Introduction

The consolidation of a corpus of case law is linked to legal certainty in several ways. First, case law compensates legislators’ inability to identify all possible factual situations. Second, case law facilitates the identification of relevant regulations in a complex legal scenario fueled by institutional fragmentation, overlap** levels of government, and the multiplication of regulatory agencies. Third, a sound jurisprudential doctrine may prevent litigation and provide efficiency in judicial proceedings since conflicting interests and the burden of proof are clearly settled.

However, the positive effects of case law on the legal system are equally criticized on several grounds (Waldron 2012). The idea of judicial lawmaking is a challenge to the integrity of the rule of law and the principle of division of powers. In addition, case law may limit the flexibility courts need to apply the law according to the current social context. Finally, case law may also limit the scope of regulatory changes introduced by lawmakers.

From a different perspective, the efficiency dimension associated with case law may be overestimated since much of the litigious issues are not related to questions of law but controversies on the state of facts, proof, and evidence. Furthermore, litigation may be the consequence of the strategic behavior of the parties to the proceeding rather than questions of law. Hence, the elaboration of case law may be biased as a consequence of the strategic calculations of litigants which leads them to propose controversial cases in courts more likely to adopt a position close to their interests (Fon and Parisi 2006).

Theoretical considerations of the positive or negative impact of case law are not constant over time. Luppi and Parisi (2010) point out that those systems that are in an initial stage of development adopt the doctrine based on the binding character of case law (stare decisis) as a mechanism of consolidation and promotion of legal certainty. This principle takes preference over the objective of minimizing judicial errors. On the contrary, in those systems in which judicial errors generate a larger negative effect because of the binding nature of case law and in which a certain level of certainty has been reached, there are fewer incentives for the implementation of the stare decisis doctrine.

Beyond the potential impact of case law on legal certainty, their configuration as a source of law has been traditionally used to differentiate common law systems from civil law systems (Mattei 1988). Nevertheless, the differences between both systems may fade through an analysis of actual judicial practice and regulatory developments. Despite the fact that case law is not binding in continental law countries, the practice of courts suggests that case law has a persuasive role depending on circumstances such as the uniformity and consistency of courts’ jurisprudential doctrines (Cappelletti 1981; Fon and Parisi 2006). On the other hand, the possibility that the resolution of a case sets a judicial precedent constitutes an incentive for the innovative capacity of judges (Posner 2010) and thus the stability of case law would be lower than expected. In addition, the role of codification is increasing in common law systems. Considering the identified developments, the understanding by which civil law systems correspond to codification whereas common law systems are equivalent to case law constitutes an oversimplification (Weiss 2000).

The utilization of case law remains an unresolved issue that requires empirical examination emphasizing the incentives posed by diverse legal systems, regardless of their classification as either civil or common law systems. The fact that case law is not listed as source of law does not preclude it from being a mode of operation for the courts, and its use may be favored by circumstances such as avoiding decisions being reviewed by higher bodies or consolidating a jurisprudential criterion aimed at reducing litigation. However, the opportunity to conduct empirical research on case law is challenged for different reasons. First, the analysis required is resource-intensive as it depends on the systematic exploration of a large number of judgments covering different courts and sectors of activity. Second, the analysis of the integration of case law also requires a content analysis to codify context-related and specific variables of the issue at stake in judicial proceedings. Finally, the non-binding nature of case law and the prevalence of a codified system of regulations reduces attention to the role of case law in continental law systems.

The complexities surrounding the utilization of case law in continental law systems have contributed to a scarcity of empirical studies on the subject. To address this gap, the present study aims to identify the circumstances under which administrative courts may rely on case law. To achieve this, a content analysis of 2964 judgments from Spanish administrative courts is presented. The administrative courts in Spain are divided into single-member bodies (juzgados) and collegiate bodies (tribunales), and their jurisdiction is affected by the decentralized nature of the Spanish political system. Among the single-member bodies, the Juzgados de lo Contencioso Administrativo have jurisdiction only within the province where they are located. In contrast, the Juzgados Centrales de lo Contencioso Administrativo have jurisdiction throughout the national territory. As for the collegiate bodies, the Tribunales Superiores de Justicia (High Courts of Justice) correspond to the different regions or Autonomous Communities and represent the highest level of the judicial structure within each Autonomous Community, subject to the jurisdiction of the Supreme Court. The Sala de lo Contencioso Administrativo de la Audiencia Nacional (National High Court) has jurisdiction over the entire territory in specific matters such as appeals against acts of Ministers or Secretaries of State or against decisions of national administrative bodies such as the Central Administrative Court for Contractual Appels or the Central Economic-Administrative Court. Finally, the Sala de lo Contencioso-Administrativo del Tribunal Supremo (Supreme Court) is the ultimate judicial authority with jurisdiction throughout the entire national territory.

The remainder of the paper is organized as follows. The second section of this paper provides a critical analysis of the potential practical differences that may exist between civil law and common law systems with regard to the utilization of case law as a source of law. The third section of this paper sets forth the main hypotheses, describing two distinct patterns of integration of case law in court rulings: the self-referential model, which is based on the utilization of case law originating from the sentencing court, and the hierarchical model, which involves the integration of case law from other courts, typically those of final appeal. In the analysis of both the self-referential and hierarchical patterns of case law integration, this paper presents a novel perspective on regulatory complexity, emphasizing the importance of identifying the law in force rather than relying solely on formal indicators such as the length of laws or procedural regulations. The fourth section of this paper provides a detailed description of the database utilized in this study, as well as the essential characteristics of the judicial system analyzed. The database is composed of a representative sample of court rulings, which were classified by their respective court of origin and the administrative activity sector under consideration, covering the period from December 2016 to February 2018. In the fifth section, a logistic regression model is presented, which is used to estimate the extent of the utilization of case law and explore the circumstances that favor either the self-referential or hierarchical patterns of case law integration. Finally, the sixth section highlights the key findings of this study and identifies areas for future research in this field.

The Role of Case Law in Common Law and Civil Law Systems

The principle of separation of powers and its implications for the legislative–judiciary relationship or the understanding of legal certainty are at the basis of the different configurations of case law as a source of law. In common law systems, the doctrine of stare decisis turns case law into a source of law. This view of case law and the role of the judiciary is not reproduced in civil law systems in which the so-called doctrine of constant jurisprudence does not attribute binding character to case law, although it does recognize the influence of uniform judicial decisions (Fon and Parisi 2006).

The concept of legal certainty is a fundamental aspect that distinguishes legal systems. Legal certainty presents itself in different ways, including prohibiting arbitrary action by public authorities in applying the law to citizens, recognizing the law in force, and ensuring predictability in the application of the regulatory framework. In civil law systems, the prevalence of codified law is considered to be a necessary and sufficient condition to achieve legal certainty and predictability in the functioning of courts, which must follow the set of rules emanating from parliaments. The general assumption is that legal codes regulate all factual situations that may take place. Yet empirical analyzes that have tried to measure legal certainty present civil law systems slightly ahead of common law systems (Deffains and Kessedjian 2015). However, the indexes of legal certainty have been questioned for not taking into account the system’s capability to resolve legal issues (Siems 2017).

In practice, the objective of legal certainty may not always be achieved by the mere existence of a codified set of rules. The complexity of the law-making process with interdependent levels of government or political instability also produces a manifest lack of legal certainty in civil law systems (Merryman and Pérez-Perdomo 2007). The identification of the law in force or the vagueness of the legal terminology are also among the sources of legal uncertainty. On the other hand, systems based on the doctrine of stare decisis also present limitations in achieving legal certainty on several grounds. First, identifying accurate case law for a case can pose significant challenges. Second, the complexity of determining the similarity between case facts, changes in case law, or the existence of dissenting opinions in judicial resolutions also jeopardize the contribution of case law to legal certainty (Maxeiner 2006; Sedler 2009). In this sense, Cardozo (1921) pointed out that case law generates a constant process of correction of errors that are always present in any human evaluation. The doctrine of stare decisis is the result of a collective enterprise in which the parties to judicial proceedings and judges critically review previous judicial decisions on which their legal arguments can be based. In short, the doctrine of stare decisis involves a reasoned account of feasibility rather than a plain reference to previous judgments. The doctrine of stare decisis fundamentally implies the analysis of the elements involved in legal reasoning which are used to elaborate judicial decisions in other courts (Pin and Genova 2019). From this perspective, the two legal doctrines, stare decisis and constant jurisprudence, present a common ground, which is the reuse of legal reasoning in the analysis of the issue at stake by courts. The use of jurisprudential doctrine is then a strategy to overcome the problems of both an excessive volume of regulations in civil law systems and a limited specification of regulatory criteria in common law systems.

Several factors suggest the ongoing convergence process between the systems of both legal traditions (Pejovic 2001; Funken 2003; Mattei and Pes 2008; Rosen 1994). In the European context, the configuration of the European system of law determines that the application of regulations takes place in a decentralized manner (i.e., by national courts) while legal interpretation and the control of the principle of legality is centralized in the Court of Justice of the European Union. This configuration encourages the use of case law produced by the European court. Furthermore, non-compliance with the European Court of Justice´s interpretative criteria by national courts at any instance can be considered as a breach of the principle of loyal cooperation between European and national institutions (Article 4.3 Treaty on European Union).

The impact of case law on legal systems may not solely be determined by the legal doctrine of stare decisis and constant jurisprudence. Other factors, such as the varying interpretations of the relationship between the branches of government, also come into play. Within the common law tradition, laws establish a framework that permits the executive branch to create additional regulations that are context-specific. The flexibility of the law is seen as a desirable outcome of this system. Conversely, in civil law systems, the scope of maneuverability for the executive branch is defined more precisely. Regardless, in both types of legal systems, courts rely on case law as a means to strike a balance between situations with low regulatory density and high discretionary power on the side of the executive or high regulatory density and low discretionary power of executives. In both cases, the principle of legality may be at risk, thus courts serve as the ultimate voice of the legislative branch of government.

Other reasons related to how case law is created lead to a qualification of the differences between the two legal systems. Gerhardt has identified precedents not only from non-jurisdictional organs but also from institutions endowed with regulatory capacity in American constitutional doctrine (Gerhardt 2008). This interpretative capacity of certain regulatory agencies is a common characteristic in both systems. Moreover, the binding characteristic of case law in common law systems is not absolute. The influence of case law is weighted by circumstances such as the type of court generating the ruling, the types of questions of law and case facts, or the coherence in the jurisprudential doctrine (Dainow 1974).

In addition to their theoretical underpinnings, the distinctions and similarities between the common law and civil law doctrines have been analyzed in terms of their regulatory and economic outcomes. Empirical research has shown differences in efficiency in specific sectors, such as labor market regulations (Botero et al. 2004) or the resolution of disputes in civil and commercial matters (Djankov et al. 2001). However, the relationship between legal traditions and the efficiency of the economic system does not always yield clear-cut differences (Garoupa and Linguerre 2011; Lee et al. 2014).

To sum up, either theoretical considerations or empirical analysis blur the differences between civil law and common law systems as far as the use of case law is concerned.

Self-referencial and Hierarchical Patterns in the Aplication of Case Law

The Spanish legal system is rooted in the civil law tradition and does not recognize case law as a source of law. According to Article 1.6 of the Spanish Civil Code, only the consolidated jurisprudential doctrine from the Supreme Court is recognized as complementing the legal system. Despite this, case law still exerts some influence in the Spanish legal system, driven by factors such as the technical authority of higher courts and certain appeal proceedings aimed at ensuring uniform interpretation of the law (Díez Sastre 2008; Doménech Pascual 2013). In addition, certain jurisdictional bodies' case law (Supreme Court and Constitutional Court) may be invoked as grounds for appeals before the courts of last instance (Ortega Rivero 2002).

In spite of the fact that case law is not listed among the sources of law in the Spanish legal system, there are other legal principles that suggest considering a far-reaching impact of case law. The Spanish Constitutional Court has interpreted the principle of equality before the law (article 9 of the Spanish Constitution) in a way that has led to the use of case law to a certain extend. In its decision 103/1984 of November 12, 1984 (appeal no 94/1984), the Constitutional Court stated that the principle of equality encompasses not only equality before the law but also equality in the application of the law, which implies that the same court cannot arbitrarily modify the legal reasoning of its decisions in substantially the same cases. The Constitutional Court demands additional requirements for considering the binding nature of case law (i.e., identity of case facts, alterity of the parties to judicial proceedings, and identity of the court). The fulfillment of these additional requirements limits the scope of case law to the jurisprudential doctrine emanating from the same court (Díez Sastre 2008). Under such circumstances, the Spanish legal system may incentivize the development of a so-called self-referential system in the use of case law. In plain terms, Spanish courts would follow their own case law without considering decisions from other courts except in exceptional cases (Constitutional Court or Supreme Court in cases of appeals based on the unification of the interpretative criteria of law).

However, the limitations on the use of case law in Spanish courts are offset by several factors that incentivize judges and litigants to follow case law from other courts. One such factor is the strategic logic followed by parties in the proceedings, who may base their arguments on case law developed by higher courts that are likely to hear an appeal. In turn, courts may rely on this case law to minimize the likelihood of their decisions being overturned on appeal. Additionally, the appellate system's configuration may impact the use of case law from other courts, as it primarily reviews questions of law and precludes litigants from raising new issues and facts. This configuration thus requires courts to seek the legal reasoning of higher courts that focus on questions of law instead of questions of facts. As a result, courts are more likely to follow case law from courts of appeal and cassation than expected in a civil law system, resulting in a hierarchical pattern of using judicial decisions, usually from higher courts, in legal reasoning.

Nevertheless, the previous hypothesis presupposes that any case is subject to appeal or that the feasibility of bringing a case before the courts of appeal or the court of cassation does not change. However, the transition between the different stages in the appellate system may not be straightforward. Every stage in the appellate procedure represents a cost for litigants in terms of resources (i.e., time and economic resources). Moreover, litigants may upgrade their estimation of obtaining a positive judgment or a dismissal at every stage of the procedure. At some point, the legal risks (e.g., courts fees on the losing party) may outweigh the potential gain, and litigants may decide not to continue. Finally, the restrictive configuration of the appellate procedures may also reduce access to the courts of appeal or the court of cassation (i.e., judicial review limited to questions of law and not case facts, restriction to raise new questions, court fees, and value of claims as a threshold to access higher courts). Such circumstances would lead to the prediction of a more self-referential pattern in the use of case law when access to appellate procedures is less likely to occur.

In summary, whereas the legal principle of equality in the application of the law would lead to a limited use of case law and justify the predominance of a self-referential model, the consideration of other factors, such as the configuration of the appellate system and the litigants and judges’ strategic behavior, point to an alternative scenario characterized by a broader use of case law from higher courts.

Beyond the institutional factors and strategic choices of the various actors, the prevalence of one model or another in the use of case law may not be uniform across different areas of law. More concretely, the complexity of the regulatory framework may increase litigation and court caseloads. An increase in court caseloads may also generate a scenario of contradictory judgments from different courts due to the lack of binding force. Under such circumstances, litigants may have an incentive and a legal argument to bring a case before appellate courts due to the different decisions reached at courts of first instance. Once the question of law is settled at higher courts, case law would have a large impact on courts of first instance since a common ground for interpreting regulations would be established and litigants will predict the final outcomes of the judicial proceedings. Consequently, the higher the complexity of the legal framework, the more often courts would follow a hierarchical pattern in the use of case law.

Legal complexity can stem from both substantive and procedural regulations, and its measurement has primarily been derived from formal aspects of regulations. Scholars have focused on formal characteristics such as the length or number of laws (Kirchner 2012; Mora-Sanguinetti and Perez Valls 2020) as well as procedural laws related to contract enforcement or business formation (World Bank 2019; Djankov et al. 2001). The emphasis on formal aspects in these studies is based on the assumption that increased complexity leads to negative externalities in economic systems. Formal characteristics are more easily gathered and compared between legal systems, hence the predominance of this approach.

This paper approaches legal complexity from a different angle, focusing on the feasibility of identifying the current regulations applicable to the case. The ease of determining the legal framework applicable to the case cannot be taken for granted. First, different levels of government usually interact in complex lawmaking processes that blur the interpretation of regulations and the coherence between norms. Second, frequent revisions and adaptations of regulations also increase regulatory complexity. Third, in the context of European legal systems, European Union law introduces a new dimension of complexity since the domestic laws are supposed to be interpreted according to the wording and objectives of European law.

Based on the analysis presented above, the interplay between the complexity of the regulatory framework and the configuration of appeal procedures can result in either a self-referential or hierarchical model for incorporating case law, as illustrated in Fig. 1. In particular, the adoption of a hierarchical pattern is likely to be incentivized by factors such as greater access to appellate procedures and a high level of complexity in the regulatory framework. Conversely, the self-referential model is more likely to be observed when lodging an appeal is less common or when the regulatory framework is relatively simple, which may reduce the incentives to proceed to higher courts.

Fig. 1
figure 1

Source Own elaboration

Hierarchical and self-referential patterns of case law.

Full size image

An intermediate scenario arises when mixed incentives drive both the self-referential and hierarchical patterns. This occurs when access to appeal proceedings is limited, leading to incentives to follow a self-referential model, while a complex regulatory system incentivizes seeking the highest court. In these scenarios, the resources and incentives of litigants may be relevant factors. For example, the economic resources of litigants for legal services and court fees, the economic value of the issue at stake, or the ability to afford lengthy judicial proceedings may incentivize litigants to proceed to higher courts.

Data and Methods

This study presents the results of a content analysis of 2964 judgments from different areas of administrative activity (public procurement, taxation, liability, public employment, personal data protection, and fundamental rights) during the period 2016 (December)–2018 (February). The judgments collected were issued by collegiate judicial bodies of the Spanish administrative jurisdiction (High Courts of Justice, National High Court, and Supreme Court). Sampling was maximized based on the type of court and on judicial proceedings (first instance proceedings, appeal, and cassation). Consequently, case law was consequently classified according to the court of origin (High Courts of Justice, National Court, Supreme Court, Constitutional Court, Court of Justice of the European Union, European Court of Human Rights). Appendix A contains a description of the database in relation to the courts and sectors analyzed.

The Supreme Court is the final court of appeal for all cases except for constitutional issues. Regarding administrative cases, the 2015 reform of the cassation procedure determined that the Supreme Court can decide to hear only those appeals that present an objective interest of cassation in a similar vein to the system of certiorari. The discretionary and restrictive nature of the cassation appeal determines that the function of the Supreme Court’s jurisprudential doctrine as a complement of the legal system may not cover all controversies but only extend to those cases that present a far-reaching question of law.

On the other hand, the jurisdiction of the High Courts of Justice is territorially defined and corresponds to the 17 Spanish Autonomous Communities. The High Courts of Justice encompass three divisions (civil and criminal matters, administrative issues, and labor cases). As for the administrative division, High Courts of Justice have jurisdiction, in the first instance, over resolutions of local entities and regional administrative bodies, as well as over any other administrative act not expressly attributed to another judicial organ. High Courts of Justice also hear the appeals lodged against administrative courts of first instance. Finally, the National High Court´s jurisdiction covers the entire national territory but only on certain matters legally attributed to it (i.e., terrorism, organized crime, and economic crimes). Regarding administrative activity, the National High Court hears legal cases against the resolutions of the state level administrative organs or legal questions that exceed the territorial delimitation of one Autonomous Community.

The scope of case law analyzed in this study extends beyond the legal definition of jurisprudential doctrine as outlined in Article 1.6 of the Spanish Civil Code. Jurisprudential doctrine, as defined by the Code, is limited to the doctrine repeatedly established by the Supreme Court when interpreting law, customs, and general principles of law. Conversely, the term case law employed in this research encompasses all court judgments, regardless of their originating court. The reference to case law in this study pertains to any prior decision that forms part of the legal reasoning in the judgments under analysis.

This expanded definition of case law is supported by several justifications beyond the confines of the Spanish Civil Code. First, this study focuses on the practice of courts, which may utilize case law from various judicial bodies as interpretive criteria on a voluntary basis. In addition to the Supreme Court's legal doctrine, courts may refer to the jurisprudence of courts that may potentially hear an appeal, as per the incentives structure previously examined. Case law serves as a means of settling controversies efficiently by reapplying previous legal reasoning in analogous cases. Second, due to the limited nature of the cassation appeal, the Supreme Court cannot establish jurisprudential doctrine across all sectors. Thus, influential judicial decisions may originate from lower courts of first instance or courts of appeal. Finally, relevant case law may originate from international courts, such as the Court of Justice of the European Union or the European Court of Human Rights.

The present analysis involves a classification of administrative sectors based on the complexity of their regulatory frameworks. To this end, Table 1 depicts a classification of the complexity of the regulatory framework according to two dimensions: density and stability. Regulatory density refers to the volume of rules governing each sector of administrative activity from a dual perspective. A quantitative perspective addresses the number of regulations in force for each sector of administrative activity, while a qualitative perspective reflects the increasing complexity of the legal framework when regulations proceed from different levels of government. For example, the influence of European regulations on taxation and public procurement sectors compels national lawmakers to line up domestic regulations with the European framework. Furthermore, the European Court of Justice´s judgments serve as a benchmark for the application and interpretation of domestic regulations, and therefore awareness of the European law is required to interpret domestic regulations accurately. The level of complexity arising from a multilayered government also emanates from below since a legally decentralized system, such as the Spanish one, results in a considerable volume of regulations adopted by regional parliaments, as exemplified by public employment.

Table 1 Regulatory complexity in different sectors of administrative activity
Full size table

Concerning the stability dimension of the regulatory framework, the public procurement sector presents the highest number of regulatory changes considering all sectors of administrative activity. From the Act 13/1995 of 18 May 1995 on Public Administration Contracts until Act 9/2017 of 8 November 2017 on Public Sector Contracts, revisions and systematizations follow one another. Compared to other administrative sectors, the taxation sector undergoes more frequent regulatory changes due to the inherent complexity of the legal configuration of taxes, including deductions and quotas that are heavily influenced by economic and political factors, as well as the administrative proceedings for tax collection. This differs from other sectors, such as the protection of fundamental rights, which have a regulatory framework that derives from the Spanish Constitution and has not undergone significant reforms to its core legal texts. A more stable scenario of the regulatory framework is also found in the protection of personal data since regulations in this sector are quite recent in comparison with the rest of the sectors analyzed. Less regulatory stability would have the same impact as a high volume of regulations because both features of the regulatory system generate the possibility of contradictory judgments in the lower courts, increased uncertainty, and, therefore, greater incentives for litigants to appeal judicial decisions to higher courts or to rely on case law from higher courts.

Combining the different dimensions explored, the highest level of complexity is found in the public procurement and taxation sectors, an intermediate level corresponds to public employment, and a low level of complexity includes the protection of fundamental rights, personal data protection, and state liability sectors.

Results

Initially, the relevance of case law was not considered substantial, given the non-binding nature of jurisprudence in the Spanish legal system. Nevertheless, a self-referential pattern is expected to emerge due to the constitutional principle of equal treatment in the application of the law. The Spanish political system’s decentralized structure results in a fragmented legal system and a judicial practice that is more aligned with a self-referential pattern than a hierarchical one. This is due to the division of competences between the national, regional, and local levels and a judicial organization that is aligned with the decentralization of the State and public administration.

Origin of Case Law

After conducting an initial analysis of the data, it is evident that case law is utilized in the legal reasoning of 2257 out of 2964 judgments examined, representing a percentage of 76%, as presented in Table 2. The Supreme Court appears to be the primary source of case law, despite the fact that the jurisprudential doctrine it produces is not classified as a source of law. Both the High Courts of Justice and the High National Court primarily incorporate the case law developed by the Supreme Court. This finding indicates the existence of several factors that contribute to a hierarchical pattern in the use of case law within Spanish administrative courts.

Table 2 Number of judicial precedents by court of origin (rows) and sentencing courts (columns)
Full size table

A closer look at the results also suggests that a self-referential pattern is at work when considering the references to case law other than the Supreme Court´s decisions. The case law of the High Courts of Justice is mainly used by the High Courts of Justice themselves (44%) in contrast to the integration of that case law in the Supreme Court (20%) or the High National Court (3%). As for National High Court decisions, the Supreme Court only refers to them in 3% of cases and the High Courts of Justice in 5% of their judgments. In contrast, the incorporation of case law from the National High Court increases to 55% in the case of judgments issued by the same National High Court.

These results also show the minor impact of the case law from international courts on the legal reasoning of the administrative courts analyzed. Only 12% of the Supreme Courts’ judgments refer to European Court of Justice rulings. This percentage shrinks in the case of the High Courts of Justice (4%) and the National High Court (5%). These results also suggest a delayed reception of the European Court of Justice’s jurisprudential doctrine by domestic courts, which could be problematic considering the restrictive nature of the cassation appeal. Given that the jurisprudence of the European Court of Justice is primarily received by the Supreme Court and access to this court is somewhat restricted, the impact of European court case law on Spanish administrative courts may be limited, particularly in sectors such as public procurement and taxation, which are significantly influenced by European regulations. Furthermore, the influence of the European Court of Human Rights’ jurisprudential doctrine is negligible, as evidenced by its integration into only 1.4% of the cases resolved by the Supreme Court.

Regarding the characteristics of the regulatory framework, Table 3 depicts the origin of case law based on the different sectors of administrative activity that were analyzed. Some results are expected due to the characteristics of the sector and the jurisdiction of the courts. This is the case of the protection of fundamental rights, for which the frequency of case law from either the Constitutional Court or the European Court of Human Rights (72% and 4%, respectively) are higher than in the rest of the administrative sectors.

Table 3 Case law and sector of administrative activity
Full size table

In general terms, the integration of case law from the Supreme Court predominates across sectors. Nonetheless, the analysis reveals sector-specific disparities in the extent to which the Supreme Court's case law is incorporated. Notably, two sectors characterized by intricate regulatory frameworks, taxation, and public procurement, exhibit significant reliance on the Supreme Court's case law. Interestingly, the same pattern emerges in the state liability sector, despite its relatively low complexity. This outcome may be attributed to the existence of an established jurisprudential doctrine at the Supreme Court, which is a consequence of the regulatory framework's stability (Pastor Merchante 2015), and the incentives to appeal to the highest court to establish uniform criteria for the recognition of state liability.

Factors Explaining the Self-referential and the Hierarchical Patterns

This section aims to evaluate the patterns in the application of case law by administrative courts in Spain. Specifically, it seeks to determine the prevalence of self-referential or hierarchical patterns in the integration of case law. To achieve this objective, a logistic regression model (1) is utilized, which estimates the probability of utilizing case law from the same sentencing court vis-à-vis case law from other courts, taking into account several predictor or explanatory variables. The specification of the model is as follows:

$$Y\left( {jp} \right)\user2{ } = \user2{ }\beta_{0} + \user2{ }\beta_{1} \left( {COURT} \right) + \user2{ }\beta_{2} \left( {INSTANCE} \right) + \user2{ }\beta_{3} \left( {SECTOR} \right) + \user2{ }\beta_{3} \left( {LITIGANT} \right) + u$$
(1)

In this study, the dependent variable Y(jp) represents the probability of a court including its own case law (self-reference pattern) as opposed to integrating decisions from a third court. The constant is represented by β0, while the error term is denoted as u. The categorical variable COURT represents the courts being analyzed, which include the Supreme Court, High National Court, and High Courts of Justice. It is expected that courts will rely more on their own case law when the risk of reversal by higher courts is low, reflecting strategic incentives in the use of case law. The variable INSTANCE is also categorical, indicating whether the court is resolving the case in the first instance, on appeal, or in cassation proceedings. The restriction of appeal proceedings to questions of law may encourage the use of case law from higher courts as they typically deal with the interpretation of law rather than the evaluation of facts. The categorical variable SECTOR reflects the administrative sectors being analyzed, including taxation, public procurement, state liability, protection of fundamental rights, public employment, and personal data. This variable captures the impact of the complexity of the regulatory framework based on density and stability dimensions. Finally, the variable LITIGANT indicates whether the party initiating the judicial proceeding is an individual, a legal entity, or a third administrative organ, and serves as a proxy for litigants' resources. It is assumed that legal entities and administrative organs have more resources than individuals, which can influence their access to later stages of the appellate system. In cases where litigants have more resources, a hierarchical pattern is expected as they may have incentives to proceed to higher courts. The logistic regression models provide information about the statistical significance, coefficients, and standard errors for each explanatory variable. By examining the sign of the coefficients, it is possible to determine the effect of each variable on the likelihood of using a self-referential model. Positive coefficients indicate an increase in the probability of self-reference model in applying case law, while negative coefficients indicate a decrease, both relative to the reference category.

In the regression models, each of the previous variables are successively incorporated through different models, and their ability to improve the explanatory power is evaluated. The sequence of regression models exhibits a gradual enhancement in the goodness of fit as demonstrated by the Bayesian Information Criterion (BIC) estimations. The results, as presented in Table 4, demonstrate a remarkable improvement in predicting the self-referential pattern with the inclusion of the explanatory variables. Moreover, the results indicate that the court, instance, and sector variables significantly influence the integration of case law within administrative courts. Specifically, the prevalence of the self-referential pattern is primarily associated with the type of court and the instance of judicial proceedings. Model 1 indicates that the use of case law from the same court increases when judicial review on appeal is limited due to the specialized jurisdiction of the court, such as the National High Court, or the restrictive nature of the appellate system, such as the Supreme Court. Consequently, a greater limitation on judicial review tends to increase the role of the self-referential pattern. Furthermore, the positive coefficients for the Supreme Court and National High Court categories are significant and consistent in all four models estimated.

Table 4 Logistic regression on self-referential pattern of case law
Full size table

Despite the relevance of the type of court being subject to variation, the impact of the instance factor remains significant. To this regard, the likelihood of following a self-referential pattern diminishes from the initial instance to the appellate level. The regression analysis reveals a negative coefficient for the variable pertaining to courts operating at the appellate level, which indicates a greater propensity for these courts to adopt a hierarchical pattern as opposed to a self-referential one (models 3 and 4). This outcome can potentially be attributed to the distinct nature of appellate proceedings, wherein judicial review typically centers around matters of law rather than factual considerations. Consequently, the courts tend to place greater emphasis on the legal reasoning underlying the case law of third courts.

In order to further investigate the effect of different combinations of instance and court variables on the adoption of a self-referential pattern regarding case law, Table 5 presents the average marginal effect of logistic regression. The average marginal effect measures the average change in the probability of an event occurring as a result of a one-unit increase in a predictor variable, holding all other variables constant. In this case, the predictor variables are the different combinations of instance and court variables. By examining the average marginal effect of each combination, we can assess the consistency of the effect of these variables on the adoption of a self-referential pattern. The predicted values associated with a self-referential pattern decrease as progress is made from the first instance to appeal proceedings independently of the court analyzed. For instance, the predicted average probability of following a self-referential pattern in High Courts of Justice changes from 0.34 to 0.28 when proceedings change from first instance to appellate proceedings. These results are in line with the influence of the restrictive configuration of appellate proceedings on the prevalence of the hierarchical model. The causal mechanism for this result stresses the fact that higher court case law is more valuable when the issues to be settled are predominantly questions of law. Questions of law are more frequently analyzed in appellate proceedings than in first instance judicial proceedings. As for the latter, the average predicted value for a self-referential pattern is lower than in appellate proceedings. The previous effect associated with the judicial instance in which the controversy is settled does not cancel the effect of the type of court. The self-referential pattern is predominantly followed in the case of the Supreme Court in contrast to High Courts of Justice due to the feasibility of reviewing this latter court ruling even in the case of appellate proceedings. Yet in the case of the Supreme Court, the predicted probability of following a self-referential pattern is higher in cassation proceedings (0.68) than in first instance proceedings (0.62).

Table 5 Predictive margins by court and instance
Full size table

Regarding the sectors of administrative activity, the logistic regression results showed an influence of the SECTOR variable on the prevalence of a self-reference or hierarchical model. Specifically, models 3 and 4 in Table 4 reveal a notable negative coefficient, implying that courts tend to integrate third court case law more frequently than their own judicial decisions. Table 6 reflects the predictive margins depending on the sector of administrative activity.

Table 6 Predictive margins by sectors of administrative activity
Full size table

The logistic regression analysis results reveal a noteworthy decrease in the anticipated probability linked to the adoption of a self-referential pattern as the complexity of the administrative sector increases, specifically in cases related to public procurement or taxation, or as regulations become more recent, such as in personal data protection. This implies that courts may intentionally rely on third-party case law over their own judgments to confront the challenges posed by intricate regulatory environments. Notably, these marginal effects emphasize the importance of external sources of law in sha** judicial decision-making processes and the rule of law. This finding indicates the emergence of a hierarchical pattern in response to the regulatory complexity prevailing in certain administrative areas, which supports the proposed hypotheses. However, this complexity of the regulatory framework does not apply to cases pertaining to the protection of personal data. This exception may be due to the recent development of this sector of administrative activity and the need to seek uniform interpretative criteria from higher courts. This partially confirms the hypothesis regarding the prevalence of the hierarchical pattern in cases of complex regulatory systems.

To further explore the relationship between the prevalence of hierarchical patterns and complex regulatory systems, we analyzed the changes in the average marginal effect for all combinations of administrative sectors and types of courts. As presented in Table 7, the results partially confirm our hypothesis. The predicted probability of a self-referential pattern is highest for the Supreme Court and public employment issues (0.72) and lowest for High Courts of Justice and public procurement cases (0.29). In other words, the self-referential model is less frequent in those sectors of greater complexity and in courts whose decisions are subject to appeal. These findings provide further evidence of the importance of external sources of law and the strategic considerations that influence the use of self-referential or hierarchical patterns by courts in navigating complex regulatory environments.

Table 7 Predictive margins by court and sector
Full size table

Both variables seem to work consistently since it is observed that the effect of the regulatory complexity is intensified by the type of court which settles the case. The cases in which the complexity of the regulatory framework is lower, or there are more restrictions on lodging an appeal against a judicial decision due to the type of court, rank higher in the marginal effect compared to the prediction of adopting a self-referential pattern (e.g., Supreme Court in cases of public employment, state liability, or protection of fundamental rights). On the contrary, in cases of high complexity in the regulatory framework (i.e., taxation and public procurement) and in which decisions may be reverted by higher courts, the changes in marginal effect are lower. An example of this latter case is found in judicial proceedings before the High Courts of Justice on taxation and public procurement (i.e., predicted probability of following a self-referential pattern of 0.31 and 0.29). In these cases, the lower courts´ judicial decisions may be reverted by the Supreme Court, and the incentives to lodge an appeal before the Supreme Courts may be higher in complex questions of law that are supposed to be settled by the courts of last resort.

Finally, the results show no statistically significant results regarding the characteristics of the parties to the proceedings. Either the involvement of legal entities or administrative organs do not produce a significant effect on the prevalence of a self-referential or a hierarchical pattern. Due to the fact that a hierarchical pattern is expected in cases of appellate proceedings and higher courts, it was expected that the hierarchical pattern would be less likely in cases of individuals because resources are limited to go through the different judicial instances. Therefore, the lack of a significant impact of the type of litigant on the prevalence of a self-referential pattern may imply the appellate system does not necessarily impose a higher burden on individuals in contrast to legal entities or administrative organs.

Conclusions

The analysis provides a comprehensive examination of how case law is integrated into the legal rationale of judgments pronounced by the civil law courts as represented by the Spanish administrative courts. Based on a content analysis of 2964 judicial decisions from Spanish administrative courts, the study presents empirical evidence that accounts for differences in the use of case law across regulatory sectors, courts, and the instance of judicial proceedings. The use of case law may be considered a general feature in the legal reasoning of administrative courts in the cases analyzed. Although jurisprudential doctrine is not listed as a source of law, administrative courts have incentives to follow case law from either the same courts (self-referential pattern) or higher courts (hierarchical pattern).

The constitutional principle of equal treatment and the foundations of civil law systems would suggest that courts rely on their own prior decisions, creating a self-referential pattern in judicial decision-making. However, empirical evidence demonstrates that a hierarchical model is more common under certain circumstances. Specifically, when questions of law outweigh questions of fact, courts tend to integrate higher court case law as higher courts are focused on the former type of questions. One implication of this argument is that the use of a hierarchical pattern is linked to access to appellate courts. This means that when judicial decisions are likely to be revised by higher courts, the hierarchical model tends to be favored over the self-referential model.

In addition, the hierarchical pattern is also observed in cases with complex regulatory frameworks such as taxation or public procurement cases, where questions of law take precedence over questions of fact. These legal issues are primarily resolved in appeal proceedings before higher courts. As a result, regulatory complexity creates an incentive to use case law from higher courts.

On the contrary, a self-referential model is observed in cases in which the complexity of the regulatory system is reduced or in which the potential revision of judicial decisions by higher courts is limited due to several factors, such as in those cases settled in the court of last resort (i.e., Supreme Court), when a specialized jurisdiction is operating (i.e., National High Court), or when questions of fact take prevalence over questions of law, for example, in cases of first instance proceedings.

Furthermore, the study suggests that case law from international courts, such as the European Court of Justice and the European Court of Human Rights, has a limited impact on judicial decision-making processes. These results diminish the relevance of those international bodies as a catalyst for the convergence between civil law and common law systems.

The findings of the study contribute to the comparative analysis between civil and common law systems. Despite theoretical differences, courts in both legal systems have incentives to use case law, regardless of its binding nature. The study highlights the significance of contextual factors in sha** the judicial decision-making process, such as the complexity of the regulatory framework and the configuration of appellate procedures.