Chapter 4

THE DUAL CURRENCY SYSTEM REVISITED

RONALD I. MCKINNON

Stanford University

In 1963, after looking at data on the integration of commodity trade and applying the theory of optimal currency areas, a world with two major currencies – Western European and American – seemed best suited for reconciling internal and international policy needs.¹ Have the succeeding seven years thrown up new data or new theories which significantly strengthen or weaken the arguments in favour of such a dual currency system?

The major argument pursued then was that currency and capital-market integration should follow the flow of commodity trade. Those countries which are major trading partners should maintain a single fixed exchange rate system because continuous exchange rate adjustments are costly and inefficient (money illusion is lacking) between economies which are integrated with each other. Conversely, those blocs with a relatively small trade connection between them should rely much more on some system of flexible – even freely floating – exchange rates to secure external adjustment. Exchange flexibility can only be used efficiently (the relevant price elasticities are high) for those economies with slight trade dependence.

Furthermore, the argument ran that independent internal monetary and fiscal policies are desirable in countries with a small external trade connection as there is no reason to believe that they should experience the business cycle in unison. A continuously floating exchange rate would facilitate this independence.² On the other hand, in economies which are highly integrated in commodity trade, fluctuations in their private sectors are naturally highly correlated. Therefore there is less need for independent monetary policies within the bloc and a strong case to be made for imposing the uniform discipline that a common currency system would provide. Independent national policies are neither necessary nor desirable if exchange rate changes can upset carefully negotiated tariff, tax, and pricing policies.

First let us examine, using Tables 4.1 to 4.5, whether or not there have been any significant alterations in recent commodity trade flows which would bear on these arguments. From Table 4.1, there is a continuous upward trend in trade integration among EEC countries with intra-EEC exports as a proportion of national income rising from 5·37 per cent in 1952 to 9·8 per cent in 1968. However, from Tables 4.1 and 4.2, one can see that this EEC trend is the more pronounced part of generally increased integration among Western European countries, which have become more open to foreign trade throughout the period. Moreover, for EEC countries there is no ‘trade diversion’ effect, as dependence on trade outside the bloc has remained about the same at a ratio of exports to national income of about 12·0 per cent from 1952 to 1968.

Thus, although there is a strengthened case for a unified currency system within the EEC, the use of continuous exchange rate adjustment vis-à-vis the outside world remains questionable. If one were to define the optimal European fixed exchange rate region, it would likely include a number of countries outside the EEC like Switzerland and those in Scandinavia and perhaps dependent economies outside Western Europe.

In contrast, the formal trade connection between the EEC and North America remains small although it has grown relatively rapidly from 1–20 per cent in 1952 to 2·16 per cent in 1968. Thus, exchange flexibility remains feasible between the two major blocs although the case here may not be quite as strong as it once seemed.

Within the North American bloc, however, there has been a striking change in the position of Canada. After remaining fairly constant at about 12·5 per cent of national income between 1952 and 1962, Canadian exports to the United States have risen to 18·1 per cent in 1968. This remarkable increase has some highly visible manifestations such as the complex agreement to merge the Canadian-American automotive industry. The portents for the future are for much greater reliance of the American economy on Canadian petroleum and natural gas, thus increasing integration even further. Since complex contractual arrangements, like those in the natural gas and automotive industries, are easily disrupted when exchange rates move, the case for maintenance of Canada’s dependent monetary status is very strong. One could even envisage further moves towards complete monetary integration with the United States. The main obstacle is the very real instability in American Federal Reserve policy, to which Canadian authorities may not wish to submit but on which they have not been able to improve.

So the statistical data suggest that integration within the two major blocs is increasing and the case for common monetary policies within each has been strengthened. Among other industrial countries, however, the position of Great Britain remains as ambiguous as ever. A glance at Table 4.4 suggests British integration with the EEC is no greater than it is for the United States and Canada, although integration with both is increasing. From Table 4.5, one can see that Japan remains closely linked in trade to the United States but at a relatively much lower level than Canada. Putting these two aside as weak possibilities to be linked to either bloc, the principal conceptual problem then evolves around the relationship between the major blocs. (This is not meant to downgrade the tremendous administrative problems inherent in pulling EEC monetary systems and capital markets together, the failure of which may necessitate second-best solutions such as gliding parities.) However, there remains the major question as to whether the industrialized non-communist world is optimally a single currency area or whether one should work towards some organized exchange flexibility between the two blocs on the assumption that each can be unified internally.

Have any major institutional changes occurred in the last seven years which would alter our thinking on the dual currency system? Although the dollar was used heavily as an official reserve currency prior to 1963, its role as a private international currency, through the New York financial community and through the Eurodollar markets, has expanded enormously since then. Indeed, the dollar can now be viewed as international ‘money’ and one can show how this dollar-based monetary system plays a very useful and non-exploitive international role. There is a fairly sensitive supply mechanism working through financial intermediaries and through the U.S. trade balance which permits internationally held dollar balances to expand more or less in line with the international demand for them.¹

The question then becomes whether or not this dollar-based monetary system should be the nucleus for a currency area covering the industrialized non-communist world. Financial integration with the EEC as proposed in the Barre plan, which may now be really beginning after years of little or no progress, should continue as fast as possible. However, should such integration ultimately aim to construct a monetary system which has a high degree of independence from the United States, possibly insulated to some degree by a flexible exchange rate? In doing so, should it then lessen the dependence of individual European countries on dollar-based financial mechanisms (the Eurodollar market)?

Financial integration within Europe is necessary but not necessarily sufficient to displace the dollar as a European currency. Obviously, no final answer can or should now be given to the desirability of this displacement occurring. However, the welfare gains or losses to Europeans will revolve around:

Economic Integration through Trade Flows among Industrial Countries¹

Table 4.1 TRADING POSITION OF THE EEC

Table 4.2 COMBINED POSITION OF EEC AND EFTA

Table 4.3 POSITION OF UNITED STATES AND CANADA

Table 4.4 POSITION OF GREAT BRITAIN

Table 4.5 POSITION OF JAPAN

¹ R. I. McKinnon, ‘Optimum World Monetary Arrangements and the Dual Currency System’, Banca Nazionale del Lavoro, No. 67 (December 1963) pp. 366–96.

² Although there may be some limitations on deficit-financed fiscal policy if capital is perfectly mobile.

¹ A fairly complete description of the demand for international money, its supply, the seigniorage gains from its issue, and the question of price level stability can be found in R. I. McKinnon, ‘Private and Official International Money: The Case for the Dollar’, Essays in International Finance, No. 74 (Princeton University Press, Princeton, April 1969).

¹ Taken from most recent data provided in the United Nations Monthly Statistics. The tables are modernized versions of those found in R. McKinnon, ‘Optimum World Monetary Arrangements and the Dual Currency System’, Banca Nazionale del Lavoro, No. 67 (December 1963) pp. 366–96.