header image

The Ditchley Foundation Annual Lecture V

8 July 1966

by H.  V.  HODSON, Provost of Ditchley
Mr. Dillon’s lecture was the fifth Foundation Lecture to be delivered since operational work began at Ditchley in the Spring of 1962.  After my Inaugural as Provost, the successive Foundation Lecturers have been General Lauris Norstad (1963), Mr. Arnold Heeney, Canadian Chairman of the United States—Canada Joint Commission (1964), and Lord Caccia (1965).  Lord Caccia is now Chairman of the Ditchley Council of Management, in succession to the Earl of Perth, who presided over the 1966 Lecture, and to whom tribute was then paid for his indefatigable leadership through three constructive years.
Mr. Douglas Dillon is among the most distinguished of contemporary public servants in the United States, having been American Ambassador to France (1953-57), Under Secretary of State for Economic Affairs (1957-59), Under-Secretary of State (1959-61), and Secretary of the Treasury (1961-65), under both Republican and Democratic Administrations, the last office being equivalent to that of Chancellor of the Exchequer.  He honoured us by using the occasion of the Ditchley Lecture to give not only a profound review of the causes of inadequate international monetary liquidity, but also clear and constructive proposals for their remedy.  The pages that follow, coming from such a great authority, are a very important—and, in view of the urgency of the problem very timely—contribution to study of this vital economic issue, whether by governments, central banks, financial experts, parliamentarians or any others concerned to affect official policies and support public action.
The lecture is printed exactly as it was delivered on 8th July, 1966.  The Foundation itself is precluded by its impartial and educational purpose from advocating opinions or entering into political controversy.  But it wishes to take this opportunity of thanking Mr. Dillon most warmly for coming to England at our invitation and delivering an outstanding Foundation Lecture.

The International Monetary System: An American perspective
Delivered by:
The Hon. Douglas Dillon, former United States Secretary of the Treasury
For the past five years the functioning of the international monetary system has been a primary concern of governments as well as a perennial subject of discussion among economists and financial experts.  During that period there have been a number of important innovations.  The General Arrangements to Borrow, the gradual construction of a network of swap arrangements between Central Banks, and the recent increases in the quotas of the International Monetary Fund have all served to strengthen and improve the system under which we operate.
But these developments, important as they have been, have only served as curtain-raisers to the debate which is currently under way, a debate which looks to further changes, considerably more radical in nature.
What I will have to say today represents my own personal thinking and does not purport to represent in any way the views of the Government of the United States or to mirror any clear-cut consensus in the American financial community.  Since this debate is breaking new ground, one finds a wide variety of opinion on our side of the ocean with the differences fully as great as when the problem is discussed internationally.
All of the changes currently under discussion have one objective in common—a system which will develop international liquidity in amounts adequate to finance continued growth in world trade and investment, while at the same time providing assurances against an oversupply of liquidity which could generate unwanted inflationary pressures.  The current discussions are of vital interest to us all, since a properly functioning international monetary system is necessary, not only to settle residual payments deficits and surpluses between nations, but also to facilitate the free flow of private funds needed to finance international trade and investment.
In the present system, the reserve currencies, sterling and dollars, ‘have a special place and importance.  They are called reserve currencies because national monetary authorities have been willing to hold part of their assets in these two currencies, along with their holdings of gold.  This willingness to hold dollars and sterling as official reserve assets developed slowly and as a direct result of the strength and usefulness of these currencies as vehicles for settling private transactions through out the world.
As world trade grew it became imperative to have one or two currencies in which private ‘transactions could easily be settled no matter where or between whom they might take place.  For this purpose currencies were needed’ whose value was known ‘to all, whose stability was unquestioned, which could readily be traded in any desired volume, and which could be converted at any time into other currencies.  First sterling and, more recently, the dollar have met this need.  It should be emphasized that this function of dollars and sterling as “vehicle” or “transaction” currencies is quite separate and apart from their use as reserve assets.  It is this “vehicle” use of our two currencies, rather ‘than their use as reserve assets, which provides us both with substantial earnings and so serves to strengthen our payments position.
The “vehicle” and “reserve” characteristics of the dollar and sterling are sometimes confused.  It has been said that the large-scale use of sterling and dollars as reserve assets must be continued in order to preserve their position as vehicles for world trade and investment.  This seems to me to be putting the cart before the horse.  Sterling and dollars do not owe their pre-eminence as “transaction” currencies to the fact that they are also acceptable as reserve assets.  Quite the contrary.  It was the extensive use of our two currencies in private transactions throughout the world that led monetary authorities to build and maintain substantial working ‘balances in dollars and sterling, balances that were required to meet the needs of their own traders and investors.  The use of dollars and sterling as reserve assets was a natural outgrowth of their position as “vehicle” currencies rather than the other way around.
The points I wish to make are simply these: irrespective of what the future may hold for sterling or for the dollar as reserve assets, the need for one or two “vehicle” currencies will continue on a steadily enlarging scale.  As long as dollars and sterling meet this need through the maintenance of convertibility and the assurance of stable value they will continue to play a unique and profitable role in the functioning of the international monetary system, no matter what the extent of their use as reserve assets.  And finally, this widespread utilization of dollars and sterling as “vehicle” currencies is a much more valuable asset to our two countries than their use as reserve assets.
Since the end of the war we have seen an unprecedented growth in world trade and international capital flows.  Not unexpectedly this has brought with it an enlargement in the size of the residual balances between nations—the surpluses and deficits that must be financed by the use of reserve assets.
This in turn means that larger national holdings of reserves are required to meet the bigger swings.  New gold supplies have been inadequate to fill this need.  Nor has there been any increase in reserve holdings of sterling.  This was not surprising since sterling holdings were so extensive in the early post-war period that they could not easily be enlarged.
The needed additional reserves have been found in the growth of official holdings of dollars, the counterpart of continued deficits in the U.S.  balance of payments.  These dollars have provided about three-quarters of the increase in reserves over the past fifteen years.  At first our American deficits were welcomed but, when they jumped to an average annual level of some $3 billion in the years 1958-64, there was mounting concern, both over the willingness of the United States to pursue policies that would maintain the value of the dollar, and over the inflationary impact that these large and con tinued deficits were considered to have in some surplus countries.
A school of thought developed, particularly in France, that challenged the whole gold exchange system as unfair — as giving the reserve currency countries the option of running permanent deficits to be financed by ever-growing foreign holdings of their currencies.  While this criticism had a ring of plausibility and attracted a certain following, it ignored a number of important factors.
First, throughout this period, despite continued payments problems, the United Kingdom never made any use of this supposed advantage.  It is hard to believe that, if the French thesis were correct, there would not have been some growth in official holdings of sterling.  In fact there was no such growth.
Second, although increased official holdings of dollars have served to finance a substantial portion of our U.S.  deficits, this period may be drawing to a close.  During 1965 official holdings of d remained substantially stationary.  It seems dear that there are limits beyond which reserve currency countries can no longer finance their deficits through ever-growing foreign holdings of their currencies.  It seems equally clear that the United States is now beginning to approach this limit.
The other side of this coin has been a growing realization on our side of the Atlantic that it is no longer in the national interest of the United States to see further increases in foreign official holdings of dollars.  Obviously large reserve holdings of dollars constitute a danger.  In time of stress they can be offered for conversion into gold in embarrassing amounts.  While this danger is potential and not immediate, it is none the less real.  It is largely the recognition of this danger which underlies the determined and continuing efforts of the United States to reach and maintain equilibrium in its international payments.
Success in these efforts will shut off dollars as a source of new reserves.  With inadequate supplies of newly-mined gold and no further growth in official holdings of reserve currencies, some other reserve asset will clearly he needed to make possible the steady growth in international liquidity that will be required to finance continued expansion in international trade and investment.  And that is why both the Group of Ten countries and the International Monetary Fund have been searching over the past year for an asset that could supplement, gold, dollars and sterling in the reserve holdings of the free nations.
While I understand that there is fairly general agreement among the experts of the Group of Ten that a new reserve asset will be needed at some time in the future, there is, as yet, no agreement as to when this need will arise.  All agree that up to now international liquidity has been adequate.  When it will become inadequate can be seen with certainty only after the fact — when overly restrictive policies have been generally applied.  The lack of agreement as to timing is largely due to the feeling by many that no action will be necessary until the U.S.  reaches payments equilibrium.  This feeling is often coupled with uncertainty as to when the United States can or will achieve this much-to-be-desired result.  This uncertainty in turn leads some to feel a lack of urgency in searching for a solution.
This lack of a sense of urgency may well be based on a false premise.  It could be dangerous.  For if our analysis is correct, and a situation is developing where official dollar holdings are approaching their practical limit, the need for a new asset may be upon us sooner than we think.  Certainly the 1965 deficit of the United States, which was met entirely in gold, provided no additional liquidity to the international system.  If the United States will have to settle any future deficits primarily in gold, free world reserves will no longer show the growth that marked the years prior to 1965.  Thus, it may well be that, irrespective of the date at which the U.S.  reaches equilibrium, dollars will no longer serve in the future as they have in the past as a steady source of growing reserve assets.  If this should prove to be the case it would lend considerable urgency to the search for agreement on the form of a new reserve asset.
While it may well be possible, and even desirable, to post pone the activation of any new mechanism for reserve creation until the need is clearer, it is essential to set up the framework of the new system well before the hour of need strikes.  In this situation the concept of contingency planning adopted by the IMF and the Group of Ten countries seems most appropriate.  It is to be hoped that this effort will be pushed vigorously to an early and successful conclusion.
At this point it might be well to digress for a moment and consider a suggestion for an alternative solution that has developed considerable popularity among economists, while eliciting little or no interest in government or financial circles.  This is the concept of variable exchange rates.  In this concept the need for reserves is minimized.  In its extreme form every payments surplus or deficit would! be compensated for, not by a flow of reserves, but rather by an appreciation or depreciation of the currency involved to whatever extent necessary to achieve balance.
It is not a mere coincidence that this theory enjoys most popularity among those economists who have been primarily engaged with the problems of an intractable domestic economy.  For them it is a highly attractive theory because it obviates the necessity for actions designed to reduce payments imbalance that might have unwanted effects on the domestic economy.  Their lives would be greatly simplified if it were possible to counteract any loss of international competitive position by an appropriate devaluation, and similarly to penalize surplus countries by appropriate upward valuations of their currencies.
While this proposed solution is very attractive on the surface, it suffers from one fatal defect.  The problem is quite simply that, for a steady growth in world trade, private traders must be able to make their contracts in a medium of constant value.  This means, at the very least, continuing stability of the “vehicle” currencies.  And the experience of the past ten years clearly indicates that the possibilities for continued growth of trade and investment are greatly improved by stability of the currencies of all the industrialized countries.  This fact is well recognized by practically all students of the subject.  But the other attractions of flexible exchange rates are so powerful that the search for ways out of the impasse continues.
We ‘have seen a number of ingenious suggestions designed to limit the variability of exchange rates so as to provide the benefits of flexibility, albeit only over a period of time, while also maintaining enough stability to allow private trade to flourish.  While I have the greatest personal respect for many of the proponents of such systems, it is difficult to avoid the conclusion that they are engaged in the never-ending search for a way to have one’s cake and eat it too.
It appears to me inevitable that, under any such system, human ingenuity would immediately concentrate on finding ways of discounting prospective shifts in exchange rates not just six months or one year into the future but for considerably greater periods of time.  This would so increase the financial costs of international trade as to pose severe and unacceptable barriers to its continued development.  If I were to go into full detail as to the pros and cons of the arguments on this matter, I could easily talk of nothing else for the rest of the evening.  Suffice it to say that I know of no one in a position of government authority in the United States, and no American active in international finance, who gives any sup port to the concept of a flexible exchange rate for the dollar.  Since floating exchange rates have been equally unpopular among responsible officials in the other countries of the Group pf Ten, there seems to be no possibility of movement along these lines.  Therefore, we are brought back once again to the necessity of working out the framework of a new reserve asset.
But before we take up the possible characteristics of such an asset, we must recognise that the mere creation of new assets will by no means solve the whole problem.  In the first place, it is clear that any new asset should be created only in amounts designed to meet global needs for increased inter national reserves, given a reasonable distribution of such re serves.  A new asset cannot and should not be looked to as a prop for the payments problems of any particular country.
But any system of reserve asset creation that is geared strictly to global needs will involve the necessity for somewhat more rapid adjustments of payments imbalances than we have seen in the past.  Hence the emphasis by all concerned on the search for ways and means of improving the adjustment process.  This search has not been easy.  So far no general rules of behaviour have been developed.  It seems clear that each case will have to be treated individually, taking domestic economic imperatives fully into account.
It is also clear that a properly-functioning adjustment mechanism requires appropriate policies on the part of surplus as well as deficit countries.  While we have made progress in this direction, much more remains to be done.  For surplus countries can still be found that are maintaining restrictive policies when such policies no longer appear appropriate to the state of their domestic economics, restrictive policies that seem to be attempting to maintain payments surpluses even at the cost of inadequate domestic growth.
One area that clearly requires improvement is that of longer term capital flows.  Methods must be found to improve international capital markets so that they can be responsive to balance-of-payments realities.  This is a subject that I have talked about a number of times over the past four or five years.  Unfortunately the difficulties caused by the inadequate development of capital markets on the European Continent, which I foresaw in my earlier speeches on this subject, have now all materialized.
Although the importance of this problem was soon generally recognized in Europe, nothing really effective was accomplished until the advent of the voluntary control program in the United States.  Then the pressures of demand for new capital from expanding American affiliates in Europe, coupled with the ingenuity of British and American investment bankers, succeeded at least in breaching the walls that had hitherto prevented the formation of long-term capital markets on the Continent appropriate to the present state of European industrial development.
The volume of financing done by these American affiliates during the past year has far exceeded the expectations of the bankers.  The potentialities of a free and active European long-term capital market have been convincingly demonstrated.  It remains to build on this experience until Continental long-term capital markets can approximate the efficiency that characterizes the New York and London markets.  Until this has been accomplished, undesirable capital controls of one sort or another, voluntary or otherwise, may well prove to be necessary as a part of the adjustment process.
In summation, suffice it to say that an improved adjustment process will he required for the effective functioning of any new system based on deliberate reserve creation; and that such an improved mechanism will require closer and better co ordination of monetary and fiscal policies between the leading industrial countries, surplus and deficit alike, than anything we have seen to date.
Just as a new reserve asset cannot and should not be de signed to facilitate the solution of the payments problems of individual countries, it cannot and should not be designed to carry any specific portion of the development burden.  The promotion of growth in the developing countries through coordinated foreign-aid efforts on the part of industrialized nations is highly desirable.  But he complex problems of foreign aid and growth patterns in developing countries should be handled on their own through the various agencies and forums available for this purpose.  We must not further com plicate the already difficult task of devising new ways of assuring adequate supplies of acceptable international liquid assets by saddling that task with the added problems that stem from the needs of developing countries for enlarged inflow of foreign investment capital.
Now for the new asset.  Whether it be conceived as an asset that can be owned, such as reserve units of one sort or another, or as an extension of automatic access to borrowing facilities, it will be credit in one form or another.  This means that it must be based on assets of real value.  For a reserve unit, the currencies making up the unit must be useable currencies.
This is why it has been appropriate, in the first instance, to pursue the search for agreement within the framework of the Group of Ten.  These ten countries plus Switzerland, which has joined them in the current studies, possess practically all the useable currencies of the world.  While contributors to a new reserve unit ought not to be limited to the Group of Ten countries, ‘there should be self-qualifying criteria based on the usefulness of the contributing countries’ currencies.  At the present time such criteria might add three to five countries to the Group of Ten.  Others could be added in later years as they achieved the requisite financial strength and stability.
In considering the credentials of the Group of Ten in connection with any new reserve asset, it must be remembered that this Group evolved as the result of essentially automatic criteria.  Any country willing and able to contribute $100 million or more in useable currencies to the General Arrangements to Borrow at the time of their negotiation in the fail of 1961 was welcome.  At least one country, Austria, that could have met the requirements was unwilling to do so at the time and so did not join the Group.
There are some ho have said that the creation of a new reserve unit based on those currencies that are readily useable would be some form of favoritism and would be unfair to the developing countries.  This reasoning is difficult to follow.  For it has always been the intention of the Group of Ten to see to it that a portion of the new assets to be created would be made available, free and gratis, in one form or another, to non-participating countries.  On the other hand, currencies that are inconvertible or not generally useful in international dealings cannot contribute to the strength and acceptability of a new reserve asset.  They have no place in such an asset, which must be carefully constructed out of useful currencies o as to ensure its ready and widespread acceptability.
Furthermore, I can see little logic to the notion that the countries with useful currencies which are the contributors to the creation of a new reserve asset should not have a special voice in deciding how their currencies should be used.  I firmly believe that the final decisions on a matter of world-wick importance such as this should be taken only in an appropriate world-wide forum, in this case IMF.  But I am equally firm in my defence of a special position for those who contribute the required assets.  A formula of this nature was developed for the General Arrangements to Borrow, with a special place for the contributing countries in a framework of overall IMF arrangements.  This formula has worked reasonably well.  I can see no reason why an improved and updated version of such a formula could not work even better in the administration of a new reserve asset.
The chief unresolved problem in the development of a new reserve unit is its relationship to gold.  This is of vital importance.  Any new unit must be acceptable to be useful, but it must not be more attractive than existing reserve assets or it would have disruptive effects that would outweigh any possible benefits.  While it is possible to imagine that a new unit might replace reserve currencies in the portfolios of Central Banks over a long period of years, it is required in the first instance as a supplement to reserve currencies not as a replacement for them.  Since reserve currency holdings now account for about one-third of the free world’s holdings of reserves, anything that threatens the stability of these holdings must be avoided at all costs.
For this reason no tie to gold in the creation and initial distribution of a new reserve unit can be accepted.  To create a link of this nature would be to precipitate a destabilizing rush by many to acquire more gold so as to obtain more units.  Conversely, to ensure acceptability, any new unit should probably have a gold guarantee as to its value.
This leaves open the question as to whether or not there should be any tie to gold in the transferability and use of a new unit after its original distribution.  Such a tie has many attractions.  A proper ratio, say fifty-fifty, would greatly facilitate the automatic operation of the new system.  On the other hand, it might limit the usefulness of units for countries that do not maintain substantial gold stocks.  There is also the danger that such a tie would imperil the position of the reserve currencies.  The difficulty with this argument, how ever, is that it is not susceptible to abstract proof one way or the other.  The only proof of such a pudding would be in the eating, and if it were to result in some fatal poisonings there would be no way to bring the patient back to life.  Under these circumstances discretion indicates that we should move.  with great care and only after giving every reasonable benefit of the doubt to those who fear that a direct tie between gold and reserve units in official transactions would imperil the current status of the reserve currencies.
So far I have only mentioned in passing the possibility of providing additional reserves through improved access to enlarged borrowing facilities rather than through the creation of a new reserve asset.  This approach suffers from the fact that the experience of the past six or seven years has shown that convertibility brings with it the probability of recurrent heavy movements of short-term capital.  To finance these swings countries with convertible currencies and reasonably well- developed short-term capital markets have a particular interest in adding to the reserves which they own rather than only to those which they may borrow.
A good technical case can be made that repayment obligations can be fashioned so that they will be no more onerous ‘than the need to rebuild’ depleted stocks of owned’ reserves.  Nevertheless, the psychological difference in approach to an asset which one owns outright from that to a right to borrow at the bank is so deeply rooted that it is unlikely to be over come.  In spite of this there are some who consider that primary reliance should be placed on improved borrowing facilities.  They feel that the creation of a new reserve asset would be an unnecessarily radical and dangerous solution.
While one must respect these doubts they do not appear to be compelling.  This does not mean that there are no advantages in improved access to credit.  Such advantages appear to exist, particularly in the case of developing countries that must rely heavily on receipts from the export of raw materials.  Here temporary price dislocations can cause severe balance of payments difficulties.  Such difficulties would seem to be best handled by enlarged: access to credit such as that available from the International Monetary Fund.
A possible solution would be a combination of increased and relatively automatic access to borrowing facilities and a new, reserve unit that would make possible larger owned reserves wherever they may be appropriate.  One way to achieve such a result would be for the countries that participate in the creation of the new reserve units to set aside an appropriate portion of the new units for use by the non-participating countries through the IMF.  The non-participating countries could receive increases in their IMF drawing rights on a basis equivalent to the gold tranche.  They would not be required to make any payment for these quota increases.  The liquidity of the TMF would be strengthened by the receipt of units convertible, into the useable currencies of the participating countries which, for their part, would not receive any quota increases.  Such, increases for participating countries would be unnecessary since they would have received appropriate amounts of new units in return for subscriptions in their own currencies.
I should emphasize that a combination of new units and increased borrowing facilities, such as I have described, would not mean any more rapid increase in overall liquidity than the use of either route alone.  The desired increase would simply be allocated in appropriate amounts between the two methods of reserve creation.
To conclude, I would like briefly to summarize my thoughts.  First and foremost, the United States, in its own interest, must promptly act to bring its international payments into equilibrium and to keep them there.  This is necessary to preserve the “vehicle” currency status of the dollar as well as to protect our gold stock which serves as the bedrock of the present international monetary system.  No supposed benefits from the dollar’s reserve currency status will permit us to postpone the drive for equilibrium.  In fact, the burdens of the dollar’s position as a reserve currency equal, and may well outweigh, any benefits it draws from this position.  Conversely, the “vehicle” currency status of the dollar is not only essential to the continued orderly growth of world trade, but is also a thoroughly profitable venture for the United States and must be maintained at all costs.  Similarly, the long-range interests of the United Kingdom would seem to lie more in the continued use of sterling as a “vehicle” or “transaction” currency than in any enlargement of its position as a recognized reserve asset.
Second, and flowing largely from the requirements of a stable “vehicle” currency for private international transactions, the concept of variable exchange rates must be rejected as unworkable, in favour of a system of relatively fixed rates.  Changes in rates should generally be limited to the currencies of developing countries where the international repercussions would be limited.  Changes in the parity rates of the currencies of industrialized countries should be avoided and only resorted to where no alternative solutions are avail able.  For the dollar, in particular, there can be no thought of any change in its gold value of $35.00 an ounce.
Third, growing world trade and investment can be expected to lead to enlarged temporary swings in payments deficits and surpluses.  Larger stocks of reserves will be required to finance these swings.  Newly-mined supplies of gold will not be large enough to carry the burden alone.  Nor can dollars any longer be counted on to provide the needed growth in international reserves.  This is so, either because official holdings of dollars appear to be reaching their natural limit, or as the consequence of the inevitable end of U.S.  payments deficits.  We are left with the necessity of finding a new source of inter national liquidity.
Fourth, additional liquidity can be provided either through the creation of a new reserve unit to supplement present owned reserves or through enlarged and more automatic access to borrowing facilities.  Some combination of both approaches may well prove to be the best answer to the problem.
Fifth, there must be improvements in the mechanisms of adjustment so as to minimize the burden on reserves in meeting payments imbalances.  This should include appropriate actions by surplus as well as by deficit countries and will demand a higher degree of international cooperation than any we have so far known.  One specific area requiring improvement is the functioning of international long-term capital markets.
Sixth, in order to ensure the continued healthy functioning of the international monetary system, it is absolutely essential that any new reserve asset serve as a supplement to the reserve currencies rather than as a replacement for them.  Since about one-third of the free world’s reserve assets are now held in the form of reserve currencies, any proposal that weakened confidence in the reserve currencies would serve to reduce rather than to increase overall international liquidity.
Seventh, because of the need to preserve the status of the reserve currencies, great care must be taken in designing any new reserve asset and in developing rules for its use.  To be useful such an asset must be acceptable along with gold and reserve currencies, but it cannot be preferable to either without risking serious damage to the orderly growth of commerce and investment.  For these reasons there can be no tie to gold in the original distribution of any new asset, although its value probably should be guaranteed in gold.  On the more difficult question of a tie between a new reserve unit and gold in transactions subsequent to the original distribution, great care must be observed to avoid any decisions that run a substantial risk of damaging the status of the reserve currencies.
Finally, as to tuning, we should urgently push forward so that contingency plans for new sources of liquidity can be completed, and the framework put in place before the need arises.  For when it does arise we will have to move relatively rapidly to avoid serious deflationary problems.  The solution to such a complex problem inevitably takes time.  We should give it all the time that is needed for intelligent decision, but we should not procrastinate.  The risk of delay is both great and unnecessary.  Good progress has been made over the past year.  It is to be hoped that arrangements for the new system can be completed and put in place during 1968 at the latest, subject only to a final decision on activation as the need arises.
A successful conclusion of the current negotiations on these problems would represent a landmark fully equivalent in importance to Bretton Woods.  It would be a tremendous step forward in assuring the continued growth and development of the Free World.  Failure could lead to a calamity far too serious to contemplate.  I am confident that these matters can and will be resolved successfully, and that a new and glowing page will soon be written in the development of international monetary cooperation.