Introduction

There is a widespread belief in some popular and academic analyses that Germany prevails in the European Union, dictating the negotiation outcomes and being the most successful in attaining its preferences. In the literature, this country is often described as “embedded hegemon” (Crawford 2007), “benign hegemon” (Morisse-Schilbach 2011) or “reluctant hegemon” (Paterson 2011). Generally, Germany’s prevalence is explained by its superior power resources, especially by the proponents of distributive bargaining theory. Some scholars argue that this state is likely to succeed in EU lawmaking because it wields the largest voting power in the Council, thus being pivotal in building winning or blocking coalitions under qualified majority voting (Barr and Passarelli 2009; Hosli and Machover 2002; Pajala and Widgrèn 2004; Selck and Kaeding 2004; Thomson 2011; Tsebelis and Yataganas 2002). Others suggest that Germany dominates the EU decision-making process due to its superordinate economic strength (Bulmer and Paterson 2013; Moravcsik 1998; Schimmelfennig 2015; Schoeller 2017). The 2010–2015 negotiations to reform the Eurozone are shown as an example of how this state used this advantage to achieve its preferred outcomes (Bernhard and Leblang 2016; Brunnenmeier et al. 2016; Bulmer and Paterson 2013; Matthijs and Blyth 2015). For instance, Bulmer and Paterson (2013: 1396) argue that “The eurozone crisis has thus presented evidence of Germany being “number 1,” (…) being the most influential player in the process of finding solutions,” while Schimmelfennig (2015: 188) adds that: “[Euro area] reform generally bears the hallmark of Germany (…), the rest of the EA largely accepted the terms that Germany preferred.” Several studies also found Germany to dominate the euro crisis debate by imposing its traditional ordo-liberal economic paradigm on other member states (Bulmer 2014; Crespy and Schmidt 2014; Matthijs and McNamara 2015; Schäfer 2016). Against this backdrop, sociologist Ulrich Beck concluded that the Eurozone crisis “has given birth to a political monster: a German Europe” (Beck 2013).

Yet, the presented literature has two lacunae. First, there is a shortage of empirical studies that systematically examine Germany’s bargaining success in “everyday” decision-making, that is when EU secondary legislation is adopted. So far, scholars have mainly focused on how Germany fare during intergovernmental negotiations taking place in the European Council, where EU treaties or policy guidelines are decided. However, daily EU legislation is adopted by ministers in the Council of Ministers (with the participation of the European Parliament depending on the legislative procedure) where states’ power resources combined with the possibility of majority decision-making may have considerable implications for bargaining success. It is thus crucial to assess the alleged German prevalence in this less investigated venue.

Second, there are several more general studies that challenge the narrative of Germany’s pre-eminence in EU lawmaking. Golub (2012) found that this country was not the biggest winner in negotiations on the most controversial EU legislation in the period 1999–2001. Likewise, Lundgren et al. (2019) showed that Germany did not dictate the resolution to the 2010–2015 Eurozone reform. Hence, there is a need to take a closer look at Germany’s bargaining success in the EU.

This paper aims to fill those gaps by offering the first comprehensive assessment of Germany’s success in relation to actual (secondary) EU legislation. Specifically, it conducts three quantitative studies to test the hypothesis that this country is the most successful in daily EU lawmaking. The first examines its success in terms of how many times it voted against legislative proposals in the Council, as compared to other member states. For this purpose, I scrutinize Council voting records between December 1, 2009 and July 1, 2019. The second study investigates Germany’s bargaining success in terms of its preference attainment, i.e., the extent to which the negotiated outcomes on EU legislation overlap with this state’s initial preferences. To this end, I use the DEUII dataset covering actors’ policy positions and final outcomes on issues discussed in 1999–2009. The third study assesses Germany’s success in relation to secondary legislation for the Eurozone reform, based on the EMU Positions dataset which includes actors’ preferences on the most important proposals negotiated during the Eurozone crisis (2010–2015).

The results provide no support for Germany’s prevalence in daily EU lawmaking. Not only does it not dictate the shape of secondary legislation, but it also emerges as one of the biggest losers in negotiations. The paper suggests a possible explanation for such poor showing: the highest extremity and intensity of Germany’s preferences. It also questions the power-based distributive bargaining argument that states’ with greater power resources are more successful in the EU.

Hypothesis on Germany’s prevalence in EU lawmaking

In this section, I formulate the hypothesis expecting Germany to be the most successful member state in daily EU lawmaking. This supposition can be derived from a rational choice institutionalism, in particular its power-based distributive bargaining argument (Knight 1992; Schelling 1960). This approach assumes that actor’s bargaining success is a function of its power resources: negotiation outcomes primarily reflect the preferences of the most powerful actors since they can make use of their superordinate resources to gain concessions from weaker partners. Here, I distinguish three power resources relevant for bargaining success in the EU: voting strength, economic size and network capital. Given that Germany has the highest stocks of all three, it should prevail in EU lawmaking.

First, Germany is the most powerful member state in terms of voting strength. States’ voting power in the Council is often estimated by several indices, i.e., the Shapley-Shubik index (SSI, Shapley and Shubik 1954) and the Coleman index (CI, Coleman 1971). The former measures the likelihood that an actor will be pivotal in transforming a losing coalition into a winning one, while the latter determines the power to block a decision, i.e., the number of situations in which an actor turns a winning coalition into a losing coalition by moving out of it. Under the double majority rule established by the Lisbon Treaty,Footnote 1 Germany holds the highest level of both indices, far exceeding other member states. According to SSI and CI calculated for the 2015 population, the presence of this country is pivotal in forming 14.9% of all possible winning coalitions and 75.45% of blocking coalitions (Antonakakis et al. 2014). The same coefficients for France, occupying second place, are much lower: 11.68% and 61.10%, respectively. Given that member states with larger voting power are expected to have more influence on policy outcomes (Warntjen 2017), Germany should be the most successful in daily EU lawmaking.

Second, Germany possesses the highest economic power among all member states. It is the biggest EU national economy in terms of nominal GDP, accounting for more than a quarter the bloc’s output and having the largest trade surplus. The literature on EU bargaining argues that states with greater structural power have a larger say in the negotiations since they can exploit their comparative advantages in relevant sectors to get concessions from economically dependent partners (Bailer 2004; Moravcsik 1998). Hence, Germany should be the most successful in shaping the EU secondary legislation due to its huge share in the EU economy.

Third, the literature also emphasizes the role of network capital in determining a state’s bargaining success (Huhe et al. 2018). It can be defined as the quantity and quality of cooperation connections (working relations) an actor managed to build with fellow negotiators. The distribution of network capital in the Council was measured by Naurin (2007) who conducted interviews with officials from all EU permanent representations, asking them to mention the member states they cooperated most often with. He discovered that Germany possesses the highest stock of network capital since this country was most frequently considered as the cooperation partner by interviewees. Given that having close network ties to many states translates into greater bargaining success (Arregui and Thomson 2009; Lundgren et al. 2019), Germany should prevail in daily EU lawmaking. To sum up:

H1

Given its highest voting strength, network capital and economic power, Germany is more successful in achieving its preferred outcomes in relation to actual EU legislation than other member states.

Methodology

To test the hypothesis, I conducted three quantitative studies. The reason is to evaluate the hypothesis more rigorously on several different sets of data and in a longer time frame. Below, I explain the methodology of the three tests.

The first study

The first study assesses Germany’s bargaining success in terms of voting results in the Council. Specifically, it investigates how often this country voted against legislative proposals, compared to other member states. The analysis is carried out on the Council roll-call data retrieved from the “Monthly Summary of Council Acts” documents which juxtapose all definitive acts adopted by the Council and—when voting took place—which member states explicitly voted against or abstained from voting.Footnote 2 However, this data were restricted to votes on proposals that meet two conditions. First, they were legislative acts, i.e., regulations, directives or decisions adopted under the ordinary or special legislative procedures. As a result, non-legislative acts, e.g., on the CFSP affairs, were excluded from the analysis. The reason is that voting results on these files were frequently missing. Second, the dataset includes legislative acts that were adopted after the entry into force of the Lisbon Treaty and during the 7th (2009–2014) and 8th (2014–2019) EP terms, that is between December 1, 2009 and July 1, 2019 (the day before the first sitting of the 9th EP term). Overall, the sample involves 1146 votes on proposals meeting those criteria.

The dependent variable—a state’s success—measures the number of contested votes (i.e., negative votes and abstentions) cast by each state in a given year. It is ideally assumed that a state is successful if it supports a legislation during a final vote. Thus, the more proposals a country contests in a year, the less successful it is. However, this measure should be used with caution. It must be emphasized that a positive vote does not always indicate that a state is satisfied with the outcome. Novak (2013) showed that the absence of explicit opposition is often a strategic behavior—a government does not register its dissent even when it is not content with a proposal in order to avoid being accused by their constituencies of having failed to defend national interests at the EU level. Likewise, a state may decide not to vote against a proposal despite disagreeing with the Council majority due to irrelevance of a legislation.

The second study

The second study investigates Germany’s bargaining success in terms of its preference attainment, i.e., the extent to which the negotiated outcomes on EU legislation overlap with a state’s initial preferences. For this purpose, I use the DEUII dataset which includes information on 331 contested issues related to 125 most controversial legislative proposals discussed in the period 1999–2009 and subjected to either consultation or co-decision, the two most important legislative procedures (Thomson et al. 2012). The DEUII dataset includes: (1) the policy positions of all member states, the EP and the Commission on each legislative issue represented spatially on a scale of 0–100; (2) the level of importance each actor assigned to an issue on the same scale; (3) the final outcome on an issue.

Figure 1 shows how controversial issues are represented in DEUII. It illustrates actors’ policy positions and saliencies regarding one issue raised by the postal services directive (2006/0196/COD). The aim of this proposal was to fully achieve an internal market for postal services by 2009. The main controversy concerned the timing of liberalization of national postal services. On this issue, actors’ policy positions divided along three lines. One group of member states wanted the postal services market to be fully opened later than 2010 without specifying the exact date (position 0). This coalition was made up of member states that did not have liberalized postal services, notably France, Italy and countries that joined the EU in 2004 and 2007 with the exception of Estonia. The second position was taken by Austria, Denmark and the Parliament (position 60). They called for a full market opening by 2010 with the possibility to extend this deadline by two more years for member states that: (1) acceded to the EU, (2) have a small population and geographical size; (3) have a difficult topography, e.g., a number of islands. The third group, backed by the Commission, called for a rapid liberalization by 2009 (position 100) or 2010 (position 80) without any derogations. This position was favored by states that had already carried out a complete or partial liberalization of postal markets, i.e., Germany, the UK and the Nordic countries. Notably, this issue was of particular importance to Germany and the Netherlands (salience = 100). The outcome aligned with the second position: a delay of full liberalization until December 31, 2010 with the option that some member states can postpone it until December 31, 2012.

Fig. 1
figure 1

Source: Thomson et al. (2012)

Actors’ policy positions regarding the liberalization of national postal services. Notes: Salience scores in parentheses. EP—European Parliament, COM—European Commission, AT—Austria, BE—Belgium, CY—Cyprus, CZ—Czechia, DE—Germany, DK—Denmark, EE—Estonia, EL—Greece, ES—Spain, FI—Finland, FR—France, HU—Hungary, IE—Ireland, IT—Italy, LT—Lithuania, LU—Luxembourg, LV—Latvia, MT—Malta, NL—The Netherlands, PL—Poland, PT—Portugal, SE—Sweden, SI—Slovenia, SK—Slovakia, UK—The United Kingdom.

Full size image

The dependent variable in this study is a state’s bargaining success. It is measured as the absolute distance between its policy position and the decision outcome on an issue, weighted by the salience a state attaches to an issue.Footnote 3 Thus, the dependent variable ranges from 0 to 100, where 0 indicates that a state attains the highest level of bargaining success since the outcome is fully in line with its salience-weighted position, while the value of 100 means a complete failure—the largest possible salience-weighted distance between a state’s position and the outcome. The salience-weighted operationalization of bargaining success is advocated in the literature because it captures the various intensity of actors’ preferences (Arregui 2016; Golub 2012). In fact, two actors with similar policy positions could attain distinct success when they attach different levels of salience to an issue. Based on Fig. 1, the EP, Austria and Denmark emerge as clear winners in negotiations over the postal service issue since the outcome ideally reflects their policy position. By contrast, the biggest losers are all new member states (except for Estonia) as well as Belgium, Luxemburg, Portugal, Greece and France. While they attached a high level of salience (90) to this issue, they obtained the outcome being 54 salience-weighted points away from their position—the highest distance out of all actors.

The third study

The third study assesses Germany’s bargaining success in relation to secondary legislation proposals reforming the Eurozone. Here, I rely on the EMU Positions dataset which contains 47 contested issues related to the 10 most important legislative and non-legislative proposals discussed between 2010 and 2015 in response to the Eurozone crisis (Wasserfallen et al. 2019).Footnote 4 For each issue, this dataset includes states’ policy positions and the outcome, both represented on a scale ranging from 0 to 100. In addition, it reports the intensity of actors’ preferences, measured on a scale from 0 (low salience) to 10 (high salience). Given that the paper focuses on daily EU legislation, I dropped issues belonging to proposals related to the EU primary law or not adopted under the ordinary or special legislative procedure, i.e., the Fiscal Compact, the Treaty establishing the ESM, the EFSF Framework Agreement and the Agreement on the Single Resolution Fund. Moreover, since the EMU Positions dataset provides no information about decision outcomes on the Eurobonds, the Financial Transaction Tax and the “Five Presidents” Report, I excluded them from the analysis as well. Thereby, the final sample covers 20 issues.

Figure 2 presents an illustration of how the EMU Positions dataset is used on an example of negotiations over the first pillar of the Banking Union—a proposal establishing the Single Supervisory Mechanism (2012/0242/CNS). Its aim was to equip the European Central Bank with specific tasks previously belonging to national authorities concerning the prudential supervision of all credit institutions established in the Eurozone countries. One of the contentious issues was the scope of the SSM: Actors were divided on whether the ECB should directly supervise all or some banks. Five member states, namely Germany, Italy, Austria, Malta and Cyprus, called for the coverage of the largest banks only (position 0). Their main goal was to exclude their smaller domestic banks from the ECB supervision (Schäfer 2016: 965). This preference was extremely important for Germany as indicated by high saliency (9.63) it attached to this issue. On the opposite side were 17 member states supported by the Commission and EP. They argue that all Eurozone banks must be directly supervised by the ECB (position 100). The central position, albeit closer to the latter coalition, was taken by Greece. It wanted the SSM to cover all Eurozone banks, wherein only about 130 largest banks would be supervised directly by the ECB, while the rest would be covered by national authorities unless the ECB decides to involve itself directly (position 70). The final compromise was in line with the Greek preference.

Fig. 2
figure 2

Source: Wasserfallen et al. (2019)

Actors’ policy positions regarding the scope of the Single Supervisory Mechanism. Notes: salience scores in parentheses. See Fig. 1 for abbreviations.

Full size image

Since the structure of the EMU Positions dataset is similar to that of DEUII, I adopted the same spatial formulation of the dependent variable. A state’s bargaining success is thus measured as the salience-weighted distance between its initial policy preference and the outcome on an issue.Footnote 5 It ranges from 0 to 100, where lower values denote higher bargaining success (a state’s policy position is closer to the final outcome), whereas higher scores signify lower bargaining success. Looking at Fig. 2, the highest bargaining success for this issue was attained by Greece, because its position perfectly overlaps with the outcome. By contrast, Germany can be understood as the biggest loser since its preference was 67.41 salience-weighted points away from the outcome, being the most distant among all member states.

Testing the hypothesis

Germany’s success in terms of voting results in the Council

Figure 3 displays the number of contested votes by each member state during the period 2009–2019. Overall, Germany voted “no” or abstained 55 times, which represents 4.8% of all votes. As a result, it emerges as one of the most frequently contesting countries in the Council, occupying third position in that respect behind the UK and Austria. However, Fig. 3 shows that Germany was more likely to contest EU legislation in the 7th (2009–2014) than in the 8th EP term (2014–2019). This pattern may stem from the governmental change in 2013 when a center-right cabinet (CDU/CSU/FDP) was replaced by a “grand coalition” (CDU/CSU and SPD). Nevertheless, even in the 2014–2019 period Germany topped the list of “no” voters and abstainers. Overall, Fig. 3 shows that, contrary to expectations, Germany is the most outvoted state in the Council after the UK and Austria.

Fig. 3
figure 3

Source: Monthly Summary of Council Acts

Number of contested votes by member states, 2009–2019.

Full size image

To check whether Germany’s level of contestation is statistically different from that of the other member states, I performed a t test. I created two groups—one for Germany and one for other members states—and compared their yearly mean number of contested votes in the period 2009–2019. The means are 5.5 and 2.8, respectively. A t test yields a significant p-value (t =  − 2.278, p = 0.023), demonstrating that there is a significant difference between these two groups. This result suggests that Germany’s propensity to contest EU legislation is significantly higher than that of the other countries. As a robustness check, I also fitted the fractional logit models with country dummies (see Table A in the “Online Appendix”). The dummy for Germany is positive and significant, confirming that this state expresses dissent more often than others.

In addition, I explored which member states are more or less likely to vote negatively compared to Germany. I conducted a series of paired t tests between Germany and each other member state to assess differences in their yearly mean number of contested votes in the period 2009–2019. Table 1 reports the results. Taking p < 0.05 as the criterion for significance, Germany is found to dissent with the Council majority significantly more often than 21 out of 27 member states. Such a large number of superior countries indicates that Germany experiences considerable difficulties in achieving satisfactory negotiation outcomes in the Council compared to other members.

Table 1 Paired t tests comparing the yearly average number of contestations between Germany and other member states, 2009–2019
Full size table

I also investigated Germany’s propensity to contest EU legislation across policy areas (see Fig. A in the “Online Appendix”). For this purpose, I used the EP committee responsible for the proposal as a proxy and selected the policies with at least 25 votes. Overall, Germany was found to be among the three most contesting countries in 7 out of 14 policy areas under investigation: agriculture and rural development, economic and monetary affairs, employment and social affairs, internal market and consumer protection, civil liberties, justice and home affairs, transport and regional development. Given the highest importance these policies have for member states, it can be said that Germany is on the losing side primarily in the most salient domains.

Generally, the first empirical study does not corroborate H1. Germany is found to cast the highest number of contested votes, especially in important policy areas, being considerably less satisfied with legislative outcomes relative to the vast majority of member states.

Germany’s bargaining success in terms of preference attainment

Figure 4 illustrates the average bargaining success of individual member states and two institutional actors—the EP and the Commission according to the DEUII. Contrary to expectations, it shows that Germany punches below its weight in daily EU lawmaking. In the years 1999–2009, its policy position was on average 23 salience-weighted points distant from the final outcome on the 0–100 scale. This result gives Germany 27th place out of all 29 actors, with only France and the Commission attaining lower success. However, with the exception of the United Kingdom, the other countries holding the highest voting power—France, Italy, Spain and Poland—are also located at the bottom of the lower part of Fig. 4. Importantly, Germany’s poor showing in the period 1999–2009 was not caused by the “big-bang” enlargement of the EU that took place on May 1, 2004 (see Fig. B in the “Online Appendix”). Prior to the 2004 accession, its policy position was on average 23 salience-weighted points away from decision outcome, while this distance remained at almost the same level in the post-2004 era (23.34). No effect of the 2004 accession is also confirmed by a t test comparing Germany’s mean bargaining success before and after this event (t =  − 0.11, p = 0.912).

Fig. 4
figure 4

Source: Thomson et al. (2012)

Actors’ salience-weighted bargaining success, 1999–2009. Notes: the red reference line indicates the average level of bargaining success for all countries.

Full size image

Figure 4 reveals that actors’ average bargaining success is relatively evenly distributed as the distances are small and concentrated in the span between 15 and 25 points. Hence, to investigate whether Germany’s low preference attainment is statistically distinguishable from that of the other member states, I carried out an additional t test. Specifically, I compared the mean salience-weighted bargaining successes of two groups: Germany and other members (treated as one group). The results are as follows. First, the mean bargaining success for Germany (23.17) is lower than that for other member states (19.85). Second, a t test produces a significant p-value (t =  − 2.57, p = 0.010), demonstrating that Germany attains significantly different and lower bargaining success than the rest of member states. Additionally, I ran linear regressions with country dummies (see Table B in the “Online Appendix”). The dummy for Germany is positive and significant (albeit only at the 0.1 level), confirming that this country is less successful than other countries.

But which countries are better or worse than Germany? To address this question, I conducted a series of paired t tests between the average salience-weighted bargaining successes of Germany and each member state. Table 2 reports the results. To understand how to interpret the table, consider the Germany-Slovakia dyad. Germany’s mean bargaining success is 23.16 and Slovakia’s is 16.66, so the difference is equal to 6.50 salience-weighted points. A t test comparing those means yields a p-value of 0.009, indicating that they are statistically distinguishable and that the former country is significantly less successful than the latter. When taking p < 0.05 as the level for significance, it can be seen that Germany has significantly different and lower bargaining success than 12 out of 27 member states. Surprisingly, none of the actors are found to be substantively less successful than Germany, even France and the Commission. Generally, these results confirm the impotency of Germany vis-à-vis most member states in achieving its preferred outcomes in daily EU lawmaking.

Table 2 Paired t tests comparing the average salience-weighted bargaining success between Germany and other actors.
Full size table

Since Germany’s bargaining success may differ across policy areas, I scrutinized how member states fare in 15 key EU policies (see Fig. C in the “Online Appendix”). I used the Commission’s Directorate-General responsible for a proposal as a proxy. The results show that Germany’s score is higher than the average in only four policies: agriculture, education and culture (EDU), enterprise and industry, and external relations. Specifically, this state is among the most successful countries in only one policy area (EDU), while it emerges as the biggest loser in four: justice and home affairs, information society (INFSO), Secretariat (SEC), and energy and transport (ETRAN). To investigate whether the observed variation is significant, I conducted a series of t tests, comparing the mean bargaining success of Germany and other member states (as one group) in each policy area (see Table C in the “Online Appendix”). The results are significant only in three policies where Germany has been previously found to be the least successful: SEC, INFSO and ETRAN. Thus, it can be concluded that in these policy areas Germany exhibits significantly lower bargaining success than other states, while there is not a single policy domain in which it fares much better than the rest.

Overall, the second study also provides no support for H1. However, the question arises: what are the reasons for Germany’s poor showing? Several studies argue that the positioning and intensity of policy preferences strongly influence state’s bargaining success (Arregui 2016; Bailer 2004). In line with this reasoning, I used the DEUII dataset to juxtapose the extremity of actors’ policy positions (measured as their absolute distance from the mean position of other member states) as well as the mean levels of salience they attached to all the issues (see Fig. D in the “Online Appendix”). A correlation analysis confirms that states’ bargaining success is strongly dependent on the extremity and intensity of their preferences (r = 0.396, p < 0.001 and r = 0.42, p < 0.001, respectively). Not surprisingly, Germany exhibits the highest values of both measures, being on average 30.4 points distant from the mean position and attaching on average 65 points to an issue on the 0–100 scale. Hence, Germany’s poor performance may stem from high levels of its preference extremity and intensity as compared to other states.

In fact, there are many qualitative examples in DEUII confirming this expectation. Consider two issues related to the VAT proposal (2008/0143/CNS) on which Germany took extreme and intense policy positions as it opposed granting the right to all member states to apply the reduced level of value added tax to certain labor-intensive services (position = 0; salience = 100, extremity = 63) and restaurants (position = 0; salience = 100, extremity = 66.6). However, it suffered a disastrous defeat in both cases as final legislation conferred such powers to member states (position = 100). Another example concerns the Return Directive (2005/0167/COD). Germany strongly contested granting minimum rights to illegally staying third-country nationals awaiting deportation (position = 0; salience = 90; extremity = 58.6). However, the final outcome provided illegals with basic social services, such as housing and emergency health care (position = 70), thereby being far from what the German government wanted. A controversy on the liberalization of postal services, presented in the “Methodology” section, is also relevant. On that issue, Germany held the most extreme and intense preference, opposing the postponement of the full market opening and the establishment of derogations for some member states (position = 100; salience = 100; extremity = 66.6), but the final compromise introduced both these solutions.

Germany’s bargaining success in the Eurozone reform

Figure 5 gives a graphical overview of member states’ bargaining success in relation to secondary legislation proposals reforming the Eurozone. Given that the EMU Positions dataset exhibits many missing values on preference saliencies, I also reported the unweighted bargaining success for comparative purposes. As a reminder, the larger the score, the lower its bargaining success.

Fig. 5
figure 5

Source: Wasserfallen et al. (2019)

Member states’ unweighted and salience-weighted bargaining success on secondary legislation proposals reforming the Eurozone. Notes: the red reference line indicates the average level of bargaining success for all countries.

Full size image

Figure 5 shows that Germany was the least successful country. On average, its policy position was 41.73 salience-weighted points distant from the outcome on a scale of 0–100. Since this is the lowest level of bargaining success attained by an actor, Germany ranks last among all 28 member states. The same conclusion applies when examining the unweighted bargaining success: Germany still occupies the last position, being on average 51.50 policy scale points away from the outcome. Hence, the above observations complement previous findings (Lundgren et al. 2019) by showing that Germany did not dictate the shape of the EU secondary legislation reforming the Eurozone. However, this finding only pertains to the legislative files that did make it to the negotiation table. Thus, it does not exclude that Germany may have been successful in shaping the agenda by pre-selecting or blocking other potential proposals.

It is also important to check whether Germany’s score is statistically distinguishable from the outcomes achieved by other countries. For this purpose, I conducted a t test by comparing the mean bargaining successes of two groups: Germany and other member states. The result is statistically significant (t =  − 4.21, p < 0.013), implying that there is a substantive variation between the two groups.Footnote 6 Given that Germany’s mean salience-weighted success is considerably lower (42.46) than that of the second group (21.81), one can say that Germany was significantly less successful than the rest of the member states. This finding is corroborated by the regression analysis with country dummies, revealing a statistically significant dummy for Germany (see Table D in the “Online Appendix”).

Similarly, I conducted pairwise comparisons of Germany’s and individual states’ average salience-weighted successes. Table 3 reports the results of paired t tests. As indicated by the positive and significant differences, Germany did worse than 23 out of 27 member states. Such a large number of superior countries displays how poorly this country did in the negotiations on secondary legislation reforming the Eurozone. Hence, the third study does not support H1.

Table 3 Paired t tests comparing the average salience-weighted bargaining success between Germany and other actors.
Full size table

As in the previous analysis, I investigated why Germany fared so badly. To this end, I compared member states’ preference extremity and salience based on the EMU Positions dataset. Again, Germany ranks first in terms of both these measures (see Fig. E in the “Online Appendix”). Its policy position was the most extreme, being on average 45.86 points away from the mean position of all other countries on the 0–100 scale. Likewise, Germany held the most intense preferences as it attached on average 8.4 points to an issue on a scale of 0–10. Again, there is a noticeable correlation between states’ bargaining success and both their preference extremity and salience (r = 0.58, p < 0.001 and r = 0.200, p < 0.001, respectively). Hence, the poor performance of Germany may stem from its more radical and intense policy positions compared to other actors.

This expectation can be evidenced by several qualitative cases. For instance, despite having held an extremely intense preference for opposing the stabilization program for Greece (position = 0; salience = 10; extremity = 80), Germany lost negotiations on this issue as the outcome envisaged the establishment of such a program (position 100). Another example concerns the scope of the Single Supervisory Mechanism, already reported in the “Methodology” section. Again, Germany took an extreme and intense position by requesting the SSM to cover only systemically important banks (position = 0; salience = 9.6; extremity = 80). But the outcome was considerably distant from the German preference as it widened the SSM coverage to all Eurozone banks with the supervision of smaller banks by national agencies unless the ECB decides otherwise (position = 70). Germany was also on the losing side on the issue of the suspension of EU funds that arose during the six-pack negotiations. While the German government strongly called for the withholding of EU funds from the member states that breach the fiscal deficit limit (position = 100; salience = 10; extremity = 57), the final compromise did not foresee such a solution (position = 0).

Conclusions

This paper has offered the systematic assessment of Germany’s bargaining success in relation to EU secondary legislation. Building on the power-based distributive bargaining theory, it has tested the hypothesis expecting Germany to be the most successful country in daily EU lawmaking. Overall, the results provide no support for this hypothesis: There is little evidence that Germany prevails in the negotiations on actual EU legislation. Quite the contrary, three empirical analyses have revealed that this state is one of the biggest losers of decision-making.

First, Germany was found to be among the least successful member states in terms of voting results in the Council. During the period 2009–2019, it was one of the most outvoted country in this institution, particularly in the most important policy areas: economic and monetary affairs, employment or internal market. A high propensity to contest EU legislation indicates that Germany is not capable of achieving satisfactory negotiation outcomes in the Council despite holding superior power resources. Second, Germany was found to be among the least successful countries in the negotiations on key EU legislative proposals in the period 1999–2009. It ranked third from the end in terms of preference attainment, achieving one of the largest salience-weighted distances between the policy position and the outcome. Third, Germany was found to be the least successful country in relation to legislative proposals reforming the Eurozone in 2010–2015. It attained the lowest level of unweighted and salience-weighted bargaining success out of all 28 member states. Hence, supplementing the findings of Lundgren et al. (2019), this result questions the narrative of Germany dictating the shape of the EU secondary legislation on the Eurozone reform.

The paper argues that the reason for such poor-showing of Germany is the nature of its preferences. It revealed a negative and significant correlation between a state’s bargaining success and both the extremity and intensity of its preference. Since Germany is found to hold the most extreme and salient policy positions on EU legislation in the years 1999–2009 and during the Eurozone reform negotiations, it may therefore be the least successful in achieving its ideal outcomes. However, it must be noted that choosing an extreme position can also be a strategy to pull the decision outcome closer to actor’s “real” preference. By holding extremist preferences Germany may extend the negotiation space, causing that the final compromise slightly shifts toward its stance. Hence, while being the least successful in attaining its “revealed” preferences, the German government may achieve higher success in relation to its “hidden” preferences by forcing other countries to move the outcome toward its extreme position. Unfortunately, I am not able to check this since none of the datasets used in the analysis contain information on hidden preferences.

Overall, this paper contributes to the scholarly knowledge on EU bargaining in two facets. First, it provides descriptive evidence from various datasets that seems to suggest that Germany has not been a particularly successful negotiator in daily EU lawmaking. Thus, while some earlier literature has portrayed this country as a hegemon in the European Union, this paper challenges this narrative with regard to the EU legislative process by showing that Germany fares surprisingly poorly in the negotiations on EU secondary legislation.

Second, the paper challenges the power-based distributive bargaining theory. Contrary to its expectations, holding large resources of economic, voting or network power does not automatically translate into higher bargaining success. This is evidenced by the very low bargaining success attained by Germany, but also by the other most powerful countries like France, Italy, Poland or Spain, as shown in this paper. Instead, the extremity and intensity of preferences seem to play a more relevant role. That is why I argue that Germany, despite being a “colossus” in terms of power resources, has “clay legs”—too extreme and intense preferences that trigger its bargaining success to “collapse.” Additionally, the paper questions the sociological institutionalism argument that bargaining success is evenly distributed among states due the operation of a consensus norm in the Council (Heisenberg 005). In fact, the case of Germany shows that there are clear “losers” in daily EU lawmaking.

However, the above findings do not explicitly mean that Germany is totally unsuccessful in EU lawmaking. The reason is that my analysis covers only formally adopted legislation, while it does not encompass legislative proposals that did not make it on the agenda or were not voted on. Hence, one cannot rule out that Germany may be successful in deciding what issues came up for negotiation or blocking them from being discussed. In fact, there are at least two examples of proposals that were prevented by Germany from being adopted: the Eurobonds during the Eurozone crisis (Schimmelfennig 2015) and the Takeover Directive in 2001 (Bailer 2011). In addition, although Germany was found to be the least successful member state on actual EU legislation, this country may still prevail in determining the EU primary legislation (i.e., treaties) which constitutes the legal basis for the secondary legislation analyzed in the paper or shaping the policy agenda in the European Council. Also, it is possible that Germany has a larger say in relation to less controversial proposals that were not covered in the DEUII and EMU Positions datasets. Future studies should explore these issues.