Lack of Communication Can Be Trouble: Brazilian Federal Supreme Court Decision-Making Behavior and Self-Promotion Strategies

Abstract

The post-1988 period in Brazil has brought as a consequence a greater role for the Judiciary in the political game, a fact that can be derived either from institutional arrangements or from strategies undertaken by its agents. Focusing on the second approach, the article explores whether exists a communication strategy between the Brazilian Federal Supreme Court (STF) and the public. According to the literature, in spite of the fact that the Judiciary differs from other political actors because it does not rely on the electorate, it is similar in terms of the need to provide transparency in its decisions, in order to achieve institutional legitimacy. In the Brazilian case, there is a constant externalization of messages directed at symbols of justice, objectivity, and impartiality and, since mid-2000’s, there has been an increase in the institution’s transparency, with emphasis on the televised transmission of its plenary sessions and a detailed institutional web page containing news and complete content of rulings. Considering that, from a comparative perspective, the constitutional court has become extremely exposed, one wonders: is there institutional selectivity in relation to its decisions? In order to answer this question, press releases from the Federal Supreme Court and judicial decisions were examined, over a 16-year time frame, from 2000 to 2016. The research utilized mixed methods: textual data mining techniques and statistics. The findings suggest that there is an intention, consistent throughout the time period studied, to transmit an image of an activist court.

Keywords
Judicial behavior; Judiciary and politics; Brazilian Federal Supreme Court; communication; institutional legitimacy

Resumo

O pós-1988 trouxe maior protagonismo do Judiciário no jogo político, fato que pode derivar tanto de arranjos institucionais quanto de estratégias de seus agentes. Focando na segunda abordagem, discute-se se há uma estratégia de comunicação do Supremo Tribunal Federal (STF) com o público. De acordo com a literatura, embora o Judiciário se diferencie de outros atores políticos por não necessitar do eleitorado, assemelha-se quanto à necessidade de proporcionar transparência às suas decisões, de modo a alcançar legitimidade institucional. No caso brasileiro, há constantemente uma exteriorização de mensagens direcionadas a símbolos de justiça, objetividade e imparcialidade, e houve, desde meados dos anos 2000, um recrudescimento da visibilidade daquela instituição, destacando-se a transmissão televisionada de suas sessões plenárias e uma detalhada página institucional que contém notícias e inteiro teor de julgados. Considerando que, em perspectiva comparada, a corte constitucional tornou-se extremamente exposta, questiona-se: há seletividade institucional em relação às decisões promovidas? Para responder a essa pergunta, foram observados comunicados de imprensa do STF e decisões judiciais, em um recorte temporal de 2000 a 2016. A pesquisa utilizou métodos mistos: técnica de mineração de dados textuais e estatística. Os resultados encontrados sugerem que há uma intenção, constante ao longo da série temporal, de enfatizar a imagem de tribunal ativista.

Palavras-chave
Comportamento judicial; Judiciário e política; Supremo Tribunal Federal; comunicação; legitimidade institucional

“[S]upreme courts occasionally play an enlightening role.” Ministro Roberto Barroso

Introduction

Baum (2006) introduces his book, Judges and their Audiences: A Perspective on Judicial Behavior, by describing several examples of the Supreme Court of the United States of America (USA), exposing their opinions in major newspapers and noticing that, despite the existence of miscellaneous speeches, it is possible to observe a clear intention of communication with different audiences. A similar phenomenon can also be observed in the various courts of other democratic countries. In Brazil, the frequency of such opinions has been spurring debates and opening new channels of communication with society.

For ordinary political actors, it is mandatory to establish a positive relationship with the public, considering that elections are the main motivation for seeking a communication strategy (DOWNS, 1957; POWELL JR., 2000; VANBERG, 2005; STATON, 2010). However, what motivation would make the communication strategy of Superior Court judges interesting, as they do not directly depend on the electorate? It seems to be a matter of great relevance, given the enhanced level of protagonism currently experienced by the Judiciary.

The debate is also related to the theory of Institutional Support outlined by Easton (1975), according to which it is possible to conclude that people who are often more attentive to the courts would be more likely to accept the mythology of judicial neutrality and objectivity in decision-making (CALDEIRA and GIBSON, 1995; GIBSON, CALDEIRA and BAIRD, 1998), in order to sustain and popularize a deeper belief in the legitimacy of the court as an institution.

Such a view would be associated with the perception that judges are, in fact, different. Thus, they trust the law rather than values of decision-making considered “objective”² and technical, which would strongly contribute to the legitimacy of the courts before society. Therefore, greater security for monitoring and complying with court decisions would be afforded.

Unlike elected politicians, Federal Supreme Court members are cooperatively nominated among the Executive and Legislative branches (appointed by the Federal President and confirmed by an absolute majority of the Federal Senate). They hold permanent mandates and, as they occupy unelected roles, the chances of dismissal are almost null.

Nonetheless, it is possible to observe a consensus in the literature that, although there are contextual specificities, members of constitutional courts are not exempt from politics. Posner (2005) mentions that conflicts between Congress and the US Supreme Court often yield remarkable cases, which can also be observed in Brazil, especially in recent decades. Therefore, parallel behavior to that of ordinary politicians can be expected of the Supreme Court during its exposure to public observation.

The visibility of the Court’s decisions was enhanced alongside the strengthened position it received from the Federal Constitution of 1988 (CF/88). Thus, new configurations emerged during the same period, such as the creation of public hearings, live transmission of sessions on open-air television, as well as an increase in the frequency of interviews and other articles in the press.

The Supreme Federal Court is constantly quoted as the only such court to have its plenary sessions aired on television, however, Mexico and the United Kingdom are also examples of this practice.

It is possible to find, in the literature, various works that seek an explanation from a strategic perspective. Some focus on the Supreme Court of the United States (FEREJOHN and WEINGAST, 1992), others on comparative politics (EPSTEIN, KNIGHT and SHVETSOVA, 2001; HELMKE, 2002), others on the Federal Constitutional Court in Germany (VANBERG, 2005), and yet others on the Supreme Court of Mexico (STATON, 2010).

Some suggest that constitutional courts are responsive to Legislative and Executive preferences, to public opinion (MISHLER and SHEEHAN, 1993; MCGUIRE and STIMSON, 2004; CARRUBBA and ZORN, 2008) and, finally, that judges consider both public preferences and the government (EPSTEIN, KNIGHT and SHVETSOVA, 2001; RAMSEYER and RASMUSEN, 2003; HELMKE, 2002; VANBERG, 2005).

In the disciplinary field of judicial institutions (judicial politics), the intriguing relationship between judicial institutions and public opinion remains an open topic. Even though Political Science offers theories for explaining this association which focus on the search for legitimacy, esteem, or institutional survival on the part of constitutional courts, empirical studies that focus on strategic models of judicial behavior are still scarce, mainly in Brazil.

Considering the relationship between the Supreme Court and the press as well as the institution’s communication practices with the public, the study proposed by Arguelhes, Oliveira, and Ribeiro (2012) is considered an exception. However, it consists of a different perspective, one related to the construction of an activist image that ocurred in the media discourse during the period of democratic transition, a phenomenon observed through the analysis of newspaper content, which are, in fact, the object of the study.

Finally, this article analyzes the behavior of the Federal Supreme Court, throughout the time under analysis, based on its decision on abstract constitutionality control, by placing as a variable the constitutional choice of promoting certain decisions to be communicated to the public and the press. In other words, is there, in fact, an institutional communication strategy?

1. Public Support Theory and Judicial Power

Political actors undoubtedly need public support to remain in power. The Judiciary - crucial for the operation of contemporary democracies and in permanent interaction with other representative institutions, due to its institutional shields as well as means of recruitment and dismissal - does not have an umbilical reliance on the public, as there is no relationship with the electorate. However, if there is, in fact, an increasing interaction with civil society, is it possible to identify communication strategies? If so, what are the incentives for such strategies to exist?

Evidence found in Political Science derives from the aimed mechanism for strengthening judicial decisions. Such a mechanism, which will be discussed in detail later, identifies the ideal conditions in which courts can be sorted and able to restrict the government, if necessary - once they are operationalized as political actors, whose function is legitimized through the representative process. Vanberg (2005), one of the authors who considers the theory that the greatest incentive would be the strengthening mechanism, summarizes both conditions under which the structure may be effective: (1) courts must have sufficient public support; (2) information about judicial decisions must be transparent.

The first condition is endogenous to Judiciary-Government interactions. As a result, judges hold only partial control over the limits on their power, in other words, only over transparency.³

In this sense, communication with the public would be necessary to develop legitimacy. Judicial actors tend to be more empowered in a context in which the public is informed and aware of their decisions.

On the other hand, as a side effect, public support for a more activist Judiciary may constitute a threat, as it may enable it to effectively influence political outcomes (CARRUBBA, 2009; STEPHENSON, 2004).

It is worth considering the term “activist” as it is used in this article. The most remote record of the term “judicial activism”, which can be distinguished as “unelected judges on the dispute of the last word with legislators”, is attributed to a 1947 article published in Fortune Magazine, written by the historian Arthur Schlesinger Jr. (ARGUELHES, OLIVEIRA and RIBEIRO 2012).⁴

Activism, as an expansion of the institutional space of judicial actors, can be expressed in different ways - creative interpretation of constitutional rules (mainly in principle); interaction between laws; expansion of procedural tools and the scope of decisions based on the Judiciary’s own initiative, or intervention in the elaboration or execution of public policies (with regard to social rights, such as health, work or even social security).

The historical antithesis of judicial activism - self-restraint - also known as self-restraint or deference - also faces obstacles related to conceptual fraying. Its nuances travel between deference and prudence as a strategy of institutional self-preservation. Therefore, the judicial institutions would retract their area of action before others.

Control of the docket limiting litigant access to constitutional jurisdiction or creation of procedural barriers (minimum requirements for case admissibility) are found among the tools that drive self-restraint. However, there are also interpretative tools, in the sense of promoting greater adherence to the texts (status quo), binding to precedents and lesser expansion of theorization of constitutional rights.

Both activism and self-restraint operate as “second-order preferences”: they guide the way in which the judge, before a policy whose constitutionality is under discussion, will take into consideration their first-order preferences regarding that policy” (ARGUELHES, 2015, p. 233).

Finally, among all the previously mentioned tools that lead to the expansion or restriction of judicial power, there is a dividing line that separates those considered endogenous (on the initiative of the judicial institutions themselves) and those deemed exogenous, such as institutional design, democratization, social demands and the strategic behavior of political actors. Although they can be analytically divided, they are able to influence one other:

The initial allocation of competences to each institution, according to the constitutional text, consists of the rule of the game and of that which is at stake in interactions among these powers. In day-to-day politics and deployment of the constitution, the various powers (mainly, but not exclusively, the Judiciary) can interpret the current Constitution in such a way as to expand and retract these competences or even to invest it with entirely new ones. (ARGUELHES, 2015, p. 216, author’s emphasis)

Such concepts are operationalized based on how the challenge to a constitutionally questioned policy/piece of legislation is answered, focusing the object of study on the judicial behavior observable from the results of the institutional decisions.

By revealing the limitations of such generalization, it is possible to consider a court as activist if it positively responds to judicial review actions. On the other hand, one that uses procedural tools to abstain from responding or respond negatively to demands of this nature, can be considered self-contained.

Faced with such a theoretical framework, the public can assist judicial actors as a form of support for more freedom of action in case of confrontation with incumbents. In other words, according to the theory of the strengthening mechanism, this public may represent a high political cost, insofar as the possibility of punishing those responsible for elections, considering potential conflicts between the judiciary and the government (CARRUBBA, 2009; VANBERG, 2005; STATON, 2010; STEPHENSON, 2004).

Therefore, the competence that courts detain to communicate with society, or even to institutionalize a process of transparency vis-à-vis their (sometimes selective) decisions, is mandatory for maintaining their reputation with positive impact on the legitimacy of its performance.

2. Institutional transparency and reputation

Transparency can be a very broad or trivial concept. Once inserted in theories of public support, it can be a variable that measures either the awareness of individuals of an “attempt toward non-compliance or evasion of a judicial decision” or even the mere possibility of public awareness. This section discusses those influences particularly central.

With regard to transparency, the literature highlights the following variables:

1. visibility achieved through press coverage;

2. institutional selection for disclosure;

3. case content (range of society’s interests or political institutions involved);

4. complexity of the case.

In the definition employed for this study, “transparency” means visibility or exposure. In the Brazilian case, there are few empirical studies on the behavior of the Federal Supreme Court focused on interactions with the public or the press (as an intermediary in this relationship) in order to illustrate how to operationalize such a concept as a variable. However, with due regard for some local idiosyncrasies, a study published by Staton (2010) was used as the main reference for the case of the Mexican constitutional court, which shares several configurations with the Federal Supreme Court,⁵ in spite of the contrasting number of annual cases it receives.

In Brazil, all judicial decisions are transparent and accessible to the external public (unless they are under legal secrecy). Due to this, the term “transparency” may be confusing when transferred to the study of judicial institutions.

However, from 1990 to 2016, a total of 1,558,019 cases was distributed in the Federal Supreme Court, which comes to an average of 59,923 per year,⁶ ten times greater than that of Germany and Mexico, for example.⁷ Faced with a Federal Supreme Court under an extensive workload and unimaginable numbers as compared to most Constitutional Courts in the world, how would it be possible to attract the attention of the public?

A small portion of the population influences public opinion - newspaper editors, journalists, writers, politicians, experts and interest groups. Several studies attribute to these specific groups considerable effects on the development of knowledge and public opinion on particular topics, done through filters and other tools that channel the discourse of such elites.

Some specific groups, such as the opposition under a given Administration, undeniably have a greater interest in transparency, as it represents a greater awareness of the topics debated in the judicial sphere, the final position of the highest court and possible maneuvers to circumvent a decision.

However, the ability of the oppositions to increase transparency is conditioned by its institutional environment and by the characteristics of the issue at stake. Legislative processes that are open to the public, that allow opposition groups to initiate public hearings, or to shape the agenda of committee hearings or floor debate, enhance the ability of the opposition committee hearings or floor debate, enhance the ability of the opposition to call attention to legislative evasion. (VANBERG, 2005, p. 46).

Some of these groups can be intuitively identified by observing which party is the author of the lawsuits. A clear example is the opposing parties, identified by Taylor (2008) as the ones that make the most use of direct actions of constitutionality, as they use them to extend previously lost or jeopardized legal battles to the legislative.⁸

Transparency also depends on factors other than the institutional environment or actions of external actors interested in mobilizing the community. Characteristics of the case itself can be decisive, hampering the ability to monitor the significance of the decision. The technical language usually employed by judges in rulings and votes considerably hinders the understanding of the broader public. Therefore, when a theme is more complex, such understanding and transparency are significantly limited.

The increase of complexity may be a tool intentionally used to reduce transparency. Several methods can be applied, such as accurately wording a position on a vote. The more precise it is, with clear foundations in constitutional principles and direct implications for a given policy, the easier it is to monitor its outcomes.

For the present study, complex thematic issues that simultaneously affect several policy areas and require more complex technical responses were considered. The classification was intentionally assigned by the authors based on the hypothesis raised by the literature (VANBERG, 2005; STATON, 2010), which suggests that more complex issues hamper the public’s ability to convincingly establish whether a decision was made less transparent to them.

Judicial institutions can also increase the probability of exposing certain decisions. Considering such a hypothesis, Staton (2010) explored the strategic use of the court’s press office to increase public awareness of decisions that may be challenged by legislative authorities.

3. Selective disclosure of decisions: ex post ⁹ international publication

This article aims to capture how communication between the Federal Supreme Court and the public arises. Naturally, estimating whether there is greater decision visibility by institutional initiative would be a herculean task, considering the difficulties in assuming certain conclusions related to associations between the outcomes of rulings and news coverage by the press, as suggested by the design proposed by Staton (2010).

It is understandable, however, that some evidence can be found on the institutional page of the Federal Supreme Court, which demonstrates some particularities. There is a section responsible for compiling some decisions and disclosing them as “News” by using journalistic or even simply less technical language, which would be the institution’s equivalent of a press release. Finally, there is also the selection of some rulings that become “informative”,¹⁰ which not only hold greater prominence, but also influence and shape the position of the entire legal class.

It is evident that the position of the press office is commissioned, and the person in charge is periodically appointed by the acting chief justice, who is responsible for communicating the contents of rulings and the docket for a given period (month or week) to the external public. The effort to translate legal technical language into more accessible and understandable terms for the lay public is quite noticeable. Recalling the aforementioned description of the theory of institutional support (EASTON, 1975), it is argued that, detaining more knowledge about a given institution is a mandatory condition to obtaining more validation.

Considering the methodology and the data, the results of previous Direct Actions of Unconstitutionality (ADI) rulings were used as the unit of analysis (proxy for judicial behavior), as they are the main procedural instrument of constitutionality control.

The “well-founded” result is considered by the literature as counter majoritarian, which means that the institution used its power of veto to annul a certain policy (legislation), and is the most widely used approach to estimate its behavior as a political actor - also known as “self-contained” versus “activist”.

Not only were considered as unfounded the ADIs not admitted on the merits, but also those that “await judgment” and those that faced “object loss”, considering that they result in a failure for the petitioner (the political actor who initiated the process) to obtain the desired veto, thus maintaining the status quo of the disputed legislation.

The first selection was carried out between 2000 to 2016. Over that period of time, 2,705 ADIs were judged. From this range, 358 cases were randomly selected, with a confidence level of 95%.¹¹

With a sample of 358 ADIs, the variables of interest were coded in the database which, in addition to containing the results of decisions, provides a list of keywords corresponding to the subject addressed in the process of judicial review, intentionally assigned by the authors. It is possible to evidence information about the petitioner of the ADI, contested legislation, date of the final ruling and the occurrence of articles published on the institutional page.

By observing the following points, it was decided to randomly select the decisions of the ADIs within the following range: (1) the result of the ADI as valid or not - dummy variable; (2) if the object of the ADI consisted of the Federal or State scope and location; (3) whether the lawsuit deals with a complex issue or not - with complex issues related to budgetary, tax, financial, social security and procedural law; non-complex ones dealing with civil, criminal, electoral and fundamental rights.¹²

On the news page, all articles referring to each ADI located by word search were individually analyzed by using their identification number in order to estimate the ex post institutional disclosure of judgments. The following observations were considered: (1) if the decision was the subject of any specific News published on the institutional page, which would be coded as a dummy variable (excluding articles that display the weekly or monthly schedule or that deal with several cases en bloc);¹³ (2) if there actually is publication, whether it was before or after the final decision.

As for the news, the institutional page publishes articles about decisions or ones still under deliberation, defined by the institution itself as a source responsible for publishing summaries of the factual and procedural circumstances and orally presented arguments in trial sessions. Since 1995, the court has been publishing bulletins based on notes taken during court sessions of the Panels and the Plenary, so that they contain unofficial summaries of decisions delivered by the court.

According to the press office of the Federal Supreme Court,¹⁴ the principal communication vehicles in Brazil employ specialized journalists (sectorists) in their coverage of the Federal Supreme Court. These journalists daily attend to gathering information in order to support their stories. In the year in which the interview with the press office was carried out, there were around 40 journalists from national and international vehicles licensed to conduct journalistic coverage.

Although all¹⁵ of the cases computed by the Federal Supreme Court can be accessed on its page (https://portal.stf.jus.br), via the “procedural follow-up” field, it is necessary to simply inform the identification number or the name of the litigants, thus, all plenary sessions of the Federal Supreme Court are fully transmitted by TV Justiça and are available on the internet. Even so, the news page represents a selection bias of the institutional urge to ward disclosure to the external public or not.

By referring to the discussion of institutional survival based on the search for specific support, it is argued that there would be no reason for the institution to concern itself with the probability of ex post coverage being low. Conversely, it already detains its own coverage from the disclosure, regardless of the results. There is, however, the possibility (and interest) of amplifying or undermining the disclosure of certain decisions.

Not all decisions of constitutionality control are broadcasted, as they do not go through plenary sessions. This was actually one of the changes that arose in the period after the year 2000, with the evident motivation of alleviating the Court’s schedule, enabling monocratic decisions in precautionary measures. All court case content is provided in its entirety to the external public (except when judicial secrecy is determined). However, the audience’s impossibility of comprehension, given that they are not familiar with the technical/legal language, is undeniable.

Though it is possible to evidence several theories that support motivations for the development of an institutional image, the present analysis still does not provide information regarding which image is intended to be transmitted. Based on the data and theory raised, the debate of the research analyzes, within behavioral discussions relevant to studies of political institutions, whether it is possible to identify an intentional choice to disclose specific decisions.

There are several arguments stating that the very object of political dispute would be to establish itself as a strong, independent and counter majoritarian court. This is exactly the role presented not only to the public, but also to other institutions - as an essential part of democratic ground rules.

It is essential, for Political Science, to observe a pattern that indicates rational choices (of institutional interest) by this institution, highlighting certain decisions to the detriment of others. By considering the triviality of publishing only the decisions that represent a “change in the status quo”, it is possible to argue that there are, within the sample, decisions of greater and lesser scope that refute the discussion of publication for “public utility”. Secondly, decisions of such (constitutionality control) nature embody a change in the applicability of a given policy, regardless of the counter majoritarian or self-contained result.

3.1. Exploratory data analysis

In this part, analysis of descriptive statistics for data display and exploration will be provided. As previously mentioned, the dependent variable “Publication” corresponds to the existence of at least one article on the institutional news page, communicating a decision to the external public, only considering the publications made after the decision.

“Valid Adin”, an independent variable, refers to the result of the ADI decision, which may be well-founded or unfounded. The other variables are: “Federal”, which indicates whether the object of the ADI represents Federal or State legislation; “Previous Publication”, which identifies whether there was any occurrence of an article on the institutional page before the final decision; “Administrative Law”, “Procedural Law”, “Social Security Law”, and “Tax Law”, which correspond to the complexity of the ADI themes. According to the literature, certain topics would be more likely to be disclosed to the general public, regardless of the result (VANBERG, 2005; STATON, 2010). Four cases were excluded from the analysis: ADI numbers 296, 1484, 2621 and 2778,¹⁶ respectively.

Graph 1 displays the distribution of the dependent variable. As can be observed, approximately 44% of the results of ADI judgments are the subject of articles published on the institutional page of the Federal Supreme Court, representing 157 of the 355 ADIs in the sample. Therefore, although all cases can be accessed in the procedural movement, it is mandatory that they are not published as news, evidencing a selectivity to publish certain cases.

Graph 1 -
Distribution of the dependent variable

Even considering that transparency is complete, as all cases are available for access by the external public, and that the plenary sessions are broadcasted, a further question still remains even after analyzing the data: why does the Federal Supreme Court decide to only provide accessible language for certain cases?

The theoretical model presented by Staton (2010) suggests that constitutional courts would be more interested in encouraging veto decisions to be widely publicized in the press, justified by the author as a potential risk of non-compliance.

The decision promotion model argues that the court applies such cases to advance the absence of reports regarding certain decisions as well as to ensure public monitoring, whether non-compliance is a potential problem that can be partly resolved by transparency or not. Empirically, if the self-promotion model is accurate, a positive interaction must be estimated between the variables “Supported ADI” and “Publication”.

Graph 2 displays the distribution of the variable of interest, the dichotomous variable that measures whether the ADI was approved or not by the Federal Supreme Court. It is observed that, in the sample, only 32% are considered valid, which represents 114 observations out of a total of 355.

Such a proportion among those that are well-founded and unfounded (in the latter group, combined to “without judgment on the merits” and “unfounded”) corroborates what literature affirms: most Federal Supreme Court decisions on actions of unconstitutionality challenging laws and normative acts reject the request, once made. According to Pogrebinschi (2011), 86.68% of the decisions maintain the status quo, thus confirming the constitutionality of the questioned rules and endorsing the majority intention expressed in them.

By observing Graph 2, it would be intuitively possible to conclude that the low number of lawsuits published as news after the final decision holds some relation to the low number of approved (well-founded) ADIs, however, this was not the result discovered, as discussed below:

Graph 2 -
Distribution of the Interest Variable

Graph 3 displays the distribution of the Federal dummy. It is possible to observe that a minority of ADIs deal with topics of such level, representing 45% of the total (161 of 355 ADIs).

The sample reflects the guidance of the Supreme Court, based on the Supreme in Numbers report (FALCÃO et al., 2013), according to which the Federal Supreme Court docket is affected by three different types of thematic areas: national, local and nationalized local:

The court, on the other hand, combined with what can be outlined as the “Court of the Federation”, is called upon to resolve issues of national interest. It is possible to observe, in this category, topics such as “public servant”, that arise among the main cases driven by the vast majority of states to the Federal Supreme Court. On the other hand, the Supreme Court often needs to rule on issues that seem to be relevant only from a local point of view. They consist of matters that are not quite representative for the federation as a whole, even on a small scale, the agenda of the Federal Supreme Court, as they may benefit from a certain expressiveness in a few states […] and, finally, issues tha can be characterized as nationalized local. These are issues that have a decisive impact on the agenda of the Supreme Court, even if they are not considerably significant throughout the federation. They constitute typical subjects of federal unit that are overrepresented in court - of which the Tax Law, a fundamental theme in São Paulo, is the most representative case (FALCÃO et al., 2013, p. 45-46).

Based on the classification used by Vanberg (2005) to measure the scope of the topic to be judged, a binary control variable was established by considering whether the policy is Federal or State in scope. According to the pattern of promoting judicial review proposed by Staton (2010), the visibility of constitutional cases would be endogenous to the nature of the conflicts themselves.

However, a greater probability of a press release based on the scope of the policy would be expected. Such a result would be consistent with the decision itself to carry out the press release. Therefore, the Federal Supreme Court would be more likely to disclose cases as the significance of the issue to be judged increased. In addition, more ADIs were identified as dealing with matters of State level, most of which did not have their results reported.

Graph 3 -
Distribution of the “Federal” variable

Graph 4 indicates the distribution of the variable “Previous Publication”, which corresponds to the existence of previous announcements on the news page, in other words, those made before a final decision. It is possible to observe that most of the ADIs were published on the Federal Supreme Court website prior to their judgment, representing 63% of the total (223 of 355 ADIs).

Graph 4 -
Distribution of the “Previous Publication” variable

However, a paradox emerges at this point: although most of the ADIs had been disclosed to the external public (focusing on the press) at a previous moment, the fraction drops considerably in terms of result coverage once a final decision is reached. In other words, the selectivity of reported decisions is related to their results, considering that previous publications (regardless of the decision reached) do not determine subsequent communication.

Based on the literature, there is consistent evidence to permit an analysis of court decisions to issue press releases responsible for announcing the results of judgments. It is presumed that the court would do so when the ADI is well-founded, especially if the case is difficult to cover and if there is no institutional initiative in this regard.

Graph 5 displays the distribution of the frequency of dummy variables for the ADI areas of administrative, procedural, social security and tax law, included in the logistic regression model setting. The casual mechanism responsible for justifying an expected negative impact of these variables on the probability of publication would be related to the complexity of such issues.

Graph 5 -
Frequency distribution of dummy variables on types of ADI

According to Falcão et al. (2013), the main topics argued before the Supreme Court are, respectively: administrative law; tax law; labor law; civil procedure; social security law; civil rights; criminal proceedings; consumer law and criminal law. The authors also add that there is a greater predominance of themes involving the state: “administrative law and tax law are, in general, disputes between individuals and state bodies. Labor law involves, above all, disputes over the FGTS¹⁷ against Caixa Econômica Federal” (FALCÃO et al., 2013, p. 48).

The thematic areas were divided into two groups: (1) administrative, tax, procedural and social security; (2) civil, electoral, criminal, environmental and labor. The first group coded as “more complex” and the second, as “less complex”, for which a greater probability of publication would be expected.

The visibility of the cases, or what Vanberg (2005) titles “transparency”, does not only depend on the institutional situation and actions of external actors, but also on inherent characteristics related to the issue. Thus, the complexity regarding to the thematic area of a given contested policy is one of such factors, alongside the scope (Federal/State) outlined above:

It is more difficult to establish whether a judicial decision has been faithfully implemented in policy areas that are technically more complex and demanding. For examples, a legislative reaction to a decision declaring the death penalty unconstitutional is transparent because it is fairly straightforward to determine (at least in a liberal democracy) whether executions still take place. By contrast, a technical decision involving several tax code provisions is much harder to monitor. Does the tax code as revised satisfy court’s demands? in general, policy areas that involve regulation of substantive outcomes (as opposed to procedural rules) and cases that involve multiple policy areas rather than a single issue tend to more complex. (VANBERG, 2005, p. 47)

Thus, the greater the complexity, the lower the visibility tends to be. When observing judicial decisions, judges normally present an interest in increasing the transparency of their decisions, unlike legislators, as it is mandatory for them to be grounded, considering that such characteristics would lower the risk of non-compliance. There are several ways to increase transparency, whether through the transparency of arguments, achieved by casting votes, precisely elaborating the position and principle that support it, or as political implications.

By observing the decisions of the Federal Supreme Court after the period during which the plenary sessions were broadcasted, Lopes (2018) evidenced a considerable increase in the extent of the votes of each minister, indicating the impact of transparency on the judicial behavior. Some aspects allow courts to issue highly focused decisions, while others require more diffuse ones, which are more challenging to be delivered in precise language. The fact, however, is that there has been an institutional interest in taking advantage of the space to expose each individual positioning.

The other way of increasing transparency, which is relevant to the question of this research project, consists of the communication of these decisions in other spaces, such as what was found by Staton (2010) by observing the communication strategies the Court of Mexico employed with the press, drawing attention to decisions that could be contrary to the interest of legislative majorities. There are still few studies regarding the effect of the Federal Supreme Court’s decisions and association with its salience/visibility, which would be a perspective in the opposite direction to that proposed by Lopes (2018).

From the perspective of the thematic in the decision-making process, some studies differentiate cases according to the ability of the court to reliably introduce implementation (HALL, 2014).

This theory differentiates enforcement power of “vertical” cases - those in which decisions could be directly enforced by lower courts -, such as criminal prosecution, civil liability, judicial administration; from “lateral” cases - those in which the court relies on “non-judicial cases”, in which implementation depends on a set of other institutions -, such as budgetary and tax law.

Relating such conceptions to the Federal Supreme Court, with due regard for the contextual characteristics that differentiate the US from the Brazilian judicial system, it would be possible to expect that an external influence would be more decisive in the decision-making of the court on some issues.

However, considering that the decision argued by Hall (2014) is effective for the case under analysis, it would be intuitive that thematics considered “lateral” would have a greater chance of institutional publication, based on the logic of disclosing cases in which there was greater institutional difficulty to be enforced. Based on the data, the area that faced greater impact was civil law, labeled as “vertical” cases and contrary to what was expected.

As evidenced in Table 1, the ADI that may have a negative impact on publication add up to 266 of the total 355 ADIs (75%). Most of these deal with administrative law. It is actually a remarkable fact, as this area of law is the main subject discussed in ADIs, as highlighted by the Supreme’s report in numbers, especially with regard to the topic of public servants, which includes queries about remuneration and other working conditions.

Table 1 -
ADIs by area of Law

EXPECTED EFFECT ON THE DUMMY VARIABLE (D.V.)	AREA	FREQUENCY	%
NEGATIVE	LAW	150	42.3
	TAX LAW	65	18.3
	PROCEDURAL LAW	30	8.5
	SOCIAL SECURITY LAW	21	5.9
POSITIVE	CIVIL LAW	49	13.8
	ELECTORAL RIGHTS	14	3.9
	CRIMINAL LAW	13	3.7
	ENVIRONMENTAL LAW	8	2.3
	LABOR LAW	5	1.4
	TOTAL	355	100

• Source: Authors’ own elaboration based on data from the Federal Supreme Court website.

Among the areas of law that may have a positive impact on publication, civil law stands out from the rest, with a total of 49 cases. Such issues, related to the labor law, were included in this thematic area by considering the low occurrence of specific cases on labor relations, and also because they correspond to individual rights and interactions among individuals.

In the obtained sample, most cases deal with issues related to the discussion of individual freedoms, such as the issue of secrecy of public documents (ADI n. 4047), exclusive spaces for women on rail transportation systems (ADI n. 4231), smoking of cigarettes in collective spaces (ADI n. 4239), same-sex marriage (ADI n. 4277) and biography authorization (ADI n. 4815).

It was decided to only include in the model the thematic areas with a negative impact, which account for most of the ADIs, in order to avoid multicollinearity problems.

3.2. Results of logistic regression model setting

Based on the binary nature of the dependent variable, the logistic regression model is adopted to describe the relationship between the variable that represents the publication of ADI judgment results on the Federal Supreme Court news page and the explanatory variables. Such a model is defined as follows:

${\gamma }_i=\alpha \mathit{}+{\beta }_{1}ADIN+{\beta }_{2}FED+{\beta }_{3}PANT+\mathit{}{\beta }_{4}ADM+{\beta }_{5}PRO+{\beta }_{6}PREV+{\beta }_{7}TRIB+{\epsilon }_i$

In the equation, γ_i corresponds to the dependent variable, which is binary categorical and represents the possible publication of the ADI result on the Federal Supreme Court website; α is the model constant; β₁ADIN represents the dichotomous variable that identifies whether the ADI was approved or not; β₂FED is a dummy that measures whether the ADI is a matter at the Federal level (value 1) or State level (value zero); β₃PANT is the binary level that indicates whether there has already been a publication on the Federal Supreme Court website about the ADI before its judgment; β₄ADM, β₅PRO, β₆PREV e β₇TRIB respectively represent dummies that indicate whether the ADI was a matter of administrative, procedural, social security or tax law, ε_i is the error term.

As observed in the analysis of the ROC curve in Graph 6, the model represents a high predictive performance. Compared to the diagonal line, which represents a random binary classification, the model has presented significant improvement. The area under the ROC (Receiver Operating Characteristic Curve), specifically, represents a predictive power of approximately 85%.

Graph 6 -
Setting Curve of the Logistic Regression Model

As observed in the contingency table below (Table 2), only the explanatory variables “Well-founded Adin” and “Previous Publication” are effectively correlated with the dependent variable, as both appeared exclusively with the significant chi-square test.

Table 2 -
Contingency of explanatory variables and the dependent variable (%)

		PUBLICATION
		0	1
WELL-FOUNDED ADIN*	0	74	26
	1	18	82
FEDERAL	0	53	47
	1	59	41
PREVIOUS PUBLICATION*	0	77	23
	1	43	57
ADMINISTRATIVE LAW	0	53	47
	1	59	41
PROCEDURAL LAW	0	56	44
	1	53	47
SOCIAL SECURITY LAW	0	55	45
	1	62	38
TAX LAW	0	56	44
	1	54	46

• Source: Authors’ own elaboration based on data from the Federal Supreme Court website.
• *The chi-square statistic is significant at 5%

Approximately 82% of the “Well-founded Adins” and 57% of those which already had news published before the trial have their final results published on the Federal Supreme Court website. Such a relationship can be fully understood in the Venn Diagrams presented in Graphs 7 and 8.

Graph 7 -
Venn Diagram of the variable “Well-founded Adin”

Graph 8 -
Venn Diagram of the variable “Previous Publication” and the dependent variable

The tests of Hosmer and Lemeshow were used in order to analyze the general setting of the model. According to Garson (2011), this test is considered stronger, compared to the chi-square test. A non-significant result (p > 0.05) suggests that the model estimated with the independent variables is better than the null model. As observed in Table 3, the estimated model presented a chi-square of 6.983, with a p-value of 0.539, suggesting an adequate setting.

Another commonly used measure of setting is the Omnibus coefficient test. Differently from the Hosmer and Lemeshow tests, a significant result (p < 0.05) suggests an adequate setting (GARSON, 2011). According to the data in Table 3, the model presented a statistically significant chi-square of 155.667 (p-value < 0.000), in other words, the null hypothesis of statistical independency between the independent variables and the dependent variable is rejected.

Table 3 -
General setting tests of the model

TEST	CHI-SQUARE	SIG.
HOSMER AND LEMESHOW	6.983	0.539
OMNIBUS COEFFICIENT TEST	155.667	0.000

• Source: Authors’ own elaboration based on data from the Federal Supreme Court website.

It is initially evident that when all the seven variables are included in the analysis (model 2), approval of ADIs remains with the positive significant coefficient of model 1, corroborating the main hypothesis that a greater probability of ADI publication results on the news page happens when it is properly judged.

Table 4 presents the results of the decreases. In parentheses, the coefficients are displayed in relation to the odds, as the interpretation through the logarithm is not intuitive.¹⁸

Table 4 -
Determinants for publishing a result of an ADI judgment on the Federal Supreme Court website

	(1)	(2)
WELL-FOUNDED ADIN	2.586***	3.189***
	(13.279)	(24.263)
FEDERAL		0.195
		(1.216)
PREVIOUS PUBLICATION		1.893***
		(6.642)
ADMINISTRATIVE LAW		-0.731**
		(0.481)
PROCEDURAL LAW		0.041
		(1.042)
SOCIAL SECURITY LAW		-0.457
		(0.633)
TAX LAW		-0.006
		(0.994)
CONSTANT	-1.039***	-2.209***
	(0.354)	(0.110)
PSEUDO R² NAGELKERKE	0.342	0.475
% PREDICTED	75%	85%
N	355	355

• Source: Authors’ own elaboration based on data from the Federal Supreme Court website.
• Reported regression coefficients (β); Exp (β) (odds ratio) in parentheses. Significance: **p < 0.05 ***p < 0.01.

Secondly, it is observed that the variable of interest also detains a greater explanatory power over the dependent variable, compared to the other variables included in the analysis. Based on the values in parentheses regarding the odds ratio, it is possible to understand that, by keeping the other variables constant when the ADI is well-founded, the chance of publishing the judgment result is 24 times greater. Such a result can be interpreted in a more intuitive way in Graph 9, which demonstrates the average marginal effect between the results of the judgments and the selection of those that will be reported on the institutional news page.

Graph 9 -
Average marginal effect of ADI approval on the chance of publishing the judgment result on the Federal Supreme Court website

Part of the literature suggests that being transparent when externalizing judicial decisions promotes compliance, which would imply the intention to disclose the decisions that annul the status quo in a more explicit way, described in simpler and less technical language.

Public support, crucial to the maintenance of the judiciary, would have as a requirement the awareness of these decisions, in other words, a greater visibility (also known by the literature as term transparency). Specific decision support refers to the satisfaction of certain policy outcomes, on the part of an institution, regarding specific judicial decisions.

Although there are some obstacles in empirically distinguishing specific support (for certain policies) from diffuse support (for the institution as a whole), several studies claim that both types of support are equivalent. Considerable evidence suggests that substantive evaluations of specific court decisions affect the level of public support for a court, specifically for opinion leaders (CALDEIRA and GIBSON, 1992). Possibly, the press office of the Federal Supreme Court believes in such a statement, as it is the institution responsible for selecting final decisions to be reported as press releases.

In model 2, the Nagelkerke test showed a result of 0.475 with all the independent variables. When interpreted according to the R² logic of linear regression, it is understandable that the explanatory variables collaboratively explain approximately 48% of the variation in the publication of ADI judgment results, which is a relevant quantity for a research study in the field of Social Sciences.

The equation sought to control three potential influences for the choice of cases that would be reported as press release: (1) the scope of the questioned legislation, dividing it into Federal and local levels, assuming that the first group would have a positive impact, as the relevance of the policy increases; (2) existence of the ex ante, in other words, if the case had been published on the page before the result of the decision, which would indicate that certain processes are considered more relevant than others, regardless of the merits or rejection of the ADI; (3) thematic area according to complexity, as certain subjects would have a greater chance of publication ex post regardless of the result.

It is vital to point out that only the origin dummy of the ADI (model 1) is able to explain 34% of the variation in the dependent variable, which demonstrates the centrality of the ADIs origin to understand the publications of the ADI case results on the institutional page of the Federal Supreme Court. Such a model holds a predictive power of 75%, therefore, it is only 10% lower than the model that contains all the explanatory variables.

Another variable that proved to be relevant was the dummy responsible for measuring whether there had already been a publication on the ADI before its judgment or not. The results suggest that, when an ADI has been mentioned on the Federal Supreme Court website before the judgment, the chance of publishing the judgment result is six times greater. This result can be interpreted in a more intuitive way in Graph 10, as it represents the marginal predictions, in which the independent variable varies while the others are fixed in their respective means.

Graph 10 -
Average marginal effect of the publication presence on the ADI prior to its judgement over the chance of publishing the judgment result on the Federal Supreme Court website

Understanding which themes are more frequently reported before the respective results is an attempt to find issues intrinsically considered “important” by the institution. The most relevant questions for the court, regarding the thematic classification or level of scope - the level of government to be confronted - and the value it represents as precedent-building are equally relevant issues.

It is possible that the publication does not relate to the public support mechanism, so that the court could be prone to release ADIs that veto policies, in the sense that many decisions reflect potential changes in legislation that are worthy of announcement.

The third and last variable that presented a significant coefficient was the dummy that identifies whether the ADI is an administrative law. The results suggest that, when an ADI belongs to this area, the chance that the judgment result will be published is 52% lower.¹⁹ This result can be interpreted in a more intuitive way in Graph 11, which displays the marginal effects of this variable, maintaining the other constants.

Graph 11 -
Average marginal effect of the ADI that deals with Administrative Law on the chance of publishing the judgment result on the Federal Supreme Court website

When observing the control related to the intrinsic characteristics of each case, it is possible to state that the administrative law would belong to a more complex area, as it involves several sparse pieces of legislation with a lower impact on publication or, in the classification proposed by Hall (2014), “lateral” cases, as they presuppose the joint action of other institutions.

However, changes that involve salary readjustments for members of the Judiciary or a reduction in the attributions of the National Council of Justice would be considered examples in which a reduction in transparency could be preferable. It could be explained either by reasons related to institutional reputation or the lower need for coordination with other spheres of power, thus, these could be plausible explanations for the variable to present a negative coefficient, the most recurrent theme in ADIs.

Although the literature considers that Federal-level news may have a positive impact on the publication of ADIs, according to the previous discussion and by using this model, it was not possible to rule out the null hypothesis that federal-level news does not have any influence on the variation of the independent level.

Conclusion

In the tested model, transparency (visibility) is endogenous, limited to institutional publications (press release) and it can also be controlled by its agents, inspired by the model of strategic disclosure (STATON, 2010).

Observing which decisions are reported to the public - that the probability of intentional disclosure by the institution will be greater when the decision is counter majoritarian (full or partial origin of the ADI); based on the values, in parentheses, referred to the odds ratio by keeping the other variables constant, it is possible to understand that, when the ADI is valid, the chance of publishing the judgment result is 24 times greater. Finally, it is possible to conclude that the model presents significant evidence that there is a positive relationship between the outcome of decisions and the ex post publication on the initiative of the Federal Supreme Court.

At first, such a relationship may indicate a similar strategy to that identified by Staton (2010) in the Mexican constitutional court. In relation to the Federal Supreme Court, the probability of publication of an ADI by the institution is greater when the result is valid or counter majoritarian, even when other control variables are evident. The results suggest that, when an ADI has already been mentioned on the Federal Supreme Court website before its judgment, the chance of publishing the judgment result is six times greater.

Either through the theory of public strengthening of decisions or through the development of an institutional reputation, it is possible to evidence a constant intention, throughout the time period examined in this article, aimed to emphasize valid ADIs, either fully or partially.

Certain variables were considered, such as the complexity of the theme²⁰ and scope (Federal or State) as well as a possible previous preference for the theme (ex ante publication). However, none of these variables demonstrated considerable significance.

Based on the results, it is possible to conclude that, although most of the ADIs are judged as unfounded when observing the judgments appointed for publication on the institutional page of the Federal Supreme Court and their respective results, it was possible to evidence a positive relationship, with high statistical significance, between the publication of articles on the ADIs and the well-founded result.

Even if they are considered control variables with a considerable explanatory power, the findings indicate that the Federal Supreme Court prefers to display counter majoritarian results to the public. This finding can be explained by the intention to enhance the image that such decisions predominate in the court, contrary to the actual data.

Although it is argued that the choice is attributed to the press office of the Federal Supreme Court, the commissioned position is appointed by its presidency, which shifts every two years. As the sample refers to a wide range (2000 to 2016), it is possible to conclude that it consists of an institutional policy which does not change according to the agent which occupies the role of president.

The results consist of evidence that the theory of strategic disclosure (STATON, 2010) is applied to the Federal Supreme Court at some point, considering that institutional news represents the means of disclosure to the press and the public.

It would also be possible to infer that the counter majoritarian result or the “Well-founded Adin” represents an objective of the institution, given that the behavioral pattern was found to be solid over the years and across the configurations of ministers analyzed, as well as at least eight different press office administrations. Considering such probability, it would be interesting to evaluate the variation over time for future studies.

However, by mentioning the link between diffuse and specific support, it is possible to evidence implications for judicial behavior from the perspective of a decision’s outcome. Considering that public support is a vital resource, the quest to maintain such support for the court could possibly influence judicial deliberations, as court actors recognize that their decisions are related to future support.

This fact could be considered a motivation, along with the urge to maintain institutional support or reputation to the sensitivity of public opinion. Such support would be a valuable resource that could be deployed rapidly, as many unpopular decisions convince citizens that the court exerts an undesirable influence on policy.

The research findings motivate the extension of the debate about the casual mechanisms behind the phenomenon. Based on the theory of institutional support, it is possible to conclude that, publics which are more attentive to the courts are able to support their strengthening as a political institution, however, it is not possible to assert that the Federal Supreme Court strongly relies on it, nor that it motivates its institutional behavior.

Detaining support is a dynamic process - data suggest that courts develop specific support for its “salience”, which allows its exposure to the wider public. This satisfaction evolves into constitutional legitimacy, and the level of connection between specific and diffuse support is accumulated over time. Several areas of policy-making appeal to different constituencies, thus, legitimacy would be achievable through the mechanism that Gibson, Caldeira and Baird (1998) explain as the successive satisfaction of “minority coalitions”.²¹

Therefore, it is possible that the selection bias of what is published is a determining factor in the probability that the institution will evidence certain decisions. This conclusion seems consistent with the theory discussed, as well as with the sample discovered.

Finally, not only do we hope to have covered a gap but also slightly contributed to advancing the agenda of studies on judicial behavior, mainly in Brazil, with an emphasis on the relationship with the public as a variable of interest.

ACKNOWLEDGMENTS

The authors would like to thank the anonymous reviewers of Revista Direito GV for their attentive observations.

REFERÊNCIAS

• ARGUELHES, Diego W. Poder não é querer: preferências restritivas e redesenho institucional no Supremo Tribunal Federal pós-democratização. In: SARMENTO, Daniel (ed.). Jurisdição constitucional e política Rio de Janeiro: Forense, 2015.
• ARGUELHES, Diego W.; OLIVEIRA, Fabiana Luci de; RIBEIRO, Leandro M. Ativismo judicial e seus usos na mídia brasileira. Revista Direito, Estado e Sociedade, [s. l.], n. 40, p. 34-64, Jan/Jun, 2012.
• BARROSO, Luís Roberto. Contramajoritário, representativo e iluminista: os papéis dos tribunais constitucionais nas democracias contemporâneas. Revista Direito e Práxis, [s. l.], v. 9, n. 4, p. 2171-2228, dez. 2018. Available at: Available at: https://www.e-publicacoes.uerj.br/index.php/revistaceaju/article/view/30806 Accessed on: Mar 20, 2023.
  » https://www.e-publicacoes.uerj.br/index.php/revistaceaju/article/view/30806
• BAUM, Lawrence. Judges and Their Audiences: A perspective on Judicial Behavior. New Jersey: Princeton University Press, 2006.
• CALDEIRA, Gregory A.; GIBSON, James L. The Legitimacy of the Court of Justice in the European Union: Models of Institutional Support. The American Political Science Review, [s. l.], v. 89, n. 2, p. 356-376, Jun, 1995.
• CALDEIRA, Gregory A.; GIBSON, James L. The Etiology of Public Support for the Supreme Court. American Journal of Political Science, [s. l.], v. 36, n. 3, p. 635-664, Aug, 1992.
• CARRUBBA, Clifford J. A Model of the Endogenous Development of Judicial Institutions in Federal and International Systems. The Journal of Politics, [s. l.], v. 71, n. 1, p. 55-69, Jan, 2009.
• CARRUBBA, Clifford J.; ZORN, Christopher J. Executive Discretion, Judicial Decision Making, and Separation of Powers in the United States. The Journal of Politics, [s. l.], v. 72, n. 3, Jul, 2008.
• CARVALHO, Ernani. Trajetória da revisão judicial no desenho constitucional brasileiro: tutela, autonomia e judicialização. Sociologias, [s. l.], v. 12, n. 23, p. 176-207, 2010.
• CARVALHO, Ernani. Judicialização da política no Brasil: controlo de constitucionalidade e racionalidade política. Análise Social, Lisbon, n. 191, p. 315-335, Apr, 2009.
• CARVALHO, Ernani. Revisão abstrata da legislação e a judicialização da política no Brasil 2005. Tese (Doutorado em Ciência Política) - Universidade de São Paulo, São Paulo, 2005.
• CARVALHO Ernani et al Judicialización de la política y grupos de presión en Brasil: intereses, estrategias y resultados. América Latina Hoy, [s. l.], v. 72, p. 59-88, 2016. Available at: Available at: https://revistas.usal.es/index.php/1130-2887/article/view/9426 Accessed on: Mar 14, 2023.
  » https://revistas.usal.es/index.php/1130-2887/article/view/9426
• DOWNS, Anthony. An Economic Theory of Democracy New York: Harper and Row, 1957.
• EASTON, David. A Re-Assessment of the Concept of Political Support. British Journal of Political Science, [s. l.], v. 5, n. 4, p. 435-457, 1975.
• EPSTEIN, Lee; KNIGHT, Jack; SHVETSOVA, Olga. The Role of Constitutional Courts in the Establishment and Maintenance of Democratic Systems of Government. Law and Society Review, [s. l.], v. 35, n. 1, p. 117-164, 2001.
• FALCÃO, Joaquim et al II Relatório do Supremo em Números: O Supremo e a Federação. Rio de Janeiro: Escola de Direito do Rio de Janeiro da Fundação Getulio Vargas, 2013. 184 p.
• FEREJOHN, John A.; WEINGAST, Barry R. A Positive Theory of Statutory Interpretation. International Review of Law and Economics, [s. l.], v. 12, n. 2, p. 263-279, 1992.
• GARSON, G. David. Statnotes: Topics in Multivariate Analysis. 2011. Available at: Available at: http://www.statisticalassociates.com Accessed on: Mar 21, 2023.
  » http://www.statisticalassociates.com
• GIBSON, James L.; CALDEIRA, Gregory A.; BAIRD, Vanessa A. On the Legitimacy of National High Courts. American Political Science Review, [s. l.], v. 92, n. 2, p. 343-58, 1998.
• HALL, Matthew E. K. The Semiconstrained Court: Public Opinion, the Separation of Powers, and the U.S. Supreme Court’s Fear of Nonimplementation. American Journal of Political Science Association, [s. l.], v. 58, n. 2, p. 352-366, 2014.
• HELMKE, Gretchen. The Logic of Strategic Defection: Court-Executive Relations in Argentina Under Dictatorship and Democracy. American Political Science Review, [s. l.], v. 96, n. 2, p. 291-303, 2002.
• LOPES, Felipe de M. Television and Judicial Behavior: Lessons from the Brazillian Supreme Court. Exchange Organizational Behavior Teaching Journal, [s. l.], v. 209, p. 41-71, 2018.
• MCGUIRE, Kevin T.; STIMSON, James A. The Least Dangerous Branch Revisited: New Evidence on Supreme Court Responsiveness to Public Preferences. The Journal of Politics, [s. l.], v. 66, n. 4, p. 1018-1035, 2004.
• MISHLER, William; SHEEHAN, Reginald S. The Supreme Court as a Countermajoritarian Institution? The Impact of Public Opinion on Supreme Court Decisions. The American Political Science Review, [s. l.], v. 87, n. 1, p. 87-101, 1993.
• POGREBINSCHI, Thamy. Judicialização ou representação? Política, direito e democracia no Brasil. Rio de Janeiro: Elsevier, 2011.
• POSNER, Richard A. Foreword: A Political Court. Harvard Law Review, [s. l.], v. 32, 2005.
• POWELL JR., G. Bingham. Elections as Instruments of Democracy: Majoritarian and Proportional Visions. New Haven: Yale University Press, 2000.
• RAMSEYER, J. Mark; RASMUSEN, Eric B. Measuring Judicial Independence: The Political Economy of Judging in Japan. Chicago: University of Chicago Press, 2003.
• SEGAL, Jeffrey A.; SPAETH, Harold J. The Supreme Court and the Attitudinal Model Revisited New York: Cambridge University Press, 2002.
• STATON, Jeffrey K. Judicial Power and Strategic Communication in Mexico New York: Cambridge University Press, 2010.
• STEPHENSON, Matthew C. Court of Public Opinion: Government Accountability and Judicial Independence. The Journal of Law, Economics & Organization, [s. l.], v. 20, n. 2, p. 379-399, 2004.
• TAYLOR, Matthew M. Judging Policy: Courts and Policy Reform in Democratic Brazil. Stanford: Stanford University Press, 2008.
• VANBERG, George. The Politics of Constitutional Review in Germany New York: Cambridge University Press, 2005.

• 1
  This phrase became popular worldwide through one of the greatest communicators on Brazilian television, José Abelardo de Medeiros, also known as Chacrinha.
• 2
  Baum (2006) presents an ironic metaphor of the “judge as Mr. Spock”, comparing him to the fictional character (not human) whose attitudes were devoid of selfishness and emotions, in the search for the common good: “Thus there is an element of Mr. Spock in the judges of the pure legal and attitudinal models, judges who act without emotion or self-interest in order to advance the general good. As the leading expert on Mr. Spock explained, ‘In the Vulcan culture, one simply does what is right’ (Nimoy 1977, 104)” (BAUM, 2006, p. 18).
• 3
  In this study, the concept can be understood as an effort to grant publication or visibility to the work carried out by the institution. The term “transparency” relates to the theory of public support, as outlined throughout this article. The hypothesis is that the initiative to disclosure represents a choice made by the institution, considering the possible purpose of developing legitimacy.
• 4
  In this widely quoted text, a profile of the ministers of the Supreme Court was drawn and they were further classified into groups such as “activists with emphasis on the defense of minority rights”, “activists with an emphasis on the right of freedom”, “defenders of self-restraint”, and “representatives of the balance of power”. With no intention of restricting itself to a discussion of such a constitutional model, the debates were enhanced in proportion to the global expansion of the judicialization of politics.
• 5
  The Mexican constitutional court has competence to exercise centralized judicial review, concrete, abstract and individual constitutional actions - writ of mandamus, support, habeas corpus, etc. The nature that stands out regarding the participation of these constitutional courts in the political process consists of the possibility of deciding disputes in the absence of a concrete dispute, which allows a policy to be immediately contested, after being approved by the Legislative or Executive and even before being completely effective. It also bears similarities with regard to the number of members, means of appointment and periods in which it underwent extensive constitutional reforms (early 2000s) by giving it more protagonism, as well as more transparency.
• 6
  Source https://transparencia.stf.jus.br/extensions/recebidos_baixados/recebidos_baixados.html. Accessed on Mar 20, 2023.
• 7
  Continuing the approach on Mexico, actions under the jurisdiction of its institutional court, SCJN, are outlined in the Action of Unconstitutionality as well as in the Amparo and Constitutional Controversy. As for abstract constitutionality control, similarly to Germany, the requirements for a process of this nature to be filed consist of 1/3 of the Chamber of Deputies or the Senate (against international laws); 1/3 of Local Assemblies (against State laws); Attorney General of the Republic (against any law); and party leaders (against Federal election laws). From 1995 to 2016, 41,700 cases were introduced according to data from the official web page, considering constitutionality control and appeals from lower courts, which represents an average of 2,000 cases received per year. Calculated by the authors, based on data available at: Suprema Corte de Justicia de la Nación. Judicial statistics portal: https://estadisticajudicial.scjn.gob.mx/. Accessed on 20 Mar, 2023.
• 8
  “The legitimate assets of art. 103 of the CF were unquestionably responsible for the emptying of diffuse control. The preference for the concentrated model is present from two points of view: 1) political point of view: the absence of political cost for those who propose it, the possibility of a proposal being vetoed by the Executive or by a majority of the Legislative, the greater visibility of a direct action of unconstitutionality and its media effect cause by the erga omnes scope of the trial” (CARVALHO, 2010, p. 196).
• 9
  The concept of selectivity as an instrument of political strategy in the judicial sphere was originally used in the Brazilian case by Carvalho (2005, 2009 and 2016).
• 10
  We appreciate the comments of the members of the Politics Working Group and Law and Judiciary during of XI meeting of the Brazilian Association of Political Science (ABCP) in Curitiba, Paraná, 2018.
• 11
  The randomly selected actions were: 12; 49; 62; 71; 94; 147; 161; 168; 196; 208; 214; 220; 239; 250; 255; 273; 279; 296; 312; 325; 329;340;341; 397; 414; 424; 450; 451; 459; 463; 494; 504; 506; 510; 602; 644; 668; 711; 770; 771; 775; 783; 857;863; 872; 886; 898; 906; 944; 945; 1072; 1074; 1163; 1165; 1200; 1218; 1221; 1272; 1273; 1280; 1281;1361;1376;1398; 1435; 1484; 1493; 1502; 1517; 1551; 1583; 1595; 1620; 1627; 1633; 1659; 1667; 1671; 1699; 1717; 1720; 1722; 1726; 1731; 1767; 1779; 1814; 1833; 1851; 1864; 1868; 1878; 1892; 1895; 1920; 1923; 1931; 1952; 1856; 1972; 1994; 2018; 2021; 2049; 2062; 2102; 2134; 2193; 2219; 2229; 2232; 2238; 2251; 2273; 2288; 2294; 2315; 2316; 2348; 2358; 2366; 2376; 2380; 2385; 2398; 2411; 2412; 2413; 2416; 2430; 2433; 2443; 2454; 2455; 2463; 2480; 2484; 2490; 2497; 2512; 2514; 2541; 2556; 2547; 2588; 2597; 2610; 2621; 2622; 2624; 2629; 2672; 2684; 2703; 2709; 2710; 2717; 2626; 2732; 2735; 2741; 2758; 2777; 2778; 2793; 2797; 2813; 2829; 2842; 2843; 2863; 2875; 2885; 2904; 2906; 2925; 2928; 2953; 2966; 2992; 3027; 3030; 3046; 3047; 3067; 3073; 3075; 3087; 3089; 3101; 3105; 3112; 3128; 3130; 3139; 3145; 3149; 3167; 3168; 3192; 3193; 3200; 3203; 3246; 3249; 3251; 3267; 3271; 3293; 3330; 3346; 3347; 3369; 3374; 3376; 3379; 3390; 3410; 3439; 3440; 3453; 3454; 3472; 3484; 3510; 3527; 3548; 3566; 3583; 3599; 3603; 3606; 3609; 3647; 3650; 3654; 3672; 3681; 3686; 3690; 3706; 3726; 3741; 3745; 3750; 3756; 3773; 3809; 3812; 3817; 3827; 3831; 3858; 3867; 3869; 3876; 3896; 3906; 3932; 3937; 3992; 4005; 4025; 4043; 4049; 4050; 4061; 4077; 4087; 4095; 4096; 4103; 4126; 4127; 4145; 4152; 4167; 4172; 4176; 4189; 4200; 4209; 4231; 4239; 4255; 4266; 4274; 4277; 4284; 4315; 4345; 4352; 4357; 4358; 4376; 4389; 4424; 4433; 4437; 4439; 4451; 4534; 4552; 4569; 4602; 4607; 4627; 4642; 4650; 4655; 4679; 4680; 4696; 4698; 4713; 4714; 4717; 4722; 4739; 4777; 4793; 4811; 4815; 4883; 4887; 4901; 4917; 4951; 4964; 4978; 4986; 4989; 4991; 5000; 5015; 5230; 5240; 5035; 5053; 5058; 5089; 5096; 5118; 5178; 5221; 5240; 5418; 5446; 5494; 5500; 5513; 5521; 5530; 5539; 5548; 5579; 5590; 5618.
• 12
  The complexity of issues and the generality of the contested policies (whether Federal or State in scope) will be considered as control variables, in an attempt to understand the level of knowledge/visibility of decisions in an endogenous way. Furthermore, the term complexity represents a factor that influences transparency over a decision. The less complex a policy area is, the easier it is for citizens to determine whether a decision was respected, as its implications tend to be more direct. For example, compliance with a decision that recognizes same-sex marriage is easier to monitor than compliance with a decision that deals with interrelationships among several provisions of social security law. Considering the set of decisions for which citizens can readily determine whether the decisions of constitutional courts was accurately implemented, legislators should be more aware of the potential for public response if they comply with the decision of the court.
• 13
  The “en bloc” news cited only mentioned case numbers without any additional information. For this reason, they were disregarded during data collection.
• 14
  Email interview with Joyce Maria Magalhães Russi, press coordinator of the Federal Supreme Court, on Oct 21, 2016.
• 15
  Except when secrecy of justice, that is, when there is a risk of exposing private information of the parties and when the process contains confidential documents.
• 16
  ADI no. 296, which first arose on Jun 6, 1990 and whose final decision was handed down on Oct 17, 2000, was erroneously filed as ADI, which is, in fact, another way of centered control - Direct Act of Unconstitutionality by Omission (ADO). It differs from the scope of this study, which aims to exclusively select ADIs, as it consists of the statement of the Federal Supreme Court that the Public Power has incurred, in default of a certain obligation established in the text of the Constitution itself. In addition, it would authorize the Federal Supreme Court to spread a mere communication by informing that it is in “constitutional delay”, subject to the mandatory nature of the same decision, which is, in fact, an administrative body, with an obligation to comply with the determination of the court “with in thirty days” (CF, art. 103, §2º). For the same reason, ADI no. 1484 (filed on Jul 25, 1996 and judged on Aug 21, 2001) and the ADI no. 2778 (filed on Dec 13, 2002 and judged on Mar 11, 2003) were converted into ADO during the course of the process. The three previously mentioned cases also contained, in the reasoning for the final decision, the statement that there was “supervening subject loss”, as a result of further normative complementation. This is considered usual, especially when there is an extensive time lapse between the beginning and the end of the process. On the other hand, ADI no. 2621 (filed on Mar 5, 2002 and judged on May 6, 2003), by considering its conversion into Argument of Breach of Fundamental Precept (ADPF), which likewise escapes the object of the study, as it consists of another method of constitutionality control, intended to “avoid or repair injury that results from an act of the public power” - including those prior to the Constitution (Law no. 9882/1999).
• 17
  Employee Severance Guarantee Fund.
• 18
  A positive coefficient (+) when transformed by the logarithm, produces and Exp coefficient (β) greater than 1. A negative coefficient (-) produces an Exp coefficient (β) lower than 1. A coefficient of zero value produces an Exp coefficient (β) equal to 1, indicating that the independent variable does not affect the chance of occurrence of the dependent variable.
• 19
  Exp (β) - 1 * 100
• 20
  The typology based on complexity was grounded in the areas of Law, as used in Vanberg (2005). The categorization of the sample processes was intentionally assigned by the authors at the time of development of the database, based on the interpretation of the process menu.
• 21
  “While courts may be minority institutions, they are able to please a range of minorities, so they can gradually gain majority support by obtaining an equivalent to a “coalition of minorities” (GIBSON, CALDEIRA and BAIRD, 1998, p. 355).

• How to quote this article:

  MAGALHÃES, Rebecca; CARVALHO, Ernani. Lack of Communication Can Be Trouble: Brazilian Federal Supreme Court Decision-Making Behavior and Self-Promotion Strategies. Revista Direito GV, São Paulo, v. 19, e2320, 2023. https://doi.org/10.1590/2317-6172202320

• Declaration of Data Availability:

  The dataset of this article is available at SciELO Dataverse of Revista Direito GV: https://doi.org/10.48331/scielodata.KFYE6I.

Data availability

The dataset of this article is available at SciELO Dataverse of Revista Direito GV: https://doi.org/10.48331/scielodata.KFYE6I.

Publication Dates

• Publication in this collection
  24 July 2023
• Date of issue
  2023

History

• Received
  11 Oct 2021
• Accepted
  27 Jan 2023