A contribution to the theory of real exchange rate fluctuations

This dissertation consists of three chapters. Chapter One studies a dynamic general equilibrium model with multiple equilibrium prices in a two-good, two-country world in which the international capital movement is not allowed. The multiple market-clearing prices in this model are caused by both a strong home good bias and a low elasticity of substitution. There is no basis to assign the objective probability to these outcomes. This chapter presents a special case of the symmetrical state of the world economies, which have three equilibria; one symmetric and two asymmetric. The symmetric one is traditionally believed to be less likely due to some criterion of stability. Here, any prediction of the frequency of realization of these asymmetric equilibria are logically arbitrary.; Chapter Two follows the framework in Chapter One where quantity evolution can be predicted with certainty even though the price sequences are totally unpredictable. However, in Chapter Two, the state of the world economy is not symmetric. The important economic insight is that the world may start with price evolution in certainty but irreversibly gravitate into a zone of unpredictable disturbance. Without a global analysis of the model structure, we cannot preclude ultimate and perpetual disturbances in the future, just because of the tranquillity of the price movement in the past. Again, all of the studies in Chapters One and Two are undertaken by assuming that there is no international lending and borrowing and individuals' expectations are never unfulfilled.; Chapter Three continues with the model for immortal agents at first but allows international capital movement to take place. By focusing on the stationary equilibrium path, some surprising outcomes can occur. The early sections in Chapter Three serve mainly as an introduction to the overlapping generation model in the last section. The observational equivalence by Rao (1985) provides a convenient bridge between the model with infinitely lived agents and the overlapping generation model. Superimposing the model of immortal agents with multiple equilibria onto the overlapping generation framework, we find the self-fulfilling equilibrium in the overlapping generation framework loses much of its predictive power. Nevertheless, a careful analysis of the decision-making in the overlapping generation model provides interesting insight. The old generations are like strategical dummies in the price-taking world because they are near the end of their horizon. Thus, their expectations do not matter. Therefore, only the expectations of the young count. Yet, the younger generation is also facing an unpredictable future. The static expectation hypothesis forms an intuitive and useful benchmark, even though such expectations are not always fulfilled. Assuming steady state configurations for this production economy, the fulfillment of static expectation in each generation leads to stationary paths. The nonfulfillment of one stationary equilibrium path when it is replaced by another serves as a useful vantage point for studying the disequilibrium phenomenon.