Return to search

Ensaios sobre o poder judiciário no Brasil

Submitted by ALEXANDRE SAMY DE CASTRO (samydecastro@hotmail.com) on 2013-09-17T20:25:19Z
No. of bitstreams: 1
TESE_SAMY_2013.pdf: 1776141 bytes, checksum: c0e5dd1535e5dd10fc890494d4475f4f (MD5) / Approved for entry into archive by ÁUREA CORRÊA DA FONSECA CORRÊA DA FONSECA (aurea.fonseca@fgv.br) on 2013-10-17T13:32:54Z (GMT) No. of bitstreams: 1
TESE_SAMY_2013.pdf: 1776141 bytes, checksum: c0e5dd1535e5dd10fc890494d4475f4f (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2013-10-21T18:20:28Z (GMT) No. of bitstreams: 1
TESE_SAMY_2013.pdf: 1776141 bytes, checksum: c0e5dd1535e5dd10fc890494d4475f4f (MD5) / Made available in DSpace on 2013-10-21T18:20:38Z (GMT). No. of bitstreams: 1
TESE_SAMY_2013.pdf: 1776141 bytes, checksum: c0e5dd1535e5dd10fc890494d4475f4f (MD5)
Previous issue date: 2012-12-18 / This thesis is a collection of empirical essays on the institut ional features of the justice system in Brazil, focused on state district courts of the first degree. It relies on econometric methods and insights from microeconomic theory, which are essential towards understanding and modeling micro - level data (case dat a). The first chapter investigates another important feature of judicial behavior: whether judges exhibit any sort of systematic bias towards poor litigants. Many surveys have shown that in the process of decision - making, judges in Brazil tend to put signi ficant weight on the social consequences of their decisions, in detriment of strict observance of the law. Survey results must be confronted with patterns revealed by actual case data, which justifies the study. The statistical analysis is based on an orig inal database, constructed by the author based on the collection of millions of civil cases for the State Justice of Paraná . Relying on a discrete choice model, the paper estimates the winning chances of poor litigants relative to non - poor litigants, contr olling for all case and jurisdiction characteristics. The main conclusion of the study is that poor litigants are less likely to win a case than non - poor counterparts. These results consider the mechanisms through which poverty conditions may affect case o utcomes: i) when poverty condition is a substantive fact of the case and part of the legal reasoning (which typically occurs in family cases); ii) selection of disputes: poverty condition implies changes in fee - shifting schemes, with a direct impact on the expected value of the dispute; iii) selection and incentives of attorneys: poor litigants may face selection and incentive problems with respect to attorneys, thereby reducing their chances of victory in court. After taking into account such mechanisms, t he study finds no evidence that judges tend to favor poor litigants. In particular, the comparison between poor and non - poor plaintiffs suggest the occurrence of opportunistic behavior, related to the abuse of indigent assistance benefits, which eliminates fee - shifting and, as result, all the downside associated to the law suit. The second chapter is an attempt to test whether there exists a trade - off between procedural fairness and judicial efficiency. Put another way, the fact that a case is resolved faster is potentially compromising to the quality of the sentence. Again, conv entional wisdom has been supportive of that dilemma, on the grounds that expeditiousness is detrimental of the compliance to the right to legal due process. The study presents an empirical framework to test the hypothesis that such a dilemma exists in the Brazilian justice system. The model relies on the basic premise that the quality of the judicial decision is inversely related to its chances of being reversed by the court of appeals ( Tribunal de Justiça ). The database is constructed by the author, this t ime based on information available for thousands of cases in district courts of Minas Gerais . The data comprises various civil cases and their respective appeals. Based on a linear probability model, results suggest that the dilemma isn't corroborated by the data. On the contrary, they are suggestive of a complementary relation between expeditiou sness and quality: cases that are tried faster are associated with 'better' sentences, in the sense that they are less likely to be reversed at the court of appeals. The empirical model includes all case characteristics, and discusses the possibility of se lection bias on the sample. The last chapter is a study on the decision patterns of courts of appeals in criminal cases. The main objective is to measure the impact of judicial backgrounds on appellate decisions, based on jurisprudence from the State Supr eme Court of Santa Catarina , Brazil. The study exploits a particular feature of the judicial selection process in Brazil: the Federal Constitution reserves 10% of the seats in State Supreme Courts to attorneys affiliated to the Brazilian Bar Association (O AB), and 10% to public prosecutors ( Ministerio Publico ). The remainder of the Court seats are to be filled by career Judges from district courts. Those three backgrounds provide a clear - cut distinction on ideological cleavages. Specifically, under the null hypothesis, we expect, in criminal cases, that former attorneys adopt a pro - defendant stance whereas former prosecutors adopt a pro - plaintiff stance. Based on a large sample of criminal appeals in Santa Catarina , this study takes advantage that cases are randomly assigned to judges, to identify the causal effects of legal backgrounds on judicial decision - making. Results suggest that former prosecutors are 20% more likely to decide against defendants than career judges. On the other hand, former attorneys a re 6% more likely to do so. Considering a sample of procedural cases only, we find much stronger effects: former prosecutors are 50% more likely and former attorneys are 50% less likely to decide against defendants. The strong background effects, found on procedural matters, reveal that former attorneys (prosecutors) tend to regard procedural rights much more (less) seriously than former judges. / A tese apresenta três ensaios empíricos sobre os padrões decisórios de magistrados no Brasil, elaborados à partir de bases de dados inéditas e de larga escala, que contém detalhes de dezenas de milhares de processos judiciais na primeira e na segunda instância. As bases de dados são coletadas pelo próprio autor através de programas-robô de coleta em massa de informações, aplicados aos 'links' de acompanhamento processual de tribunais estaduais no Brasil (Paraná, Minas Gerais e Santa Catarina). O primeiro artigo avalia - com base em modelo estatístico - a importância de fatores extra-legais sobre os resultados de ações judiciais, na Justiça Estadual do Paraná. Isto é, se os juízes favorecem sistematicamente a parte hipossuficiente (beneficiária de Assistência Judiciária Gratuita). No segundo artigo, estuda-se a relação entre a duração de ações cíveis no primeiro grau e a probabilidade de reforma da sentença, utilizando-se dados da Justiça Estadual de Minas Gerais. O objetivo é avaliar se existe um dilema entre a duração e a qualidade das sentenças. Dito de outra forma, se existe um dilema entre a observância do direito ao devido processo legal e a celeridade processual. O último artigo teste a hipótese - no âmbito de apelações criminais e incidentes recursais no Tribunal de Justiça de Santa Catarina - de que as origens profissionais dos desembargadores influenciam seus padrões decisórios. Isto é, testa-se a hipótese de que desembargadores/relatores oriundos da carreira da advocacia são mais 'garantistas' ( e desembargadores oriundos da carreira do Ministério Público são menos 'garantistas') relativamente aos seus pares oriundos da carreira da magistratura. Testam-se as hipóteses com base em um modelo estatístico que explica a probabilidade de uma decisão recursal favorável ao réu, em função da origem de carreira do relator do recurso, além de um conjunto de características do processo e do órgão julgador.

Identiferoai:union.ndltd.org:IBICT/oai:bibliotecadigital.fgv.br:10438/11239
Date18 December 2012
CreatorsCastro, Alexandre Samy de
ContributorsCunha, Alexandre dos Santos, Machado, Cecilia, Oliveira, Fabiana Luci de, Moura, Rodrigo Leandro de, Escolas::EPGE, Braido, Luís H. B.
Source SetsIBICT Brazilian ETDs
LanguagePortuguese
Detected LanguageEnglish
Typeinfo:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/doctoralThesis
Sourcereponame:Repositório Institucional do FGV, instname:Fundação Getulio Vargas, instacron:FGV
Rightsinfo:eu-repo/semantics/openAccess

Page generated in 0.0026 seconds