Introduction
The 20th century witnessed the emergence of an important, structural change in the way European countries had thus far understood the relationship between politics and the court system. Until that time, the governance of the judiciary had been mainly a responsibility of the executive and more particularly of the Ministry of Justice. However, at least since the end of the World War II, concerns about judicial independence, judicial accountability, and a better judicial performance have started to alter this scenario (Guarnieri, Reference Guarnieri2004; see Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009b: 57). Ministries of Justice have lost their monopoly on judicial governance, and in many countries, separate institutions with powers over the careers of judges or the management of the judiciary have been born (see Guarnieri, Reference Guarnieri2004, Reference Guarnieri2007: 174). Judicial governance no longer follows a single model. Instead, a diversity of approaches to judicial governance has emerged, enlarging the range of choices available to constitution makers and political actors.
Judicial governance can be defined as the set of institutions, rules, and practices in a jurisdiction that organize, facilitate, and regulate the exercise by the judicial branch of its function of the application of law to concrete cases. From this perspective, judicial governance comprises a wide range of functions, from allocation of resources and the judicial budget to the oversight of the quality of the system and from the control of judicial careers to the allocation of cases to judges.
Despite being central to the functioning of liberal systems based on the principles of rule of law and separation of powers, the topic of judicial governance has only recently started to gain prominence in law and social sciences literature (Voermans, Reference Voermans and Fernández Segado2003; Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009a, Reference Garoupa and Ginsburg2015; Bobek and Kosař, Reference Bobek and Kosař2014; Bunjevac, Reference Bunjevac2017; Preshova et al., Reference Preshova, Demjanovski and Nechev2017). Most likely, for this reason, there is no universally accepted classification of existing approaches to the governance of the judicial branch. Different authors propose different typologies, and although these sometimes overlap, the differences between them are still important. A first typology of models of judicial governance is that provided by Win Voermans. In his 2003 piece, the author made a distinction between what he called ‘Northern European’ and ‘Southern European’ judicial councils. According to his work, the latter were characterized as being ‘constitutionally rooted’ and as fulfilling functions related to judicial independence, such as the appointment of judges, promotion, or disciplinary measures (Voermans, Reference Voermans and Fernández Segado2003: 2134). Conversely, the functions of ‘Northern European’ councils were focused on the management of the judicial organization, such as, inter alia, the management of caseloads, the budgeting of courts, and quality care, to the exclusion of the management of judicial careers (Voermans, Reference Voermans and Fernández Segado2003: 2135). A different classification was provided by Bobek and Kosař. In their work on judicial governance in Central and Eastern Europe, the authors point to at least five approaches to judicial governance: the Ministry of Justice model, the judicial council model, the courts service model, hybrid models, and the socialist model, although the latter has already disappeared from the European continent (Bobek and Kosař, Reference Bobek and Kosař2014).
With an aim to understand the rationales behind the implementation of the judicial council model in Europe, this article follows essentially the classification provided by Bobek and Kosař, as this is a clear taxonomy that differentiates models of judicial governance according to their most important characteristics: the organ of State with powers over the judiciary and the type of powers that such organs hold. Having the notion of ‘power’ at the core of the classification, the taxonomy provided by these authors is particularly useful for a political analysis of choices on models of judicial governance. However, certain variations to this classification and the scores of cases were introduced for this research (Table A2). In this article, a judicial council is defined as a separate institution for judicial governance that has at least certain powers – even if limited – over the careers of judges. By this, I mean powers over appointment, promotion, discipline, separation from office, and so on. These institutions, judicial councils, constitute the object of study of the article. Besides them, we can classify the remaining forms of judicial governance in courts services and ministerial models. A courts service can be conceived as a separate institution that manages the judiciary and essentially lacks powers over judicial careers, its competences being mostly managerial. A Ministry of Justice model is defined as one in which the majority of powers, whether managerial or related to judicial careers, are concentrated in the Ministry of Justice. In addition, we can find a number of hybrid or sui generis models as well as authoritarian models of judicial governance in nondemocratic countries. Based on this definition, this article contributes to the literature in the field with a classification of European systems of judicial governance (Table 1) in 29 liberal democracies of the continent.Footnote 1 This classification is the basis for subsequent analyses carried out in the central part of this article.
Table 1 Models of judicial governance in European liberal democracies
a Includes the councils of England and Wales, Scotland, and Northern Ireland.
Own elaboration.
In Table 1 countries were classified into the aforementioned models of judicial governance.Footnote 2 As can be seen, the judicial council model has been implemented in 38% of countries covered. However, in analyzing the cases it was observed that although some countries had institutions for judicial governance with certain powers over judicial careers, the range of powers of these institutions was limited compared with other judicial councils. To account for this difference, the subcategory of ‘weak’ judicial councils was created, comprising Slovakia, Greece, and Estonia. These were borderline cases.
The courts’ service model covers exactly the same share of countries, being also present in 38% of cases. Therefore, this model can be considered as a solid institutional alternative to the judicial council model. It is worth remarking that courts services are not always totally deprived from powers over judicial careers, although these are very limited compared with strong judicial councils (Table A2).
The third approach is the Ministry of Justice model. As it is present in only 14% of the cases, this model can be considered as residual. Hybrid models comprise 10% of the cases.
The landscape in European liberal democracies is therefore one of a wide diversity of approaches to the governance of the judicial branch. However, we know very little about the reasons that explain the implementation of these models of judicial governance. To the best of this author’s knowledge, the only comprehensive, cross-country, large-N attempt at covering this topic is the excellent work by Garoupa and Ginsburg (Reference Garoupa and Ginsburg2015). This scarcity of cross-country causal research is striking, given the very relevant academic and sociopolitical implications of this question. First, it is essential to understand variation of modalities and subtypes of liberal systems of judicial governance. Second, it is central to the understanding of rationalities, incentives, and constraints underlying political choices of macrolevel institutions. And third, it is fundamental to understand the overlap between legal, political, and sociological phenomena in the design of legal institutions.
This article aims at contributing to the literature on this topic, empirically responding to the following question: how can we explain the diffusion of the judicial council model in European countries? To answer this research question, this article provides for a pioneering application of qualitative comparative analysis (QCA) to the topic of choices on models of judicial governance. Using QCA, this article will provide for both cross-country patterns and case-specific explanations which together shed an important light over this scarcely researched object of study.
As will be shown, the answer to the research question lies in combinations of four factors: legal families, authoritarian backgrounds, the role of constitutions in democratic transitions, and Europeanization pressures. Together, these four conditions show the usefulness of explaining political–juridical institutions in their wider social milieu. As I will show, the four aforementioned factors interacted, resulting in choices on models of judicial governance by both creating incentives and inducing mindsets as to which arrangement was best to organize the judicial branch. In particular, it will be showed that the implementation of the judicial council model is the result of incentives created by authoritarian backgrounds in countries that implement new constitutions in their transition to democracy. But these authoritarian backgrounds only lead to the creation of judicial councils in Romanistic law countries – where these institutions found a more fertile ground for diffusion – or in countries subject to Europeanizing pressures – as a signal of commitment to democracy and European integration, and expecting accession in return.
In this regard, the findings of the article differ from and complement those by Garoupa and Ginsburg (Reference Garoupa and Ginsburg2015). While these authors did not find evidence that the dichotomy civil law/common law countries had an impact in the implementation of judicial councils, this article finds evidence that focusing on a specific type of civil law family, the Romanistic one, has explanatory capacity. In addition, in an earlier work, Garoupa and Ginsburg (Reference Garoupa and Ginsburg2009b) had suggested the existence of authoritarian backgrounds as one of the explanations for the implementation of judicial councils. In their 2015 work, however, the authors focused instead on the impact of the level of democratic quality of countries. This article goes back to the idea of authoritarian backgrounds as the explanation of the implementation of judicial councils, at least in Europe, including it as part of the model. Finally, and most importantly, different from the work by Garoupa and Ginsburg (Reference Garoupa and Ginsburg2015: 128 ff.), which focuses on the autonomous impact of independent variables on the dependent variable, in this article I will use QCA to try to systematically show how different explanations interact in leading to the implementation of these institutions. This is the most important element of the article, one that structures its findings: while all of the conditions tested are important to explain the outcome, none of them is alone sufficient to produce it. Rather, it is their combination in specific manners that matters. This article will show in detail those combinations.
The remainder of this article is as follows. The next section presents the theoretical framework of this article. The following section presents the methodological aspects. The subsequent sections carry out the configurational analyses of the data. Next, I analyze qualitatively two deviant cases: Poland and Czech Republic. After this, I discuss my findings in more detail. The last section concludes.
Theory
The choices of models of judicial governance are the result of the impact of legal, political, and social factors that constrain and induce decision-makers into opting for a certain model rather than another. In this section, it is argued that such factors are related to the type of legal family, transitions from authoritarianism to democracy (and, connected to this, the relevance of the enactment of new constitutions), and Europeanization pressures. The causal impact of these conditions can be explained by a combination of rational choice calculations and sociological factors in each case.
Legal families
The use of legal origins as an explanatory variable is controversial in the social sciences. For instance, the relation between legal origins and economic development was recently contested in Dam’s influential work (Dam, Reference Dam2006). The idea that legal families could have an impact on the diffusion of models of judicial governance has only been infrequently and indirectly addressed by the literature in the field (however see Piana, Reference Piana2009: 818–819). In fact, when this explanation has been systematically tested, no evidence of its explanatory impact has been found (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2015: 133).
Against this background, however, I believe that legal families can impact choices of models of judicial governance for two reasons. One of the reasons has to do with the preexisting characteristics of legal systems and can be explained through rational choice theories. The other reason has to do with the processes of institutional imitation and diffusion, and it is better explained through sociological and ideational accounts.
From a rational choice perspective, certain characteristics of legal families might be more compatible with certain elements of specific models of judicial governance, thus creating an incentive to implement such models. For instance, Garoupa and Ginsburg associate common law countries with a model of judicial governance focused on accountability rather than independence:
Recruitment of the judiciary in common law countries has traditionally drawn from more senior lawyers who have a wider range of previous experience and socialization than do judges in the civil law jurisdictions. Therefore, external accountability has been a major motivating factor in shaping the design of judicial appointment systems. Compared to the civil law judiciaries, common law judges have relatively few opportunities for advancement, and hence there is less capacity for political authorities to use the promise of higher office to influence judicial decision making. Accordingly, appointments processes have received serious attention since judges are fairly immune from pressures once appointed (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009a: 112).
The authors also argue, although focusing on the US experience, that ‘because in common law systems, the judiciary is not a “career judiciary” in the civil law sense, there is less interest in having independent commissions handle discipline, promotions and reassignments, and greater emphasis on judicial appointments’ (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009a: 114). If judicial councils are created to take power over judicial careers out of the hands of politicians, then in common law countries, there is a weaker incentive to create such institutions because this problem is less pressing.
The sociological element has to do with processes of institutional diffusion within legal families. Elkins and Simmons define diffusion as ‘interdependent, but uncoordinated, decision making’ in which actors, such as governments, make autonomous decisions but factor in the decisions of their peers on similar issues (Elkins and Simmons, Reference Elkins and Simmons2005: 35). Interdependence means here that ‘the adoption of a practice by one actor alters the probability of its adoption by another’ (Elkins and Simmons, Reference Elkins and Simmons2005: 36). In his review of diffusion theories, Ginsburg suggests that diffusion operates better between countries with similar characteristics and that one of those characteristics would be legal tradition (Ginsburg, Reference Ginsburg, Whittington and Keleman2008: 93). This logic can be conceived as one in which ideational factors guide the choices of models of judicial governance by increasing the likelihood of a model being implemented in a country when countries of the same legal family have already implemented it. In particular, it can be hypothesized that countries in each legal family will follow the model implemented in the most important countries with that legal tradition: the French and Italian judicial council model in Romanistic countries, the German Ministry of Justice model in Germanistic countries, and the Swedish and British courts service/administration models in the Scandinavian law and common law legal families.
Authoritarian legacies and the enactment of new constitutions
The general political background of a country seems to matter for the adoption of organs for judicial governance. In their work, Garoupa and Ginsburg (Reference Garoupa and Ginsburg2015: 132) found evidence about the relevance of the level of democracy of a country to explain the adoption of judicial councils. This article tests the impact of related but different explanatory conditions: the existence of authoritarian backgrounds and of new constitutions in countries exiting authoritarianisms.
The literature in the field has suggested that judicial councils were selected as a mechanism for judicial governance in countries that had a strong tradition of executive interference in the court system, while countries that had a more established tradition of judicial independence opted for the courts service model (Bunjevac, Reference Bunjevac2017: 822–823). Of course, one of the strongest modalities of such executive interference has to do with the existence of an authoritarian regime. According to Garuopa and Ginsburg, ‘the motivating concern for adoption of councils in the French-Italian tradition was ensuring independence of the judiciary after periods of undemocratic rule’ (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009b: 58).
Again, the connection between authoritarian legacies and the judicial council model can be explained from two perspectives. The first explanation is that the implementation of judicial councils is rational for the new democratic political elites in the processes of a transition to democracy. The new democratic political elites depend on the consolidation of democracy for their survival, and this depends, in turn, on the existence of strong constraints on power and an independent judiciary. The second explanation is ideational. Once a number of countries have successfully transitioned to democracy and implemented judicial councils in that transition process, other new democracies will deem this arrangement as appropriate to facilitate the consolidation of the new system of government. The logic here would be similar to that of diffusion explained above, but instead of operating within legal families, it would follow a logic of imitation among new democracies. Magalhães et al. explain the adoption of judicial councils in Southern European countries as a reaction against the existing authoritarian background:
in each of these countries, a broad consensus was possible around a few major issues (…). The first was the transfer of all or most supervisory powers over the judiciary from the executive to revitalized ‘judicial councils’. These were now conceived as self-governing bodies of the judiciary that would absorb those executive powers that had been previously used to neutralize judicial independence: recruitment, appointment, disciplinary sanctions, promotions and transfers. (…) All four constitutions [of Greece, Italy, Portugal and Spain] included articles stating the independence of the judiciary from elected branches and assuring judicial life tenure, as well as banning special tribunals and unifying previously fragmented jurisdictional systems, a measure specially addressed at avoiding the repetition of past experiences… (Magalhães et al., Reference Magalhães, Guarnieri, Kaminis, Gunther, Diamandouros and Sotiropoulos2006: 146–147).
If the transition to democracy is associated with the implementation of a judicial council, then, sensu contrario, a lack of authoritarian experiences would mean a lack of incentives to implement such an arrangement. But of course, postauthoritarian adoption of judicial councils depends, in turn, on an additional crucial factor: the enactment of new constitutions for the postauthoritarian period. It is the enactment of this new constitution that will create the window of opportunity for the implementation of the new model of judicial governance. Conversely, countries returning to old constitutions in the postauthoritarian period will not implement this arrangement.
Europeanization
In his influential work on Europeanization, Olsen distinguished five meanings of this popular term. One of them, the most relevant to the topic of choices of models of judicial governance, defined Europeanization as the ‘central penetration of national systems of governance (…) [which] implies adapting national and sub-national systems of governance to a European political centre and European-wide norms’ (Olsen, Reference Olsen2002: 923–924). In this sense, Europeanization would be a top-down process in which policies designed at the European level are implemented in member states. Bulmer explains that a top-down Europeanization operates when there is ‘a “misfit” at the domestic level with EU requirements (…) [so that] adjustment pressure builds up’ at this level (Bulmer, Reference Bulmer, Graziano and Vink2006: 51). The subsequent national adaptation to EU norms could be explained from two perspectives. From the perspective of rational choice institutionalism, adaptation would depend on the existence of national veto points and facilitating factors, while from a more sociological perspective, adaptation is explained by the existence of ‘norm entrepreneurs’ seeking to persuade national actors to change their preferences in processes of social learning (Bulmer, Reference Bulmer, Graziano and Vink2006: 51–52).
Judicial governance became a major object of Europeanization when, from the 1990s, the European Union and the Council of Europe started to pay attention to judicial administration (Piana, Reference Piana2009: 816; Bobek and Kosař, Reference Bobek and Kosař2014: 1260). There are many milestones in the emergence of the judicial council model as a soft standard in Europe. According to Preshova et al., ‘due to the limited “hard” acquis on the judiciary in general and different legal traditions, the EU relies on so-called European standards sponsored by Council of European institutions’ (Preshova et al., Reference Preshova, Demjanovski and Nechev2017: 13). Since 1993, when the European Union established the Copenhagen criteria for accession, both this organization and the Council of Europe started to devote efforts to the creation of standards of court administration (Bobek and Kosař, Reference Bobek and Kosař2014: 1260). In 1994, for instance, the Council of Europe made recommendations in line with the creation of judicial councils as a means of guaranteeing judicial independence (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009a: 109; Garoupa and Ginsburg, Reference Garoupa and Ginsburg2015: 128). In the Agenda 2000, the Commission announced that it would report to the Council on the progress made by candidate countries from Central and Eastern Europe (CEE), including the monitoring of independence and self-government of the judiciary (Bobek and Kosař, Reference Bobek and Kosař2014: 1275–1276).
In the case of the European Union, Preshova et al. underline two reasons why this organization put so much emphasis on judicial reform: first, because of its central importance to the operation of the rule of law and democratization in candidate countries and second, because of the ‘European mandate’ for national courts with respect to accession, which places national courts in charge of applying EU law to particular cases (Preshova et al., Reference Preshova, Demjanovski and Nechev2017: 7). The implementation of judicial councils as a result of the dynamics of Europeanization could be deemed to be due to a misalignment between the institutions existing at the national level and the standard for judicial governance promoted at the supranational level. In this context, it would be rational for national actors to implement judicial councils, as the prospects of EU membership would create a strong incentive to do so. Piana also refers to the impact of Europeanization, although she offers a more sociological account:
The Commission did not dispose of any legal mechanism to oblige the candidate to adopt a specific organization model. In this policy field, there are no legally binding norms to be transmitted. The promotion of judicial independence and of judicial capacity has therefore to be relied upon the terms of a transnational discourse elaborated within epistemic communities and policy networks. This discourse is the outcome of policies aiming to promote the “rule of law” and screening used by think tanks and the international organizations in the period of democratic transition (Piana, Reference Piana2009: 823).
Configurational hypotheses
Thus far, based on the existing literature in the field, I have explained why factors such as the type of legal family, democratic transitions with new constitutions, and Europeanizing pressures should lead to the adoption of certain models of judicial governance, especially the judicial council model. However, these factors do not operate independently from each other. On the contrary, there are good reasons to believe that they interact, giving rise to causal configurations that help to explain the outcomes of each case. For this reason, in this subsection, I will present some configurational hypotheses: hypotheses regarding how conditions should interact to be sufficient for the implementation of a certain model of judicial governance. For reasons of space, rather than presenting expectations for every possible configuration of the explanatory conditions, I will present only those that are the most theoretically sound. These configurational hypotheses are tested in the empirical part of this article.
A first configurational hypothesis has to do with the interaction between authoritarian backgrounds and legal families. Above, I explained that the existence of a Romanistic civil law legal system is expected to be associated with the implementation of this approach to judicial governance. However, judicial governance in Romanistic civil law countries was, like all others, historically dominated by the Ministry of Justice. There must be, therefore, a historical event that changes political preferences, creating an incentive to break with the path-dependent survival of the Ministry of Justice model and to create a new institution for judicial governance. Such events are the processes of a transition to democracy, especially when a new constitution is created. It is this combination of factors rather than each of them separately that is sufficient to explain the implementation of a judicial council in a country. The explanation is that in the Romanistic area judicial councils became the standard option in processes of transition to democracy, as the institution is compatible with this legal family and as cases such as Italy and France set a model to be followed by other countries sharing similar characteristics and background. Therefore, the following hypothesis is proposed:
HYPOTHESIS 1: The combination of a Romanistic legal family AND a process of transition to democracy AND the enactment of a new constitution is sufficient to produce the implementation of a judicial council.
A second configurational hypothesis would, however, suggest that there is an alternative route to the implementation of the judicial council model. A corollary of Hypothesis 1 would be that countries that exit authoritarian regimes might not opt for a judicial council when they do not belong to the Romanistic legal family. The reason is that, outside of this family, the dynamics of postauthoritarian imitation and institutional diffusion will not take place. However, when these non-Romanistic countries aim at joining the European Union, a new, additional incentive is created. In a period in which the judicial council model has already become a soft standard in Europe, postauthoritarian countries willing to join the European Union will opt for this arrangement when enacting a new constitution, regardless of their legal family. With this option for a judicial council, these countries will be signaling EU partners their willingness to consolidate democracy and to join this supranational organization. For that reason:
HYPOTHESIS 2: The combination of Europeanizing pressures AND a process of transition to democracy AND the enactment of a new constitution is sufficient to produce the implementation of a judicial council.
Methodology
The main methodology employed by this article is QCA. Although the basic aspects of this methodology will be explained in this section, readers unfamiliar with QCA and who are willing to learn more can find detailed explanations about it in the different specialized books that have been published in recent years (Rihoux and Ragin, Reference Rihoux and Ragin2009; Schneider and Wagemann, Reference Schneider and Wagemann2012). To carry out the QCA, I used the fsQCA software. The crisp sets version of QCA was used, as the conditions were dichotomous in nature.
QCA of sufficient conditions seeks to identify which combinations of conditions suffice for the production of an outcome. To do so, and to attain parsimony, QCA uses a Truth Table, which presents all possible configurations of explanatory conditions. It also uses Boolean minimization, by which conditions that are logically irrelevant for the production of a result are dropped from the solution. QCA is also able to include counterfactuals in the minimization process, that is, configurations of conditions that did not empirically occur. QCA solutions are defined as follows: solutions that do not include any counterfactuals are called ‘complex solutions’, solutions that include all counterfactuals are called ‘parsimonious solutions’, and solutions that only include counterfactuals that are theoretically sound are called ‘intermediate solutions’. The traditional consensus among QCA methodologists has been that intermediate solutions were to be preferred (Ragin, Reference Ragin, Rihoux and Ragin2009: 111). For that reason, this article presents intermediate solutions only.
To test the research hypotheses, all countries included in Table 1, except for hybrid cases, are included in the analysis. Hybrid cases were excluded following a conservative strategy: as it was difficult to assign them to a particular model of judicial governance, they were excluded to increase the reliability of the analysis for the cases included. In addition, the readers will notice that some European countries are not included in Table 1. The reason is that they were not part of the sample. In QCA, case selection is based on ‘an area of homogeneity’ in which cases share ‘enough background characteristics’ as to make them comparable (Berg-Schlosser and De Meur, Reference Berg-Schlosser, De Meur, Rihoux and Ragin2019: 20). The definition of this area of homogeneity, in this article, includes all European liberal democracies to the exclusion of microstates. The 2018 Liberal Democracy Index by the V-Dem Project is followed to determine which countries classify as liberal democracies (Coppedge et al., Reference Coppedge2019): countries scoring 0.5 or more in this index are included in the sample. Note that according to this operationalization Poland is included in the sample (V-Dem Score: 0.55), despite that different authors have documented the process of illiberalization and rule of law backsliding in this country (most notably Sadurski, Reference Sadurski2019). This has analytical implications, to the point that Poland will be the object of a specific analysis later in this article. Countries that are not liberal democracies and microstates are thus excluded, as given their particular characteristics, judicial governance is deemed to follow in those cases different dynamics, not comparable to the sample of this article. Small states of the European Union, such as Cyprus, Luxembourg and Malta, are however included. The final sample comprises 26 countries. Finally, note that QCA operates on the assumption of ‘modest generalization’ (Berg-Schlosser et al., Reference Berg-Schlosser, De Meur, Rihoux, Ragin, Rihoux and Ragin2009: 11–12), so that researchers should be careful in generalizing the findings of the article beyond the sample countries included and, especially, to countries explicitly excluded because of their heterogeneity of background characteristics.
For the outcome (the model of judicial governance in each country), my classification is similar but not identical to that provided by Bobek and Kosař (Reference Bobek and Kosař2014). Table 1 shows the classification of countries, and Table A2 provides for a justification of all scores. As the definition of judicial councils focuses on the range of powers of the institution over the careers of judges, the Justice Scoreboard 2016 was particularly useful to complete the classification (European Commission and Directorate-General for Justice and Consumers, 2016). However, this source does not cover some of the countries analyzed, so an independent, qualitative research on some of the cases had to be carried out. When the Justice Scoreboard 2016 did not cover a particular case, I used as a first supplementary source Bobek and Kosař (Reference Bobek and Kosař2014). If in any instance my score is different from theirs, I justify this in Table A2. Finally, in cases not covered by any of these two sources I carried out my own research and justify the score assigned in Table A2.
Regarding legal families, to determine which countries belong to the Romanistic legal family, I used the dataset provided by La Porta et al. (Reference La Porta, Lopez-de-Silanes and Shleifer2008). Note that my focus here is in the distinction between Romanistic civil law countries and all others, contrary to Garoupa and Ginsburg (Reference Garoupa and Ginsburg2015: 132–133) who focus on the general civil law/common law distinction. As will be shown in the empirical part of this article, this has important consequences for the findings of the research. For authoritarian backgrounds, I mainly follow the CSIC database on Transitional Justice and Memory in the European Union (CSIC-IPP, 2013), although I extended it to countries not originally covered by it and corrected it when necessary. The data on whether countries enacted new constitutions at the end of repressive regimes and about EU enlargement waves were gathered autonomously by the author. Table 2 specifies the operationalization of such data into the crisp set conditions of the analysis.
Table 2 Explanatory conditions and their values
Following the best practices in the use of QCA (Schneider and Wagemann, Reference Schneider and Wagemann2010), the Truth Table (Table 3) and the data matrix (Table A1) are provided. Regarding the solution, the article only presents the analysis of sufficient conditions for the outcome ‘judicial council’ (Table 4), which is the object of study of the research. No analysis for the negation of the outcome is displayed, for a substantive reason: the negation of the outcome groups together very different models (courts services and Ministry of Justice models). An analysis of sufficient conditions for the negation of the outcome would be analytically incoherent, grouping together cases that do not belong in the same category.
Table 3 Truth Table
Table 4 Intermediate solution for judicial councils
Solution coverage: 1
Solution consistency: 0.846
Finally, the analysis of sufficient conditions in Table 4 includes indicators of coverage and consistency. Coverage indicates the proportion of cases with the outcome under study (in this case, a judicial council) that are explained by a path or a solution, while consistency explains the share of cases in a certain path or solution that have the outcome under study (see, for further explanation, Schneider and Wagemann, Reference Schneider and Wagemann2013: 123–139).
The dissemination of the Judicial Council model: configurational analysis of sufficient conditions
To understand choices behind the adoption of the judicial council model in European liberal democracies, this section performs an analysis of sufficient conditions, showing the intermediate solution.
In constructing the Truth Table (Table 3), a contradictory row was found. The deviant cases (underlined in the table) were those of Poland and the Czech Republic. Contradictory configurations often point at the need to find out additional explanations that can resolve the contradiction. To maximize the coverage of the analysis and so that all countries with a judicial council are explained, the contradictory row was included in the minimization. But the deviant cases are indicated and explained autonomously, and a specific section of this article provides for a qualitative analysis of these accounting for the causes of this contradiction.
In addition, in the Truth Table (Table 3) there are some configurations without empirical cases. These are the counterfactuals that were referred to in the methodological section of this article. QCA is able to deal with counterfactuals in a sophisticated manner: including them in the minimization when there are reasonable theoretical grounds to do so. This ‘intermediate solution’ allows for an informed, transparent use of counterfactuals. The selection of directional expectations is made explicit in this section, following the recommendation of transparency as a good practice in the use of QCA (see Schneider and Wagemann, Reference Schneider and Wagemann2010).
To produce the intermediate solution for the adoption of judicial councils (Table 4), directional expectations had to be specified. Authoritarian backgrounds, Europeanizing pressures, and the Romanistic legal family were expected to contribute to the outcome when present. The survival of an old constitution was expected to contribute to the outcome when absent. The model achieved had perfect coverage, meaning that all countries with a judicial council can be explained by this solution. The consistency of the model is 0.846, meaning that almost 85% of the cases explained by the model were countries with a judicial council, thus accounting for and acknowledging the existence of contradictory cases.
In the first path, which covers almost 55% of the cases, we observe that belonging to the Romanistic civil law tradition combined with having adopted a new constitution in the transition from a dictatorship to democracy is sufficient for the creation of a judicial council. In this regard, this path largely confirms the intuitions of the first research hypothesis (H1). The path includes the first councils in Europe, namely, the French (1946) and the Italian (1958) ones. For Garoupa and Ginsburg, the creation of these two councils had to do with a concern about the excessive politicization of the judiciary, which as a consequence, resulted in the granting of extensive independence to the judicial powers (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009a: 107). Once these two institutions were created, the template provided by France and Italy was followed by other countries that shared a similar legal culture and, especially, that were in their area of influence. Moreover, this template was followed exclusively when a critical juncture took place: a process of transition from dictatorship to democracy in which a new constitution was enacted. In the contexts of Romanistic law countries exiting authoritarian experiences, the creation of a judicial council was deemed as an optimal solution for consolidating the separation of powers and for ensuring the creation of an independent judiciary. Countries in the Romanistic area that did not undergo this type of process of transition to democracy or that did not implement a new constitution during such a transition did not implement a judicial council. Thus, this path very well fits the assessment by Garoupa and Ginsburg presented above, according to which the rationale for the adoption of judicial councils was ensuring the independence of the judiciary after undemocratic periods (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2009a: 110). Despite that in a subsequent work the authors did not find a correlation between (common law) legal families and the adoption of judicial councils (Garoupa and Ginsburg, Reference Garoupa and Ginsburg2015: 132), this article provides evidence that, in Europe, the presence of a Romanistic legal family has explanatory impact. The path also fits the assertion by Guarnieri, according to which the Italian model became ‘a model for Spain and Portugal in their post-authoritarian periods’ (Guarnieri, Reference Guarnieri2004: 178).
The second path shows the importance of the dynamics of Europeanization for the adoption of the judicial council model. The explanatory capacity of Europeanization and the pressure by European institutions had been advanced by some authors (Bobek and Kosař, Reference Bobek and Kosař2014; Garoupa and Ginsburg, Reference Garoupa and Ginsburg2015: 128). This path provides for systematic evidence in support of the theory and explains in detail how it operates. In this path, the type of legal family is irrelevant. Instead, joining the Union as part of the 2004 and subsequent enlargements becomes crucial for the adoption of a judicial council. This condition suggests that countries that joined the European Union after the point in time when the judicial council model had already been consolidated as a soft standard in the continent felt strongly compelled to adopt this institution. As stated by Bobek and Kosař, the judicial council model ‘has been exported through the European institutions and marketed as the “Euro-solution” for the judicial reform across the CEE’ (Bobek and Kosař, Reference Bobek and Kosař2014: 1274). Preshova et al. are bold in their interpretation of the impact of the waves of accession on the adoption of the judicial council model:
While judicial reforms were not seen as a very important issue for the so-called old Member States, they have become a centre of interest with the EU accession of the Central Eastern European Countries (CEECs), and now even more so for the Western Balkan countries aspiring to become members of the EU (Preshova et al., Reference Preshova, Demjanovski and Nechev2017: 7).
However, the impact of Europeanizing pressures took place in connection with two other conditions: first, the transition from an autocratic to a democratic regime and, in that context, the corresponding implementation of a new constitution. The path, thus, confirms the intuitions of the second hypothesis (H2), suggesting that countries in transition to democracy implemented this model of judicial governance because of the incentive provided by EU membership.
The deviant cases
The unexpected survival of the Ministry of Justice model in the Czech Republic
In constructing the Truth Table (Table 3), it was observed that a row was contradictory and included two deviant cases. One of such deviant cases was that of the Czech Republic, which, unlike other cases in similar circumstances, did not implement a judicial council and followed the Ministry of Justice model. An in-depth analysis of the case revealed the reason for this striking phenomenon. The Czech case has been called ‘the “black sheep” of the region’ and an ‘outlier case’, as it did not result in the implementation of a judicial council, although interestingly it has also been praised as a case of success in achieving a high level of independence (Kosař, Reference Kosař2017: 97).
The reason why a judicial council was not created in this country was simply that despite attempts to implement this system, the Czech parliament rejected it. A number of arguments have been put forward to explain this rejection, including the fear of judicial corporatism and elitism, the need to amend the Czech constitution, the lack of support for the Minister of Justice at the time, and the lack of agreement on this model among the very judges (Kosař, Reference Kosař2016: 183–184).
One additional, interesting element in the case of the Czech Republic is that, once the attempt to implement a judicial council failed, the country opted for the Ministry of Justice mode. The statements by Bobek and Kosař are telling as to why:
one may even ask why the CEE countries, most of which were heavily influenced by the German and Austrian legal tradition, were asked to opt for the Italian model of court administration. Had the choice been phrased as either the German/Austrian model or the Italian model instead of either the German/Austrian model or the “European” model, the answer of CEE countries would have been different (Bobek and Kosař, Reference Bobek and Kosař2014: 1282).
The case of the Czech Republic seems to confirm this intuition. The weight of the germanistic legal family in this country might explain why once the judicial council model failed, the Czech Republic persevered in maintaining the Ministry of Justice model, following the example of Germany and Austria. The Czech model evolved over the years into a system in which the Ministry of Justice shares responsibility with the courts’ presidents (Bobek and Kosař, Reference Bobek and Kosař2014: 1288). According to Kosař,
Czech court presidents have managed, step-by-step, to erode the Minister’s sphere of influence and have themselves become the most powerful players in the Czech judiciary, able to wield the most effective “stick” (disciplinary motion) and “carrot” (promotion) against individual judges. The Czech court presidents are thus both protectors of judicial independence and simultaneously a threat to it (Kosař, Reference Kosař2017: 97).
A very sui generis approach to the courts service model in Poland
Another interesting finding has to do with Poland. Although the 2018 V-Dem Liberal Democracy Index gives this country a score of 0.55, there is an emerging consensus in the literature that Poland has undergone a process of rule of law backsliding (Bugarič and Ginsburg, Reference Bugarič and Ginsburg2016; Pech and Schepple, Reference Pech and Scheppele2017; Sadurski, Reference Sadurski2019). It is thus not surprising that the Polish case is contradictory: following in-depth analyses such as those provided by Sadurski (2018, Reference Sadurski2019) it is very much questionable that Poland still belongs in the family of European liberal countries, and thus in the ‘area of homogeneity’ of this article. In any case, Poland was a deviant case in a path in which judicial councils were predominant. Why did Poland opt for an institution for judicial governance without substantive powers over the careers of judges?
The answer to this lies precisely in its process of rule of law backsliding. Poland underwent Europeanizing pressures at the exit of its authoritarian regime, so the judicial council model should have been the default option. But the process of rule of law backsliding might have pushed the country away from this model, and towards the adoption of an institution which is more amenable to executive control (Bugarič and Ginsburg, Reference Bugarič and Ginsburg2016; Sadurski, Reference Sadurski2018). In this context, the implementation of a pseudo-courts service model can be a good excuse to strip the independent organ for judicial governance of its powers. The case of Poland is thus exceptional. The country can no longer be said to have a proper judicial council. But it would be optimistic to consider the institution for judicial governance of this country as simply an equivalent of the liberal version of the courts service model. Instead, Poland can be simply deemed to have a disempowered organ for judicial governance that is the result of the specific political context of this country and its process of rule of law backsliding and illiberalization.
Discussion
The aim of this article was to explain the implementation of the judicial council model in European liberal countries, and to do so through the exploration of four main factors: legal families, authoritarian backgrounds, constitutions in democratic transitions, and Europeanizing pressures. Using QCA, this study has confirmed the explanatory capacity of these four conditions, showing evidence in favor of the two configurational hypotheses presented above.
With these findings, I hope to have contributed to shedding light on an important aspect of the design of an under-explored legal–political institution. More particularly, contributing to the literature on political decision-making, the article provides empirical knowledge regarding the rationales, incentives, and constraints behind the choices of institutions made by political actors. Contributing to the literature on comparative legal and political institutions, the study also shows that options for implementing models of judicial governance are not the result of random or idiosyncratic factors but instead follow cross-country patterns.
The findings of this article help us to refine the existing knowledge about models of judicial governance. They showed that the judicial council model was implemented in relation to a very important historical circumstance: the existence of authoritarian pasts. Judicial councils in the sample were always created against the background of an important historic trauma. In countries exiting authoritarianism, the priority seemed to be setting guarantees that the judiciary would remain independent once the new democracy was established to meet one of the basic requirements of a system based on the rule of law. In Romanistic law countries, the implementation of the councils after authoritarian experiences seemed to be an automatic response when new constitutions were enacted. We can talk in this case about processes of institutional diffusion and imitation within this legal family in response to a common historical trauma. However, as Bobek and Kosař suggest, for many other countries, the implementation of the judicial council model also had a lot to do with Europeanizing pressures, rather than being the result of unconstrained processes of institutional imitation.
The findings of this article also open some interesting lines of inquiry. First, as it was shown, the spread of the judicial council model has heavily relied on Europeanizing pressures. It remains to be seen what will happen in the absence of such pressures when either the prospect of accession to the European Union disappears for some candidate countries or when accession has already been achieved. In this latter regard, the examples of Poland and Hungary are telling.
Second, close attention will have to be paid to the evolution of the courts service model. The expansion of this model in its ‘natural’ area of diffusion – common law and Scandinavian countries – is almost complete. However, a number of Romanistic (Malta, Belgium, the Netherlands) and Germanistic (Latvia) countries have also opted for this model. The question that emerges is whether the courts service model will be considered an option in Romano-Germanic countries in which the judicial council model is now solidly anchored, especially in those that implemented that model as a response to authoritarian experiences. In the case of a positive response, a big challenge will have to be addressed: how to articulate a transition to this model of judicial governance which does not translate into a deterioration of judicial independence and the systems of checks and balances.
A third and final question has to do with the Ministry of Justice model. This model exists now in a tiny minority of countries. Three out of the four instances in which it exists are in Germanistic law countries. However, there are more Germanistic countries that have adopted the judicial council model than there are countries that have remained faithful to the Ministry of Justice model, so this latter model cannot even be said to represent the standard option in this legal tradition. The question is how long this model can survive in these circumstances, in an époque in which its basic features seem to be at odds with the approach to judicial independence that has become dominant in the European continent.
Conclusion
Judicial councils are a very important institutional innovation in liberal democracies. These institutions are expected to play an important role in guaranteeing the separation of powers, taking the competences over the management of judicial careers away from the hands of politicians. This article has showed the factors that explain the implementation of these institutions in European liberal democracies. Judicial councils were created when countries transiting from authoritarianism to democracy enacted new constitutions, in either Romanistic law countries or in countries subject to Europeanizing pressures.
Judicial councils have both a political and a juridical dimension. The findings of the article are in line with the dual nature of these institutions, and show the importance of combining sociopolitical and socio-legal explanations to understand the rationales behind their implementation.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1755773919000298
Acknowledgements
A previous version of this article was presented at the 2018 General Conference of the European Consortium for Political Research. I want to thank all participants in my panel, as well as Calum Young and the anonymous reviewers, for valuable comments on this article. Usual disclaimer applies.
Appendix
Table A1 QCA dataset for replication of analyses
Table A2 Justification of assigned values for outcome
JC = Judicial Council; JC(w) = Judicial Council (weak); CS = Courts Service; MJ = Ministry of Justice; H = Hybrid; JS2016 = Justice Scoreboard 2016.
