Le paysage bancaire européen se compose majoritairement de banques à capital-actions et de banques coopératives. Cette différence de structure de propriété nous amène alors à nous demander si les spécificités qui les caractérisent, tant dans leur périmètre d’activité, dans leur structure organisationnelle que dans leur mode de gouvernance, peuvent avoir un impact sur leur niveau de prise de risque et sur leur contribution à la stabilité financière du système bancaire. La récente crise financière dite des subprimes est un évènement sur lequel nous pouvons alors nous appuyer afin de répondre à cette interrogation. Dans cette recherche, et dans un premier temps, nous analysons en profondeur les raisons qui pourraient expliquer que le statut coopératif et les modes de gouvernance correspondant amènent ces banques à adopter des comportements moins risqués, à être mieux capitalisées que les banques à capital-actions. Nous tenons compte dans cette analyse de la diversité du modèle coopératif en Europe, marqué par une convergence plus ou moins forte vers le modèle dominant de banque SA., ce qui peut influer de façon différenciée sur la prise de risque et la stabilité financière. Dans un second temps, nous évaluons plus précisément, de manière empirique, la capacité de résistance et de résilience des banques au cours de la crise dite « des sub-primes », la plus grave qu’ait subi l’industrie bancaire depuis celle des années 1930. Nous nous attachons alors à vérifier si les différences que nous avons notées en termes de statut de propriété et d’hybridation du modèle coopératif (relativement à celui des banques à capital-actions), se traduisent par des différences dans leur prise de risque et leur solidité financière, caractéristiques qui sont des éléments déterminants de leur mission sociale, en particulier aux yeux des pouvoirs publics et des autorités monétaires. Pour modéliser l'influence de la structure de propriété des banques et des caractéristiques de gouvernance qui en découlent sur leur solidité financière, sur leur attitude vis-à-vis du risque et finalement sur leur capacité à gérer et surmonter la « crise des subprimes », nous avons construit une base de données comprenant 63 groupes bancaires européens coopératifs et à capital-actions. Nous avons collecté pour chacun de ces groupes des indicateurs de performance de gestion comptable et financière, puis de gestion des risques. Nous avons procédé à une classification des groupes bancaires coopératifs en fonction de leur degré d’hybridation. Nous avons eu notamment recours aux méthodes de l'économétrie des données de panel sur les années 2002-2011 pour mettre en œuvre notre démarche comparative. Au final, nous montrons que le comportement des groupes coopératifs en temps de crise s’est distingué de celui des banques à capital-actions, mais que le constat se doit d’être nuancé, selon le degré d’hybridation de ces groupes / The European banking landscape consists mainly of plc banks and of cooperative banks. This difference of structure of property brings us then to wonder if the specificities which characterize them in their scope of activity, in the organizational structure and in their mode of governance can have an impact on their level of risk-taking and on their contribution to the financial stability of the banking system. The recent financial crisis known as the “subprime mortgage crisis” is an event on which we can then lean to answer this interrogation.In this research, and at first, we analyze in depth the reasons which could explain why the cooperative status and the modes of corresponding governance lead these banks to adopt less risky behavior, to be better capitalized than plc banks, what can influence in a differentiated way on the risk-taking and the financial stability. Secondly, we estimate more exactly, in an empirical way, the capacity of resistance and impact strength of banks during the subprime mortgage crisis, the most serious crisis that the banking industry has known since the 1930s. We attempt then to verify if the differences which we noted in terms of status of property and hybridization of the cooperative model (with regard to that of plc banks), are translated by differences in their characteristic, financial risk-taking and their solidity which are determining elements of their social mission, in particular in the eyes of public authorities and monetary authorities.To modelize the influence of the structure of property of banks and the characteristics of the governance which ensue from it on their financial solidity, on their attitude towards the risk and finally on their capacity to manage and to overcome the “subprime mortgage crisis”, we built a database including 63 cooperative and plc European banking groups. We collected for each of these groups performance indicators of accounting and financial management, then of risk management. We proceeded to a classification of the cooperative banking groups according to their degree of hybridization. We turned in particular to the methods of the econometrics of the data of panel over the years 2002-2011 to operate our comparative approach. In the end, we show that the behavior of the cooperative groups in times of crisis distinguished itself from that of plc banks; and we show that the degree of hybridization of these groups is far from being neutral

Orientador : Prof. Dr. Armando Vaz Sampaio / Tese (doutorado) - Universidade Federal do Paraná, Setor de Ciencias Sociais Aplicadas, Programa de Pós-Graduação em Desenvolvimento Ecônomico. Defesa : 05/04/2016 / Inclui referências : f.143-153 / Resumo: O acesso ao crédito é frequentemente considerado como um requisito fundamental para o crescimento econômico e para o aumento do padrão de vida nas zonas rurais menos desenvolvidas. Nesse sentido, o crédito é um meio do qual dispõem as famílias na obtenção do que necessitam para produzir, estando a sua função estratégica relacionada à sua capacidade de contribuir com a viabilidade, sustentabilidade e expansão das atividades econômicas. Todavia, a maioria das redes financeiras brasileiras, principalmente os bancos convencionais, ainda não prioriza as regiões periféricas, sobretudo no meio rural, perpetuando-se as limitações no acesso ao crédito. Nesse contexto, as cooperativas de crédito, na lógica de finanças de proximidade, podem ajudar a resolver algumas das imperfeições de mercado que existem no setor de crédito rural, uma vez que podem reduzir os custos de transação e a assimetria de informações entre os agentes. Com isto, a possibilidade de acesso ao crédito, via cooperativas de crédito, pelos produtores rurais antes marginalizados pelo sistema financeiro tradicional pode gerar um impacto positivo para o desenvolvimento das comunidades em que se inserem, contribuindo para melhorar a qualidade de vida da população. Desse modo, esse estudo buscou avaliar quais os fatores determinantes do acesso ao crédito rural via cooperativas de crédito e via bancos, bem como mensurar os impactos do acesso ao crédito rural, via essas fontes, nas produtividades parciais do trabalho e da terra e na eficiência técnica dos estabelecimentos agropecuários da amostra considerada. Assim, usando microdados do Censo Agropecuário 2006 do IBGE, incluindo características do estabelecimento agropecuário e do seu responsável, considerou-se que os estabelecimentos poderiam ter três diferentes status de acesso ao crédito: nenhum acesso (restrição de crédito), desconsiderando os estabelecimentos que não acessaram crédito porque não precisaram; acesso ao crédito via cooperativas de crédito ou acesso ao crédito via bancos. Ressalta-se que foram considerados apenas os estabelecimentos cuja propriedade é de um produtor individual, tendo-se excluídos todos os demais. Dentre as técnicas econométricas utilizadas, um modelo logit multinomial verifica os fatores determinantes do acesso ao crédito rural via cooperativas de crédito e via bancos, em relação aos estabelecimentos que não acessaram crédito, controlando por características tanto dos estabelecimentos como de seus responsáveis. Os resultados sugerem que, dada a magnitude dos coeficientes, os estabelecimentos que tinham acesso à assistência técnica e cujo responsável participava de cooperativas e/ou outras entidade de classe tinham mais chances de acessar crédito rural via cooperativas de crédito e via bancos do que sofrer restrição de crédito. Tais resultados são confirmados pela análise discriminante linear realizada. A aplicação do Propensity Score Matching possibilitou que os estabelecimentos agropecuários das amostras fossem pareados em grupos equiparáveis. De modo geral, em todas as regiões, observou-se que os estabelecimentos que acessaram crédito rural via cooperativas de crédito e via bancos apresentaram melhores desempenhos nas produtividades parciais do trabalho e da terra e na eficiência técnica, quando comparados àqueles estabelecimentos que não acessaram crédito de nenhuma fonte. Entretanto, quando se comparam os estabelecimentos que acessaram crédito rural via cooperativas de crédito com aqueles que acessaram via bancos, percebe-se que os efeitos são diversos. Enquanto em algumas regiões as diferenças nas produtividades do trabalho e da terra e na eficiência técnica não foram significativas, caso das regiões Norte e Nordeste, em outras essas diferenças foram positivas (Brasil, Sudeste e Centro-Oeste) e até mesmo negativas (Sul) / Abstract: Access to credit is often regarded as a fundamental requirement for economic growth and rising living standards in less developed rural areas. However, most Brazilian financial networks, especially conventional banks, not priority peripheral regions, especially in rural areas, perpetuating the limitations on access to credit. In this context, credit unions, in a logic of proximity finance, may help solve some of the market imperfections that exist in the rural credit sector, since they can reduce transaction costs and information asymmetry between agents. In this sense, the possibility of credit access through credit unions, by farmers previously marginalized by the traditional financial system, can generate a positive impact on the development of the communities in which they operate, helping to improve the quality of life of the population. Thus, this study aimed to assess what are the determinants of rural credit access through credit unions and banks, as well as measure the impacts of rural credit access, by these sources, in partial labor and land productivities and technical efficiency of farms from the sample considered. Using microdata of Agricultural Census 2006 from IBGE, including characteristics of the rural farm households and its head, it was considered that the rural farms could have three different credit access status: no access (credit restriction), excluding those that did not access credit because they did not need; credit access through credit unions or credit access through banks. It is emphasized that were considered only the establishments owned by an individual producer, having excluded all others. Among the econometric techniques, a multinomial logit model was estimated to verify what are the determinants of rural credit access through credit unions and banks in relation to the rural farms that had no access (credit restriction), controlling by characteristics of the establishments and its head. The results suggest that, given the magnitude of the coefficients, establishments which had access to technical assistance and whose responsible participated in cooperatives and/or other entity class were more likely to access rural credit through credit cooperatives and through banks than to have credit restrictions. The implementation of Propensity Score Matching enabled the rural farm households of the samples to be matched in comparable groups. Generally, in all regions, it was noted that the rural farm households which accessed rural credit through credit unions and through banks were better in the technical efficiency and partial labor and land productivities when compared to those which not accessed credit from any source. However, when comparing the rural farm households which accessed rural credit through credit unions with those which accessed through banks, it is clear that the effects are diverse. While in some regions the differences in labor and land productivities and technical efficiency were not significant (North and Northeast), in other these differences were positive (Brazil, Southeast and Midwest) and even negative (South).