Essays on government interventions in a foreign exchange market: Asymmetric interventions to anchor inflation

This thesis includes three essays analyzing the role of government interventions in anchoring inflation in an extremely open small economy, Taiwan, whose monetary authority has been frequently using large-size interventions since the 1985 Plaza Agreement. Taiwan's relative success in controlling inflation raises the question: What types of exchange rate management contribute to price stability under the floating exchange rate regime? Essay 1 investigates specific instances to find that interventions in Taiwan were used symmetrically to stabilize the exchange rate only when there was no existing inflation problem within the economy. With an existing inflation problem, the Taiwan monetary authority tended to sell foreign assets to offset depreciation pressures but did not appear to offset appreciation pressures. Empirical analysis supports the hypothesis that such asymmetric intervention behavior was used to help stabilize domestic price levels. Such asymmetric intervention, resisting depreciation pressures more strongly than appreciation pressures, is somewhat surprising for a highly export-oriented country. Essay 2 develops a model, within the general framework of optimal exchange-rate intervention, to justify the type of asymmetric interventions that have occurred in Taiwan. The model shows that, without an existing inflation problem, the optimal intervention rules are crucially determined by the types of shocks. With a large existing inflation problem, an asymmetric intervention pattern emerges as the optimal rule: offset shocks that would cause depreciation but not those that would cause appreciation. Contrary to the view of Mishkin and Savastano (2000) against the use of interventions in inflation-targeting countries, essays 1 and 2 together provide evidence that intervention on the foreign exchange market can be an important policy tool to anchor inflation. Based on the theoretical model developed in Essay 2, Essay 3 provides further policy implications by using Taiwan and U.S. data to do an empirical analysis. I perform the Engle-Granger cointegration analysis to obtain statistically superconsistent parameter estimators that ensure policy implications from the model are reliable. In particular, the essay addresses the quantitative question: "How much can inflation rate be brought down, if 1 percent of foreign assets are used by the monetary authority to implement an intervention? ".