probably saw greater scope for their own influence with Laschet than with Merz. Accordingly, it is not surprising that many high-ranking officials favored Laschet—he can clearly be seen as the candidate of the party elite (Zeit 15 January 2021).

In the autumn, Merz sensed a party elite plot against him (Interview 1; TS 27 October 2020). The second postponement of the party congress in December was defended by the arguments that an attended party congress could not take place due to COVID-19 and national law would not allow digital elections. However, even a postal vote was not seen by party officials as a viable option. Merz interpreted the decision as a reflection of Laschet's inferior poll ratings (Figure 2) as the main reason for the postponement, as the party's elites, as has been shown, favored Laschet. To avoid an open power struggle in the public and the media, the candidates and the party headquarter finally agreed: The party congress should take place in January—in digital format (Interview 1; Alexander, 2021, p. 330ff).15


Figure 2. Popularity17 of candidates (in %).

In no time at all, the party organization generated a roadmap (, 2021a) for the party's internal election campaign. “CDU Live” was launched, where the three candidates could present themselves exclusively to all members, separately as well as in a kind of a TV debate. Nevertheless, it must be emphasized that, with regard to Laschet's popularity within the party (Figure 2), there was an increase of 10 percentage points shortly before the party congress in January, while support for Merz fell by the same amount. On selection day, therefore, a short and surprising momentum in favor of Laschet could be discerned.


For the first time in the history of the CDU, a party congress had to take place in digital format. Social interactions and communication among the delegates, which also contribute to opinion formation at party congresses, predominantly took place beforehand this time, as we showed earlier. Nevertheless, it is plausible to assume that delegates exchanged messages with each other, albeit with limitations in terms of quantity and intensity compared with face-to-face events.16 Additionally, the contenders for the party chair in 2018 had still been able to convince their audiences through rousing speeches and form opinions through spillover effects among the delegates, in the form of strong applause and small talk. These forms can be decisive in close outcomes, as undecided voters are guided by them. This time, the three candidates delivered their statements in front of technical devices that transported them directly into the living rooms and offices of the delegates—“there was no atmosphere” (Interview 3). Robin Alexander, a German journalist, who has been able to observe the power structures within the CDU very closely over a long time, states in his book:

Merz and Röttgen think that everything is as usual. Only the hall has been moved to the Internet. They are mistaken. There is no hall, not here in the TV studio and not on the Internet. [...] Laschet sees it this way: he does not speak in front of 1,001 delegates, he speaks with them, but with each one individually […]. He also takes something else into account: The overwhelming majority of the delegates are men. They are not cooped up together this time, but are at home, in most cases with their wives (emphasis in the original; own translation; Alexander, 2021, p. 333).


Bouffier, Volker; former vice party chair of the Christian Democratic Union; 8 March 2023.

Fundamentally different ways of speaking to delegates were observed. Laschet began by explicitly greeting families and then moved on without any real content or vision for the future, instead appealing to emotions. He closed by holding his father's miner's badge up to the camera, quoting him: “He said, tell people they can trust you.” The badge, that no one would have been able to recognize at an attended party congress, but whose outlines had been captured clearly by the camera, left a lasting impression on the delegates. This symbolic expression and emotional approach were more impressive than the content and the more or less visionary ideas formulated by Laschet's two competitors (Alexander, 2021, p. 333ff; Laschet, 2021; Merz, 2021; Röttgen, 2021). Importantly, the appearance at the party congress can be seen as the fourth potential factor (Interviews 1, 2). Up to this point, the race was open (Interview 2). A majority of the votes are required for the election. If this is not achieved, a runoff election is held. In this selection process, Laschet's “speech of his lifetime” (Interview 3), through which he was able to win over the necessary delegates, can be seen as the “decisive factor” (Interviews 2, 3).

At this point, nonetheless, we cannot clearly identify a causal mechanism.18 Although this potential factor fulfills the criteria of being critical in its explanation and having spatiotemporal contact with the outcome, we do not find an unexpected outcome in the election of Laschet, as the race was considered open until then. Thus, no clear causality can be demonstrated in our understanding because other potential factors also allow Laschet's victory to be an outcome of this selection process. Röttgen, who received 224 votes in the first round, lost to Laschet (380 votes) and Merz (385 votes). In the subsequent runoff, Laschet profited from Röttgen's defeat, as Röttgen's positions were closer to Laschet's than to Merz's (Interview 2). As Rehmert (2022, p. 1165) found when examining selectors' preferences for party candidates, “selectors follow cues of valence indicating candidates' quality through their commitment to the party or legislative experience,” what also points more in the direction of Laschet than Merz. Accordingly, Laschet (521 votes) prevailed over Merz (466 votes). The result of the digital ballot was legally confirmed by a postal vote by the congress delegates, in which Laschet did even better, receiving 796 votes.

However, this does not conclusively settle the question of leadership. As we saw, the union parties, the CDU and the CSU, run together in national elections—naturally, with only one frontrunner, who is considered the electoral leader during election campaigns (Costa Lobo, 2014) and on becoming chancellor, would rank as the informal highest position of the union parties. This is largely due to their self-understanding as vote-seeking, catch-all parties. The respective party chairs are the natural aspirants for this position.

Electoral leader of Christian Democrats


The starting point of the contest for the position of the CDU/CSU's electoral leader could be found within the conference of German minister-presidents, which included both Laschet (North Rhine-Westphalia) and Söder (Bavaria), and which emerged as a central, albeit informal, decision-making body during the pandemic. In the course of debates within the conference, it quickly became clear that Söder (like the chancellor) advocated strict measures to control the virus; he was continuously building his image as a crisis manager. Laschet embodied a policy style of wait-and-see, deliberation, and looser action. The fact that he chaired the conference during the crisis helped Söder to become a national figure. Accordingly, at official press conferences to explain pandemic measures, the only Christian Democrat at the Chancellor's side was Söder. Moreover, surveys indicate support for his leadership style (Zeit 24 February 2021).

Once Laschet had been elected party leader, the confrontation of the two personalities increasingly came to the fore. According to Interview 3, Laschet missed the chance to publicly declare his candidacy for chancellor immediately after his designation as party chairman, as Söder had hitherto insisted that his place was in Bavaria. Laschet's reasonable wish, however, was to close ranks with the sister party and make a joint announcement (Interview 3).

In fact, there is no formal procedure between the two sister parties to determine their candidate. The section on gatekeeping can therefore be disregarded here. The customary right of party chairs to the chancellor candidacy only exists in the CDU/CSU by archaic mutual agreement (Interview 1; Wiesendahl, 2021, p. 2). The linkage between the CDU's party leader election and the upcoming nomination of the electoral leader is clearly visible. As Laschet's highly competitive election was only a short time earlier and the CDU was in a polling slump in the spring of 2021, the larger party was under pressure to make a strong and united statement during the election campaign and to solemnly crown their candidate for chancellor. A peaceful agreement, however, was not to be this time (Interview 1). In the weeks leading up to Söder's official announcement on 11 April 2021 that he would be available as the CDU/CSU's candidate, there were minor and major digs on his part in Laschet's direction. The chancellor also took it upon herself to criticize Laschet's handling of the pandemic in a prime-time talk show (AW 28 March 2021), which Söder then echoed (Tagesthemen 28 March 2021). This criticism damaged Laschet's campaign (Interview 1). However, the 10-day power struggle that followed Söder's public proclamation of his willingness to run was to mark a watershed in the history of the two sister parties and left a “field of debris” (Interview 1).


In the CDU/CSU, the coronation of the parties' electoral leader—that is, the presentation of a single candidate to the selectorate for its approval—is usually the responsibility of the highest party bodies, especially because the parties are considered to have a representative-democratic hierarchical order (Wiesendahl, 2021, p. 2) and thus have not established direct-democratic forms in their candidate selection processes. At first, Söder fitted into the existing power structure, proclaiming his support for the big sister party (Söder, 2021a). In theory, this settled the question: the CDU's National Executive Committee and presidium agreed on Laschet. Yet Söder's idea of support did not correspond to the usual, representative-democratic practices. He tried to win the backing of the joint federal parliamentary party, various state branches, and the party on the ground (Wiesendahl, 2021, p. 2).19 In an interview, he warned that the question of who should run as an electoral leader should not be decided in a “little backroom” (BR 12 April 2021)—a wording that made the issue into more than just a personnel decision. The highest CDU bodies, duly elected at the party congress, felt compelled to restore their contested position of power (Wiesendahl, 2021, p. 2).

Politicians and parties are driven by a combination of policy influence, vote-seeking, and office-seeking (Müller and Strøm, 1999). Individuals will assign different priorities to these goals, depending partly on their positions within the party structure. When it comes to the question of who should lead the party or lead it into an election, it can be assumed that different politicians therefore evaluate which leader will bring them closer to their respective goal or be most useful to it. The leader may aim to present a unified image of the party (Kam, 2009, p. 21ff; Proksch and Slapin, 2014), through either the integration of different party wings or strict hierarchy. Plausibly, the party in the central office (inter alia the executive committee) wants to maintain its influence on the organization and thus strives for an integrative team player in the leadership position. Members of the party in public office, primarily parliamentarians, should seek, among other things, to organize their re-election in order to have further influence on policymaking and thus seek electorally effective candidates. Moreover, these are closely linked to the interests of the party on the ground, which is primarily interested in having its interests represented in the form of policymaking (Kam, 2009, p. 23f). In our case, these expectations about prioritizations tend to be confirmed.

Söder's significantly superior poll ratings, which he clearly emphasized in his argumentation, were the first potential factor in the selection process (Wiesendahl, 2021, p. 6). As can be clearly seen, Söder did not only become more popular and better known between the candidate selection (April 2021) and the previous party congress (January 2021; (Figure 3), he could boast significantly better poll ratings than his opponent on who was considered more suitable for the chancellor candidacy (Figure 4). This clear lead can be explained in part by the circumstances of pandemic policy, as outlined above (Wiesendahl, 2021, p. 6). Popularity and media impact are especially important in light of the personalization of politics (Karvonen, 2010; Rahat and Kenig, 2018).


Figure 3. Poll ratings:20 (Dis-)satisfaction and recognition (numbers in %).


Figure 4. Suitability21 to run for chancellor (numbers in %).

Nevertheless, due to the hierarchical order within the party and the related representative-democratic habits, high-ranking supporters also play a decisive role in the selection process. The chancellor did not take up the intuitive ex officio option of directing the succession as steering agent. In addition to her criticism of Laschet's COVID-19 management, she did not publicly side with either Laschet or Söder (Interview 2; see also Alexander, 2021, p. 351, 355). Regarding the state branches, the picture is different.

As Figure 5 shows, various state officials were in favor of Söder (Bavaria, Saxony-Anhalt, Saarland). In Rhineland-Palatinate and Saxony, the state leaders prevented a pro-Söder revolt by the branches by promising not to position themselves openly in favor of Laschet (Alexander, 2021, p. 356f). Meanwhile, Laschet knew that his own and the Hesse branches, led by the influential minister-president Volker Bouffier (Interview 1), were behind him. In sum, the party on the ground in most states favored Söder (TS 19 April 2021). While the state branches, through their delegates, were considered a potential factor in the first selection of 2021, with its formally regulated party procedure at the party congress, this influence can no longer be confirmed in this second selection. Nowhere is it stipulated that the state branches must be involved in this decision-making process. Of course, however, their representatives are important actors.


Figure 5. Support of state22 branches in the question of Christian Democratic electoral leadership in the upcoming national election.24

Ultimately, this decision requires decision-makers at the federal level. Accordingly, only two possibilities for reaching an agreement were discussed in public. First, the joint parliamentary group in the Bundestag or the representatives of the so-called Landesgruppen, which represents the federal states in the parliamentary group, could decide. Second, the National Executive Committee of the much larger sister party, the CDU, could decide (Interview 3). Naturally, party officials at the federal level have their own career fate in mind, or at least their influence in the next legislative period, both of which are strongly linked to the outcome of the upcoming federal election, which in turn depends to a large extent on the performance of the electoral leader. Thus, on 13 April, the joint federal parliamentary party met and held a tough debate. For four hours, one MP after another came to the lectern and explained to Laschet face to face why they thought him incapable of winning the election. Audio recordings were leaked to the public (Alexander, 2021, p. 354). The support of major party politicians embodied the second potential factor of this selection process. The momentum now seemed to point clearly in Söder's direction. From a micropolitical perspective, however, Söder's affront to the party elites unleashed political forces he had not anticipated.

Laschet, meanwhile, concentrated on shaping a majority for himself in the CDU's National Executive Committee. Although the party on the ground in many states tended toward Söder, Laschet had his eye on those officials who had voting rights on the National Executive Committee. He was able, for example, to persuade Bernd Althusmann, then CDU state chairman in Lower Saxony, to promise to vote for him on the National Executive

Committee, while the members of his state branch were more pro-Söder (Interview 2). Additionally, in the president of the German federal parliament, Wolfgang Schäuble (CDU), Laschet found an ally whose party-political influence was sufficient to turn the tide (Interview 1). While Schäuble had vehemently supported Merz a few months earlier, this time Laschet benefitted from his initiative. Schäuble, a member of the German Bundestag since 1976, a former party leader and a minister in different prestigious ministries, can be described as the grand seigneur of the CDU. However, the president of parliament was less concerned with sympathy than with defending his party's supremacy over its little sister, the CSU.

He saw himself as a bastion of the established party hierarchies, in the form of the party committees, which are not swayed by media factors. Söder's dismissal of the decision of the party's National Executive Committee and presidium as a backroom deal triggered a “solidarity effect” (Wiesendahl, 2021, p. 3) within the CDU, which found its boldest advocate in Schäuble (Interview 1). Laschet was able to profit from this without any action on his part (Wiesendahl, 2021, p. 3).

On 18 April, both contenders met in Berlin in the chambers of the President of the Bundestag to find a common solution. As Alexander (2021, p. 360f) reports, the choice of venue was a political coup by Laschet's team. Schäuble steered the spin of the debate and operated as a kind of steering agent in this process. The justifiable arguments of better electoral chances were ultimately no match for the defense of party democracy and the legitimacy of elected party bodies (Alexander, 2021, p. 362ff). Bouffier, who attended this inner-circle meeting, suggests that “it [the meeting] was not without significance, but not ultimately decisive” (Interview 3). Söder's press conference followed the next morning, at which he declared that “where and how [the final decision is taken], the CDU decides alone, and also which body takes this decision. If it is the National Executive Committee, then we will respect that […] we as the CSU and also, I respect every decision [own translation]” (Söder, 2021b). In doing so, he was referring to the CDU's National Executive Committee meeting that Laschet had scheduled—extraordinarily—for that very evening under the pretext of discussing procedural issues (Interview 2). This reading matches Söder's statements in his press conference (Söder, 2021b), which suggest that he did not expect a final decision to be made there.

What is crucial in this context, however, is that in CDU circles, the following narrative circulates: A freshly elected party leader who gives up the chance to be chancellor in these times cannot subsequently expect much tailwind within the party (Interview 1; Alexander, 2021, p. 361). To put it more drastically: “Whoever is not a candidate for chancellor cannot be party chairman either” (Interview 3). The originator of this doctrine was Schäuble (Interview 3). He thus raised, in the minds of CDU officials and members, the specter of yet another party leadership election—a matter that must certainly be considered in view of the evening session.

Thus, the third potential factor is the weighing of established party hierarchies against election-oriented factors if these are contradictory. From the CDU's perspective, the problem can be summarized as follows: “Söder would be a better electoral leader, Laschet would be a better chancellor” (Interview 3).


Members of the party in public office, primarily parliamentarians, seek, among other things, to organize their re-election in order to have further influence on policymaking. They thus seek electorally effective candidates. The party in central office wants to maintain its influence on the party organization and its unity and thus strives for a team player in the leadership position. The party on the ground is primarily interested in having its interests represented in the form of policymaking.

The convening of the National Executive Committee meeting—whether under the guise of procedural issues or not—can be called in retrospect “pulling the joker” (Interview 3) or simply “a coup” (Interview 2) by Laschet. This digital event, on 19 April, was actually intended to finally settle the question, as Laschet declared at the beginning of the meeting that it was now time to put all the arguments openly on the table, to hold an extensive discussion, and then to take a final vote (Interview 2). Additionally, details of the closed meeting were again leaked live. Ultimately, the final vote was taken and sealed the future direction of the Christian Democrats. There were 31 votes for Laschet, 9 for Söder, and 6 abstentions. The following morning, Söder publicly admitted defeat, but in the same breath emphasized how much support he had received. The CSU secretary-general later concluded that Söder was the “candidate of the hearts [own translation]” (Söder and Blume, 2021). From the CSU's point of view, everyone in the republic should know: Söder would have been clearly the better candidate.

As stated earlier, our aim, whenever possible, is to find empirical evidence for the explanatory power of identified potential factors, to show that they are necessary and sufficient for the final outcome (García-Montoya and Mahoney, 2020). In this instance, we found empirical evidence, so the third identified potential factor can be interpreted as far more influential than the other ones and can thus be amplified into a causal mechanism.

The prioritization of party-hierarchical principles over electoral ones remains one of the potential factors. However, what enhanced it into a causal mechanism was triggered by the event of a public attack (“backroom deal”) on the established hierarchy by Söder, which can be seen as a prelude for the events that followed one after the other until the final decision taken within several days. The factor is clearly necessarily critical in its explanation of the ultimate result, as the other identified potential factors, both the election polls and party-internal support, tended toward Söder, and yet the choice fell on Laschet. The wave of solidarity in defense of fundamental party structures triggered by Söder's aggressive choice of words outweighed all other tendencies and thus decided the question of the chancellor candidacy. Essential here is the part played by Schäuble as an advocate of the primacy of party hierarchy. His “interventions” (Interview 3) against Söder's arguments—the attempt to convince him in his chambers and, even more, the insistence that the CDU party leadership and the chancellor candidacy must go hand in hand—ultimately decided the course of the events. Söder “lost his nerve” (Interview 3) and submitted to the decision of the CDU. The National Executive Committee voted for Laschet. Ultimately, it is doubtful whether hierarchical principles would have been prioritized over electoral ones without Schäuble's involvement, as the German Christian Democrats are traditionally very voter oriented. As a typical people's or catch-all party, it ever tried to maximize its numbers of voters since its foundation (Walter et al., 2011).

The CDU, with Laschet as electoral leader, lost the federal parliamentary election in 2021 with its worst result ever. The idea that Söder would have been the better candidate remains speculation, albeit well-founded (see Jun, 2022; Klein et al., 2022).

As it is often the case in the aftermath of moments of crisis in political parties, the admitted defeat in the election was followed by a further process of change in the CDU (HB 2 November 2021). However, not only was a new leader being sought within the party's own ranks, as party research has often found, there was a shift toward more inclusiveness in the selection process. This time, the party on the ground took the decision.

Party leader of the CDU (…again)


In the aftershock following the lost federal election, the CDU's party on the ground demanded the right to determine the next party leader (Interviews 1, 3). Again, a loopback effect to the previous leadership selection within the Christian Democrats can be observed (Interviews 13).

Such a change in internal party decision-making structures is far from unprecedented. It has long been suggested that parties often strengthen their intra-party democracy, in the form of a higher degree of inclusiveness, after an external shock like an electoral defeat (Harmel and Janda, 1994; Jun, 2004; Pilet and Cross, 2014). In this case, the election result was a huge shock to the CDU. At any rate, the dissatisfaction of the party on the ground with the leader selection process led to a call for more participation: The decision for Laschet contradicted the preferences of the clear majority of both the electorate and, in particular, the party members (Niedermayer, 2021b, p. 821).

Due to strong internal party pressure, Laschet announced on 7 October 2021 that he would give up the party leadership. His next step was to invite all 325 district chairs of the party to Berlin, where they would discuss the consequences of the election. Laschet concluded came to the conclusion that party members wanted more influence. The meeting thus decided to ask all party members about who should become the next party leader. In addition, the entire National Executive Committee should resign and its members subject to re-election at the next party congress, in early 2022.

What must also be mentioned here is the fact that this democratization process was by no means a long-lasting phenomenon, but as Laschet declared, a single ad-hoc event, meaning that this agreement would not result in any obligation for further party leader elections. This circumstance supports the thesis that the election defeat is to be understood as the origin, reinforced by the fact that the respective preferred candidate of the party on the ground did not win in the two preceding selection processes. The latter supports the hypothesis we made at the beginning, that the modes of intra-party selection processes can become interrelated under certain circumstances.

Before the central events or potential factors of this selection process can be analyzed in more detail, the general conditions must first be clarified.


As already mentioned, German party law does not specify the process prior to the selection of a leader at the party congress. This explains why a membership ballot on the party leader is still possible; the congress is only bound by the members' decision because of the democratic-representative norm, not officially. Voting rights were reserved for all members who were included in the central register by a certain cutoff date (, 2021b, p. 1), with voting either done by post or digitally. The process was divided into two stages. After the first stage, the election committee will invite the candidates to declare their further candidacy. If no candidate achieved an absolute majority in this first ballot, and more than one candidate remained willing to stand, a further ballot would be held between those candidates. The one with the absolute majority would then be proposed to the party congress (21 and 22 January 2022) as the election proposal of the party's National Executive Committee (, 2021b, p. 3).24

Only three contenders gained the opportunity to run through the proposals of their home district branches: Merz, Röttgen, and one newcomer, Helge Braun, the minister of the Chancellery and a Merkel ally. A fourth contender, Sabine Buder, did not achieve the necessary majority in her district branch to run (Spiegel 47/2021).


A completely new dynamic, with new rules of the game, could be observed in this selection. As mentioned above, the displeasure of many party members over the miscast chancellor candidacy left its mark on this process (Interview 1). Not only was the party on the ground set as the de facto selectorate, but the party's established opinion leaders also held back on trying to influence members. Each candidate thus reached out to the party on the ground, seeking to appeal to members and to influence their mood. Some of the special organizations still made their preferences public: the youth organization (ntv 25 November 2021) and the economic wing again supported Merz (RND 17 December 2021), while the party's women preferred Röttgen (FAZ 24 November 2021). Crucial, however, is the fact that neither Röttgen nor Braun were as well-anchored in the party as Merz, who systematically mobilized the party on the ground with various activities, such as phone calls to district chairs or important figures in party branches (Interview 1). This well-established network can certainly be regarded as a potential factor.

Röttgen had often been considered a lone wolf (TAZ 2 November 2021). Braun remained largely unknown, even within his own party, until the beginning of the pandemic, when he was entrusted with high-level crisis management tasks ( 17 December 2021). He had the reputation as a puppet of Chancellor Merkel (Interview 1), without a clear profile of his own. For some observers, he was one of the faces of the unsuccessful crisis management during the second and third corona waves in 2020 and 2021 (SZ 22 November 2021). This impression was also confirmed in an opinion poll in November 2021, showing that 48% of the CDU supporters surveyed considered Merz most suitable for the position; Röttgen received 20%, Braun 14. Among the population as a whole, 30% consider Merz qualified for the job, Röttgen 22%, and Braun only 12%, so Merz entered the race as the clear favorite.25 In addition, Braun was simultaneously running for the office of chairman of the budget committee in parliament, which internally gave the impression that he himself did not believe in his victory (Interview 1).


On 17 December 2021, the general secretary of the CDU announced the result of the first round of voting in the membership ballot via livestream. There was to be no runoff election, as Merz had already attained an absolute majority, with 62.1%. Röttgen, with 25.8%, and Braun, with only 12.1%, had to concede defeat (Tagesschau 17 December 2021). For most observers, the result was the expected one (Interview 1): Merz confirmed his role as the favorite of the party on the ground. At over 64%, turnout was surprisingly high (Interview 1), which can be explained by the dissatisfaction of the membership after the lost election and its disapproval of the procedure for determining the electoral leader. However, there may have been other factors. Since the liberal migration policy of 2015–2016, resentment had built up among the party on the ground, some of whom were critical of Merkel's course in many policy fields (Kronenberg, 2020; Wiliarty, 2021). They supported Merz's more conservative approach.

The result of the membership ballot was confirmed by the Federal Party Congress on 22 January 2022, with almost 95% of the delegates' votes in favor of Merz, who reached his goal at the third attempt. In a final step, all delegates had to confirm their electronic vote via a postal vote. At this stage, Merz received even more than 95%. If the members had been allowed to decide from the start, the CDU might have saved itself some intra-party trouble.


Using the example of the CDU, this study has shown the role of the selectorate in party leader selection, in which informal processes play a central role. Moreover, it has been shown that, in our case, different selectorates unleashed different dynamics in the intra-party power struggle (Table 2). Through our analysis via causal-process tracing, we were able to inductively identify several potential factors and one causal mechanism for internal party selection processes for leaders, which need to be examined by comparing them with other parties, also in different countries, so that potential patterns can be identified.


Table 2. Findings on the intra-party contests of the German Christian Democrats.

The potential factors can be divided into three categories. First, support plays a decisive role, which in turn can be differentiated on the basis of the three faces of the party, according to Katz and Mair (1993)—that is, the party on the ground, the party in central office, and the party in public office. At this point, the three investigated processes are too few to make patterns recognizable, but it would be interesting to investigate under which circumstances the three faces act competitively. What can be assumed from our analysis and is thus a possible starting point for comparative or quantitative studies is that the party in public office is guided by the electability of the candidates when it comes to the question of the electoral leader (Kam, 2009). This is not surprising, as the careers of parliamentarians, for example, are likely linked to the performance of their electoral leader. The party in the central office, however, tended in our case to favor the integrative team player when it came to the question of party leaders. The party on the ground, meanwhile, preferred party unity through strong leadership (Kam, 2009; Proksch and Slapin, 2014). Likewise, the decentralization of the party organization can also play a decisive role. In a federal political system like Germany's, state branches and their representatives can play a significant role if, as in the first selection process, the selectorate consists of delegates and the principle of delegation finds its origin in federal party structures. In addition, parties can be divided into sub-organizations or party wings, which can also be powerful actors in internal party decision-making. Like the state branches, their influence is (among other factors) based on the number of their members. Both can exert their influence in member ballots, as well as at party congresses, by instructing their members or delegates accordingly.

In addition to the various facets of support, we were able to identify an established and systematic network as a second aspect. Here, the focus shifts from the selectorate to the candidate's integration and anchoring within the party. They can benefit blatantly in this sense if they have many influential contacts within their party. They can cultivate these contacts and can use them consciously and effectively.

The third aspect concerns the candidates' personalities. However, their persuasiveness at crucial moments is decisive, both in absolute terms and relative to the competitors. The best example of this is Laschet's key performance at the party congress, as has been shown. However, a candidate must also enjoy the trust of the party base or the electorate and be able to convince them of his or her merits.

Regarding the dynamics within the three intra-party contests examined, it can be stated that a higher degree of inclusion and decentralization seems to limit the influence of smaller subgroups or individuals. At the same time, the decision-making process of a small group can be very strongly influenced by a single personality with sufficient motivation and power. These tendencies, which we have found in this case study, must be supplemented by further studies in order to be generalized, but they serve as an instructive starting point. It is also important to note that in our third selection process, the Zeitgeist of the lost federal election, rather than the high degree of inclusion itself, could also explain the reluctance of high-ranking party officials.

Furthermore, the modes of the various elections were found to be in some ways interlinked. After the candidates of the party on the ground failed twice in the selection processes and, moreover, the candidate of the party elite failed to meet the demands of his party, the members almost inevitably had to take the decision on the third attempt. The relationships identified here—between the composition of the selectorate and its inherent distribution of power, and between successive selection processes in terms of their degree of inclusiveness—may also be relevant beyond this single case. Astudillo and Detterbeck (2020) showed that primaries are mainly held “when there are divisions and instability in the parties' organizations” (Astudillo and Detterbeck, 2020, p. 602). Again, the dynamics found here inductively and exploratively should be compared with similar cases and thus, if possible, made generalizable for a better understanding of intra-party selection processes.

The CDU, the most important party in the German system, had to make crucial personnel decisions in the Bundestag election year 2021, which contributed to the fact that the party ultimately emerged as the loser from the Bundestag election. Its interim leader Laschet was able to prevail against Merz among the middle party establishment as well as against Söder among the narrower party leadership, but without winning the favor of his own base or even the party's voters.

In line with our theoretical considerations, it was due to the electoral shock of the clear election defeat that the party on the ground ultimately had the last word in determining the CDU's party leader. In accordance with the principle of intra-party democracy, the party on the ground should not be underestimated.

When the new CDU leadership around Merz recapitulates the events analyzed, it might conclude that, for the selection of the party leader, a high form of membership inclusivity may bring advantages even in inter-party competition. It may strengthen important aspects of an election campaign, such as legitimacy, mobilization, and participation (see also Cozza and Somer-Topcu, 2021). However, the implementation of more inclusive selection procedures does not automatically guarantee better election results

(Kernell, 2013; Cordero and Coller, 2018, p. 10f). In line with the results of Cozza and Somer-Topcu (2021), we could observe short-term effects in the form of higher poll ratings, but expectations of better election results in the long term, solely because of the primary election that was held, may be severely tempered. To keep the positive effects going, the members would have to be more involved in the shaping of programs and the selection of candidates in future elections. However, as different scholars have stated, besides the positive effects already mentioned, broadening the selectorate is also associated with challenges regarding intra-party dispute, “manipulation of rules,” “low-quality participation,” and “representational outcomes” (Kenig and Pruysers, 2018). Further research on the democratization of parties must therefore continue to delve deeper into the matter in order to identify and understand the underlying causal mechanisms.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

Ethics statement

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Author contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at:


1. ^Electoral setbacks are by far not the only causes for party democratization processes. See for more details Reiser (2018) and Astudillo and Detterbeck (2020, p. 595f).

2. ^It should be emphasized that leadership selection processes are considered separately from the selection of candidates for (sub-)national elections in the literature (see, i.e., Von dem Berge and Poguntke, 2017, p. 141ff), as other dynamics come into play here. Our contribution focuses explicitly on leadership selection processes within parties.

3. ^The German Social Democrats have also had various leadership selection processes in recent years, but two of three (Olaf Scholz as electoral Leader and Lars Klingbeil as party co-leader) were purely nominations by the party elite, without a competitive component. Only the party leader selection process ultimately won by Saskia Esken and Norbert Walter-Borjans in 2019 was characterized by a high degree of competition. The case of the Christian Democrats is therefore to be classified as more interesting for our purposes.

4. ^In this study, the term “party leader” is used as a synonym for the term “party chair.”

5. ^The interviews were conducted in German. All quotes from the interviews have been translated into English by the authors. Possible self-interest can, of course, affect interviewees who were heavily involved in the processes under study. However, we have critically examined the quoted passages to see whether they fit with press reports and the other interviews.

6. ^We are grateful to all three interviewees.

7. ^The fundamental reason for the Christian Democrats' internal turbulence was the former Chancellor Angela Merkel's refugee and migration policy, which both divided the CDU into two camps and caused a programmatic dispute with the sister party CSU (Interview 3). Merkel initially tried to establish Kramp-Karrenbauer as her successor, both in terms of personnel and program. For various reasons, however, Kramp-Karrenbauer, as party leader, gradually lost the support of Merkel, who remained chancellor. As Bouffier puts it: “What is someone like that supposed to do next to the head of government?” (Interview 3).

8. ^We understand the party elite to be an informal group of high-ranking officials of the party in central and public office, which, in this special case, is largely covered by the membership in the National Executive Committee or the presidium.

9. ^See for the potential crucial role of steering agents in selection processes of party leaders, for example, Aylott and Bolin (2017).

10. ^The influence of mass and social media on party congress delegates, which is absolutely emphasized at this point, could not be investigated within the framework of this study. From our point of view, a quantitative study targeting the media behavior of the delegates would have been a suitable method for this. We believe that asking individual delegates about this in interviews would not have yielded generalizable information. However, we support and recommend follow-up studies to this effect.

11. ^Traditionally, the state branches and the CDU minister presidents have played a decisive role in both formal and informal decision-making processes within the CDU (Interviews 2, 3), so their influence can be assumed to be high here as well.

12. ^You can find a map of where which state is located and what it is called at: (accessed March 21, 2023).

13. ^In Bavaria there is no CDU State Branch because of the Christian Democratic sister CSU, which only operates there.

14. ^Own Illustration; Sources: Tagesschau 15 January 2021; SZ 27 February 2020; Welt 28 February 2020. HS 12 January 2021; Zeit 15 January 2021.

15. ^The legal hurdles were overcome, as will be described later.

16. ^It would certainly be an interesting and worthwhile contribution to examine the differences in the role of communication between delegates at digital and face-to-face party congresses.

17. ^Own illustration; According to polls by infratest dimap in March 2020, November 2020, and January 2021.

18. ^In this case, the delegates of the Party Congress form the selectorate for the Party Leader. As we have not conducted an empirical study among these delegates, we can only identify the most plausible factors for their election decision at this point. Whether it was the state branches, special organizations, high-ranking party officers, or the appearance at the party congress itself, none or several of these factors ultimately shaped the election decision cannot be precisely determined.

19. ^There are historical reasons for this approach, as a similar question of the electoral leadership in 1979 between Franz-Josef Strauß (CSU) and Ernst Albrecht (CDU) was decided by a vote in the joint Bundestag parliamentary group in favor of Strauß (Interview 2).

20. ^You can find a map of where which state is located and what it is called at: (accessed March 21, 2023).

21. ^Own Illustration; Sources: Alexander (2021), p. 356f; Tagesschau 19 April 2021; NDR 12 April 2021; SZ 12 April 2021; SZ 16 April 2021.

22. ^Own illustration; According to polls by infratest dimap, taken monthly from January 2021 to April 2021.

23. ^Own illustration; According to polls by infratest dimap in September 2020, January 2021, and April 2021.

24. ^A candidate had to be a member of the CDU, proposed by a branch of the CDU which is entitled to file an application. Entities entitled to submit nominations are among others the National Executive Committee of the CDU as well as the Executive Committees of a State, District, or Foreign Branches (§§ 1, 6 in the CDU rules of procedure).

25. ^Poll by infratest dimap in November 2021.

26. ^As has been demonstrated earlier, the party on the ground can be identified as the crucial selector in this case.


Alexander, R. (2021). Machtverfall. Merkels Ende und das Drama der deutschen Politik: Ein Report. München:Siedler.

Google Scholar

Astudillo, J., and Detterbeck, K. (2020). Why, sometimes, primaries? Intraparty democratization as a default selection mechanism in German and Spanish mainstream parties. Party Polit. 26, 594–604. doi: 10.1177/1354068818795195

CrossRef Full Text | Google Scholar

Aylott, N., and Bolin, N. (2017). Managed intra-party democracy: precursory delegation and party leader selection. Party Polit. 3, 55–65. doi: 10.1177/1354068816655569

CrossRef Full Text | Google Scholar

Aylott, N., and Bolin, N. (2021). “Conflicts and coronations: analysing leader selection in european political parties,” in Managing Leader Selection in European Political Parties, eds N. Aylott and N. Bolin (Cham: Palgrave Macmillan), 1–28. doi: 10.1007/978-3-030-55000-4_1

CrossRef Full Text | Google Scholar

Bean, C., and Mughan, A. (1989). Leadership effects in parliamentary elections in Australia and Britain. Am. Polit. Sci. Rev. 83, 1165–1179. doi: 10.2307/1961663

CrossRef Full Text | Google Scholar

Bennett, A., and Checkel, J. T. (2015). Process Tracing. From Metaphor to Analytic Tool. Cambridge: Cambridge University Press. doi: 10.1017/CBO9781139858472

CrossRef Full Text | Google Scholar

Bittner, A. (2011). Platform or Personality? The Role of Party Leaders in Elections. Oxford: Oxford University Press. doi: 10.1093/acprof:oso/9780199595365.001.0001

PubMed Abstract | CrossRef Full Text | Google Scholar

Blatter, J., and Haverland, M. (2012). Designing Case Studies: Explanatory Approaches in Small-N Research. Basingstoke: Palgrave Macmillan. doi: 10.1057/9781137016669

CrossRef Full Text | Google Scholar

Bolin, N., and Aylott, N. (2021). “Patterns in leader selection: where does power lie?,” in Managing Leader Selection in European Political Parties, eds N. Aylott and N. Bolin (Cham: Palgrave Macmillan), 217–241. doi: 10.1007/978-3-030-55000-4_11

CrossRef Full Text | Google Scholar

Buisseret, P., and Van Weelden, R. (2020). Crashing the party? Elites, outsiders, and elections. Am. J. Polit. Sci. 64, 356–370. doi: 10.1111/ajps.12457

CrossRef Full Text | Google Scholar

Campbell, A., and Cooper, H. C. (1956). Group Differences in Attitudes and Votes. Ann Arbor, MI: University of Michigan.

Google Scholar (2021a). Das Kandidatenverfahren. Available online at: (accessed October 21, 2021).

Google Scholar (2021b). Verfahrensordnung zur Durchführung einer Mitgliederbefragung. Available online at: (accessed December 6, 2021).

Google Scholar

Cordero, G., and Coller, X. (2018). “Lights and shadows of democratizing candidate selection,” in Democratizing Candidate Selection. New Methods, Old Receipts?, eds G. Cordero and X. Coller (Cham: Palgrave Macmillan), 3–24. doi: 10.1007/978-3-319-76550-1_1

CrossRef Full Text | Google Scholar

Costa Lobo, M. (2014). “Party and electoral leadership,” in The Oxford Handbook of Political Leadership (Oxford: Oxford University Press), 362–375.

Google Scholar

Cozza, J., and Somer-Topcu, Z. (2021). Membership vote for party leadership changes: Electoral effects and the causal mechanisms behind. Elect. Stud. 71, 1–12. doi: 10.1016/j.electstud.2021.102326

CrossRef Full Text | Google Scholar

Cross, W. (2008). Democratic norms and party candidate selection. Taking contextual factors into account. Party Polit. 14, 596–619. doi: 10.1177/1354068808093392

CrossRef Full Text | Google Scholar

Cross, W. (2013). “Party leadership selection and intra-party democracy,” in The Challenges of Intra-Party Democracy, eds W. Cross and R. Katz (Oxford: Oxford University Press), 100–115. doi: 10.1093/acprof:oso/9780199661879.003.0007

PubMed Abstract | CrossRef Full Text | Google Scholar

Cross, W., and Blais, A. (2012). Who selects the party leader? Party Polit. 18, 127–150. doi: 10.1177/1354068810382935

CrossRef Full Text | Google Scholar

Cross, W., and Katz, R. (2013). “The challenges of intra-party democracy,” in The Challenges of Intra-Party Democracy, eds W. Cross and R. Katz (Oxford: Oxford University Press), 1–10. doi: 10.1093/acprof:oso/9780199661879.003.0001

PubMed Abstract | CrossRef Full Text | Google Scholar

Cross, W., and Pilet, J.-B. (2014). “The selection of party leaders in contemporary parliamentary democracies,” in Selection of Political Party Leaders in Contemporary Parliamentary Democracies. A Comparative Study, eds J.-B. Pilet and W. Cross (Oxfordshire: Routledge), 1–11.

Google Scholar

D'Antonio, O., and Werwath, C. (2012). “Die CDU: innerparteiliche willensbildung zwischen gremienarbeit und grauzone,” in Wie Entscheiden Parteien? Prozesse Innerparteilicher Willensbildung in Deutschland, eds K. R. Korte and J. Treibel (Baden-Baden: Nomos), 35–61. doi: 10.5771/9783845242392-35

CrossRef Full Text | Google Scholar

Detterbeck, K. (2013). The rare event of choice: party primaries in German land parties. German Polit. 22, 270–287. doi: 10.1080/09644008.2013.794451

CrossRef Full Text | Google Scholar

Detterbeck, K. (2018). “Alte und neue probleme der innerparteilichen demokratie,” in Parteienstaat – Parteiendemokratie, eds M. Morlok, T. Poguntke, and S. Ewgenij (Baden-Baden: Nomos), 123–142. doi: 10.5771/9783845289243-123

CrossRef Full Text | Google Scholar

Detterbeck, K., and Rohlfing, I. (2014). “Party leader selection in Germany,” in Selection of Political Party Leaders in Contemporary Parliamentary Democracies. A Comparative Study, eds J. B. Pilet and W. Cross (Oxfordshire: Routledge), 77–92.

Google Scholar

Ding, H. (2015). Die Demokratisierung der Parteiführerauswahl. Wiesbaden: Springer. doi: 10.1007/978-3-658-07514-9

CrossRef Full Text | Google Scholar

Gallagher, M. (1988). “Introduction: Candidate selection in comparative perspective,” in The Secret Garden of Politics, eds M. Gallagher and M. Marsh (London: Sage), 1–19.

Google Scholar

García-Montoya, L., and Mahoney, J. (2020). Critical Event Analysis in Case Study Research. Sociological Methods and Research. doi: 10.1177/0049124120926201

CrossRef Full Text | Google Scholar

Garzia, D., Ferreira da Silva, F., and De Angelis, A. (2022). Partisan dealignment and the personalisation in West European parliamentary democracies, 1961-2018. West Eur. Polit. 45, 311–334. doi: 10.1080/01402382.2020.1845941

CrossRef Full Text | Google Scholar

Harmel, R., and Janda, K. (1994). An integrated theory of party goals and party change. J. Theor. Polit. 6, 259–287.

PubMed Abstract | Google Scholar

Ignazi, P. (2020). The four knights of intra-party democracy: a rescue for party de-legitimation. Party Polit. 26, 9–20. doi: 10.1177/1354068818754599

CrossRef Full Text | Google Scholar

Jun, U. (2004). Der Wandel von Parteien in der Mediendemokratie. SPD und Labour Party im Vergleich. Frankfurt; New York, NY: Campus.

Google Scholar

Jun, U. (2022). “Die regierungsparteien CDU/CSU und SPD: überraschender wahlsieg der SPD als wachablösung im parteienwettbewerb?,” in Das Ende der Merkel-Jahre. Eine Bilanz der Regierung Merkel 2018-2021, eds R. Zohlnhöfer and F. Engler (Wiesbaden: Springer), 45–68. doi: 10.1007/978-3-658-38002-1_3

CrossRef Full Text | Google Scholar

Jun, U., and Jakobs, S. (2021). “The selection of party leaders in Germany,” in Managing Leader Selection in European Political Parties, eds N. Aylott and N. Bolin (Cham: Palgrave Macmillan), 73–94. doi: 10.1007/978-3-030-55000-4_4

CrossRef Full Text | Google Scholar

Kam, C. J. (2009). Party Discipline and Parliamentary Politics. Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511576614

CrossRef Full Text | Google Scholar

Karvonen, L. (2010). The Personalisation of Politics. A Study of Parliamentary Democracies. Colchester: ECPR Press.

Google Scholar

Katz, R., and Mair, P. (1992). “Introduction: The cross-national study of party organizations,” in Party Organizations. A Data Handbook on Party Organizations in Western Democracies, 1960-90, eds R. Katz and P. Mair (London: Sage), 1–20. doi: 10.7551/mitpress/4570.003.0003

CrossRef Full Text | Google Scholar

Katz, R., and Mair, P. (1993). The evolution of party organizations in Europe: the three Faces of party organization. Am. Rev. Polit. 14, 593–617. doi: 10.15763/issn.2374-7781.1993.14.0.593-617

CrossRef Full Text | Google Scholar

Kenig, O. (2009). Democratization of party leadership selection. Do wider selectorates produce more competitive contests? Elect. Stud. 28, 240–247. doi: 10.1016/j.electstud.2008.11.001

CrossRef Full Text | Google Scholar

Kenig, O., and Pruysers, S. (2018). “The challenges of inclusive intra-party selection methods,” in Democratizing Candidate Selection. New Methods, Old Receipts?, eds G. Cordero and X. Coller (Cham: Palgrave Macmillan), 25–48. doi: 10.1007/978-3-319-76550-1_2

CrossRef Full Text | Google Scholar

Kernell, G. (2013). “Political party organizations, civic representation, and participation,” in Representation: Elections and Beyond, eds J. Nagel and R. Smith (Philadelphia, PA: University of Pennsylvania Press), 114–136. doi: 10.9783/9780812208177.114

CrossRef Full Text | Google Scholar

Klein, M., Springer, F., and Kühling, C. (2022). Last man standing: zur bedeutung der kanzlerkandidaten für das ergebnis der bundestagswahl 2021. Zeitsch. Parlam. 53, 17–38. doi: 10.5771/0340-1758-2022-1-17

CrossRef Full Text | Google Scholar

Kronenberg, V. (2020). “Die Öffnung der CDU im deutschen parteienspektrum als vermächtnis der ära merkel,” in Politische Parteien in der modernen Demokratie, eds R. Grünewald, S. Busch-Janser and M. Piepenschneider (Berlin: LIT Verlag), 317–346.

Google Scholar

Laschet, A. (2021). Application Speech. Berlin: CDU National Party Congress. Available online at: (accessed October 21, 2021).

Google Scholar

Lazarsfeld, P. F., Berelson, B., and Gaudet, H. (1960). The People's Choice. How the Voter Makes Up His Mind in a Presidential Campaign, 5th Edn. New York, NY: Columbia University Press.

Google Scholar

Mahoney, J. (2021). The Logic of Social Science. Princeton, NJ: Princeton University Press.

Google Scholar

Marsh, M. (1993). Introduction: selecting the party leader. Eur. J. Polit. Res. 24, 229–231. doi: 10.1111/j.1475-6765.1993.tb00378.x

CrossRef Full Text | Google Scholar

Merz, F. (2021). Application Speech. Berlin: CDU National Party Congress. Available online at: (accessed October 21, 2021).

Google Scholar

Müller, W. C., and Strøm, K. (1999). “Political parties and hard choices,” in Policy, Office, or Votes? How Political Parties in Western Europe Make Hard Choices, eds W. C. Müller and K. Strøm (New York, NY: Cambridge University Press), 1–35. doi: 10.1017/CBO9780511625695.001

PubMed Abstract | CrossRef Full Text | Google Scholar

Niedermayer, O. (2021a). Parteimitgliedschaften im Jahre 2020. Zeitsch. Parlam. 52, 373–407. doi: 10.5771/0340-1758-2021-2-373

CrossRef Full Text | Google Scholar

Niedermayer, O. (2021b). Corona als “stunde der regierungsparteien”? Die folgen der covid-19-pandemie für das parteiensystem. Zeitsch. Parlam. 52, 805–823. doi: 10.5771/0340-1758-2021-4-805

CrossRef Full Text | Google Scholar

Panebianco, A. (1988). Political Parties: Organization and Power. Oxford: Oxford University Press.

Google Scholar

Pedersen, H. H., and Rahat, G. (2021). Political personalization and personalized politics within and beyond the behavioural arena. Party Polit. 27, 211–219. doi: 10.1177/1354068819855712

CrossRef Full Text | Google Scholar

Pilet, J.-B., and Cross, W. (2014). “The selection of party leaders in comparative perspective,” in Selection of Political Party Leaders in Contemporary Parliamentary Democracies. A Comparative Study, eds J. B. Pilet and W. Cross (Oxfordshire: Routledge), 222–239. doi: 10.4324/9781315856025

PubMed Abstract | CrossRef Full Text | Google Scholar

Proksch, S.-O., and Slapin, J. B. (2014). The Politics of Parliamentary Debate. Parties, Rebels, and Representation. Cambridge: Cambridge University Press. doi: 10.1017/CBO9781139680752

CrossRef Full Text | Google Scholar

Pruysers, S., Cross, W. P., and Katz, R. S. (2018). “Personalism, personalization and party politics,” in The Personalization of Democratic Politics and the Challenge for Political Parties, eds W. P. Cross, R. Katz, and S. Pruysers (Lanham: Rowman and Littlefield), 1–18.

Google Scholar

Rahat, G., and Hazan, R. (2001). Candidate selection methods: an analytical framework. Party Polit. 7, 297–322. doi: 10.1177/1354068801007003003

CrossRef Full Text | Google Scholar

Rahat, G., and Kenig, O. (2018). From Party Politics to Personalized Politics? Party Change and Political Personalization in Democracies. Oxford: Oxford University Press. doi: 10.1093/oso/9780198808008.001.0001

CrossRef Full Text | Google Scholar

Rehmert, J. (2022). Party elites' preferences in candidates: evidence from a conjoint experiment. Polit. Behav. 44, 1149–1173. doi: 10.1007/s11109-020-09651-0

CrossRef Full Text | Google Scholar

Reiser, M. (2018). “Contagion effects by the AfD?,” in The Selection of Politicians in Times of Crisis, eds X. Coller, G. Cordero, and M. Jaime-Castillo (New York, NY: Routledge), 81–97. doi: 10.4324/9781315179575-6

CrossRef Full Text | Google Scholar

Röttgen, N. (2021). Application Speech. Berlin: CDU National Party Congress. Available online at: (accessed October 21, 2021).

Google Scholar

Sandri, G., Seddone, A., and Venturino, F. (2015). “Understanding leadership profile renewal,” in The Politics of Party Leadership. A Cross-National Perspective, eds W. Cross and J. B. Pilet (Oxford: Oxford University Press), 90–106. doi: 10.1093/acprof:oso/9780198748984.003.0006

PubMed Abstract | CrossRef Full Text | Google Scholar

Schmitt-Beck, R. (2021). Wer wählte wen bei der Bundestagswahl 2021? Aus Polit. Zeitgesch. 71, 10–16.

Google Scholar

Schoen, H. (2014). “Soziologische ansätze in der empirischen wahlforschung,” in Handbuch Wahlforschung, eds J. W. Falter and H. Schoen (Wiesbaden: Springer VS), 169–239. doi: 10.1007/978-3-658-05164-8_6

CrossRef Full Text | Google Scholar

Schönbohm, W. (1985). Die CDU wird Moderne Volkspartei. Selbstverständnis, Mitglieder, Organisation und Apparat. 1950-1980. Stuttgart:Klett-Cotta.

Google Scholar

Seddone, A., and Sandri, G. (2021). “Political elites and party primaries,” in New Paths for Selecting Political Elites. Investigating the Impact of Inclusive Candidate and Party Leader Selection Methods, eds G. Sandri and A. Seddone (New York, NY: Routledge), 228–243. doi: 10.4324/9781003022893-12

CrossRef Full Text | Google Scholar

Söder, M. (2021a). Press Conference. Berlin: Federal Parliamentary Group. Available online at: (accessed October 21, 2021).

PubMed Abstract | Google Scholar

Söder, M. (2021b). Press Conference. Munich: CSU Headquarter. Available online at: (accessed October 21, 2021).

Google Scholar

Söder, M., and Blume, M. (2021). Press statement. Munich: CSU Headquarter. Available online at: (accessed October 21, 2021).

Google Scholar

Von dem Berge, B., and Poguntke, T. (2017). “Varieties of intra-party democracy: conceptualization and index construction,” in Organizing Political Parties. Representation, Participation, and Power, eds S. Scarrow, P. Webb, and T. Poguntke (Oxford: Oxford University Press), 136–157.

Google Scholar

Walter, F., Werwath, C., and D‘Antonio, O. (2011). Die CDU. Entstehung und Verfall Christdemokratischer Geschlossenheit. Baden-Baden: Nomos. doi: 10.5771/9783845232515

CrossRef Full Text | Google Scholar

Wiesendahl, E. (2021). Hinterzimmer Versus Inszenierung: Wie Man in Deutschland Heutzutage Kanzlerkandidat:in Wird. Discussion Paper. Das Progressive Zentrum.Available online at: (accessed October 21, 2021).

Google Scholar

Wiliarty, S. (2021). “Germany: how the christian democrats manage to adapt to the silent counter-revolution,” in Riding the Populist Wave, Europe's Mainstream Right in Crisis, eds T. Bale and C. R. Kaltwasser (Cambridge: Cambridge University Press), 141–169. doi: 10.1017/9781009006866.007

CrossRef Full Text | Google Scholar

Keywords: party leadership, electoral leadership, selection process, intra-party democracy, German Christian Democrats, process tracing

Citation: Jun U and Minas M (2023) Power struggles in the German Christian Democrats. The dynamics of three consecutive leadership contests. Front. Polit. Sci. 5:1039547. doi: 10.3389/fpos.2023.1039547

Received: 08 September 2022; Accepted: 04 July 2023;
Published: 29 August 2023.

Edited by:

Nicholas Aylott, Södertörn University, Sweden

Reviewed by:

Debra Leiter, University of Missouri–Kansas City, United States
Jana Schwenk, University of Gothenburg, Sweden

Copyright © 2023 Jun and Minas. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Uwe Jun,; Marius Minas,

These authors have contributed equally to this work