Three essays on monetary search theory and exchange rates determination

This thesis consists of three essays. The first two essays employ monetary search theoretic models to examine issues related to money and specialization and bilateral exchange rate determination, respectively. The third essay, which is empirical in nature, investigates fundamental determinants of the Korean won - U.S. dollar exchange rate. Essay one investigates an individual's specialization decision in a decentralized trading environment. We allow individuals to choose the set of goods they are willing to produce. The degree of specialization is based on the number of varieties being produced, that is, greater specialization implies individuals have smaller production sets and vice versa. By undertaking greater specialization, individuals can reduce production costs but they confront a low probability of double coincidence matches. We mainly focus on two types of economies based on the individual's strategy regarding money acceptance, the pure strategy monetary economy and the mixed strategy monetary economy. In each economy we derive the optimal degree of specialization and show how it is affected by the introduction of money. When the economy is not liquid, introduction of money increases specialization and welfare in the pure strategy monetary economy. However, in the mixed strategy monetary economy, introducing money lowers specialization and raises welfare. Therefore, we conclude that if money works as a proper medium of exchange, introduction of money leads to specialization and raises welfare by extending the market. We also provide numerical examples to support our arguments. In essay two, we propose a search theoretic framework to study the endogenous determination of equilibrium nominal exchange rates. Our model consists of two countries and two currencies, and we consider two types of transactions. One type of transaction takes place in a decentralized market where prices are endogenously formed; it involves the exchange of fiat money for goods. The other type takes place in a centralized foreign exchange market; it involves the costly exchange of one currency for another. We assume that goods are differentiated by the country of origin and that agents are subject to taste shocks. Agents may be willing to diversify their currency portfolios if the cost of exchange is not too high. This work, with a simple version of the model, indicates there exists an equilibrium exchange rate for a given pattern of international transactions. Essay three involves the determination of factors influencing the Korean won and the U.S. dollar exchange rate movements. We test the real won/dollar exchange rate movements using both the traditional Purchasing Power Parity model and the sectoral productivity differential model. Cointegration analysis shows that there exists a long run equilibrium relationship between real exchange rates and productivity differentials, thus supporting the Balassa-Samuelson hypothesis. This paper also conducts Granger causality tests and innovation accounting in order to determine the relative weights of the factors affecting the movements of real exchange rates.