Collaborative Research: International Currencies in Theory and History

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

Project Summary Mounting U.S. external indebtedness, the advent of a single European currency, and the rise of China are fueling discussion of whether the dollar is at risk of losing its international currency role. The answer matters: recent estimates (e.g. Gourinchas and Rey 2005) put the annual benefits to the United States of the dollar's key currency role at 2 per cent of GDP. There is also the danger that, if the dollar abruptly loses its position as the dominant international currency, its exchange rate could fall sharply, with uncertain consequences for global macroeconomic and financial stability. The prospects for the dollar as an international currency are not something on which existing scholarship sheds much light. The theoretical literature is small and stylized. The empirical literature relies principally on a couple of dozen annual observations for the share of global reserves held in dollars and other currencies and on proprietary data available only to researchers at the IMF and national central banks. The intellectual merit of this proposal lies in the development of new data and application of new methods to these questions. The investigators' approach is theoretical and empirical, but also historical. Historical data provide the most obvious route to expanding the data base. In addition, changes in the identity of the dominant international currency do not occur very often - over the last three centuries there appear to have been only two such shifts, from the guilder to sterling and then the dollar - making any serious investigation necessarily historical. Finally, modern theoretical models invoke network externalities when explaining the dominance of particular currencies in international transactions, and where there are network externalities, there can be lock-in and path dependence, creating a role for history. The investigators will develop two historical data bases for the period prior to 1973, one on the currency composition of official reserves, the other on the actual use of currencies in foreign exchange markets. The first of these elaborates existing data, although the work of constructing it will be highly data and source intensive. The second data base is entirely new; it will be constructed by assembling information on the currencies that are actually traded in foreign exchange markets. The investigators will analyze these data using standard regression techniques but also the methods used by statisticians and sociologists to study social networks - which is precisely the approach suggested by the theoretical literature emphasizing network externalities. In this way the investigators hope to establish the roles of current policy (as manifested in external indebtedness, inflation proneness, and exchange rate instability), economic structure and past economic performance (as manifested in the weight of the key-currency country in the international system), and historical path dependence in the acquisition and loss of international currency status. The broader impact of the proposal and project will be to shed light on the dollar's prospects as an international currency, the immediacy of the challenge it will meet from current and future rivals, and the implications for United States.