Three essays on financial markets and institutions (Brazil)

This dissertation is comprised of three essays. In the first essay we study the efficacy of sterilized foreign exchange intervention in a model where exchange rates are pushed away from their fundamental level by a rational bubble. The bubble is sustained by uncertainty about the post-crash price and about the market's absorption capacity. We show that the bubble will continue in equilibrium if there is a reasonable probability that the bubble might continue in the first period. We then examine how sterilized interventions can change the equilibrium. In particular, we compare announced with secret interventions, when the central bank intervenes only once in the life of the bubble (the degree of central bank commitment in bringing the exchange rate back to its fundamental level is crucial to determine whether the central bank will prefer to announce the intervention or keep it secret). Finally, we show that the central bank must gradually increase the size of interventions under a continuous intervention strategy. In the second essay we study the effect of stochastic volatility models in simulating distribution of returns that are comparable to the historical fat-tailed distributions. We show that it is possible to increase the probability mass toward the tails by changing the decay factor appropriately and to match reasonably well the historical evolution of volatilities. We also show that, within PSA approach, stochastic volatilities can shift CTM probabilities towards lower credit categories, which is more consistent with stylized facts observed in volatile environments like Brazil. In the third essay, we have performed a large number of simulations using comprehensive data for 28 of the largest Brazilian banks, which invest heavily in government loans. Our results can be summarized as: (i) Brazilian banks appear to show significant solvency problems only when government default is accounted for; and (ii) simulating banks simultaneously allow for capturing the interbank propagation channel, which is not very significant in the case of Brazilian banks because of the low fraction of interbank lending.