Essays on banking theory and history of financial arrangements

• Ano de defesa: 2014
• Autor(a) principal: Ferreira, Murilo Resende
• Orientador(a): Cavalcanti, Ricardo de Oliveira
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Corridas bancárias; Ciclos de crédito; Bolhas de ativos; Crises financeiras; Monitoramento imperfeito; Sistema de indicadores antecedentes
• Palavras-chave em Inglês: Financial crisis; Imperfect monitoring; Early warning systems; Bank runs; Credit cycles; Asset bubbles
• Link de acesso: https://hdl.handle.net/10438/12421

This thesis contains two chapters, each one dealing with banking theory and the history of financiai arrangements. In Chapter 1, we extend a Diamond and Dybvig economy with imperfect monitoring of early withdrawals and make a welfare comparison between all possible allocations, as proposed by Prescott and Weinberg(2003) [37]. This imperfect monitoring is introduced by establishing indirect communication( trough a mean of payment) between the agents and the machine that is an aggregate of the financiai and the productive sector. The extension consists in studying allocations where a fraction of the agents can exploit imperfect monitoring and defraud the contracted arrangement by consuming more in the early period trough multiple means of payment. With limited punishment in the period of late consumption, this new allocation is called a separating allocation in contrast with pooling allocations where the agent with the ability of fraud is blocked from it by a costly mean of payment or by receiving enough future consumption to make fraud unattractive. The welfare comparison in the chosen range of parameters show that separating allocations are optimal for poor economies and pooling allocations for intermediary and rich ones. We end with a possible historical context for this kind of model, which connects with the historical narrative in chapter 2. In Chapter 2 we explore the quantitative properties of an early warning system for financiai crises based on the boom and bust framework described in more detail in appendix 1. The main variables are: real growth in equity and housing prices, the yield spread between the 10-year government bond and the 3-month interbank rate and the growth in total banking system assets. These variables display a higher degree of correct signals for recent crises (1984-2008) than comparable early warning systerns. Taking into account an increasing base-line risk ( due to increasing rates of credit expansion , lower interest rates and the accumulation of distortions) also proves to be informative and to help signaling crises in countries that did not go trough a great boom in previous years.