View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

To be published in
.t r
Aussenwirtschaf1,
Schweizerisches Institut fur AussenwirtschaftsStruktur- und Marktforschung, St. Gallen,
October 1974

CO*-L
August 22, 1974

LONGER-RUN ASPECTS OF INTERNATIONAL MONETARY REFORMBy
Henry C. Wallich,
Member, Board of Governors of the Federal Reserve System
Washington, D.C.

In the immediate future, the principal consequences of
the work done by the Committee of Twenty will be measures to which
the Committee, in its final report, refers to as "Immediate Steps."
These relate to guidelines for floating, strengthening of the IMF
through establishment of a high-level Council, a pledge against
trade restrictions, and a method to value the SDR, along with
others.—
The longer-run future of the international monetary system,
however, may well be shaped, in important respects, by what the
Committee refers to as "The Reformed System."

H

This long-run plan

was left as a torso, agreed in part but with important matters
unresolved.

It is generally understood that evolution, rather than

negotiation and explicit decision and agreement, will have to be

JV

The views expressed are the personal interpretation of the author
and do not necessarily reflect
views held in the U.S. Government,
or in the Committee of Twenty, nor those of the two principal
architects of the American plan, former Under Secretary Paul A.
Volcker and former Secretary George P. Shultz.

2/

"Outline of Reform," IMF Survey, June 17, 1974, Supplement,
Part II, p. 197. I have discussed these "Immediate Steps," in
Finance Magazine, September 1974, and Challenge Magazine,

september/october 1974.
3/




Ibid, Part I, p. 193*

federal Reserve Bank
of
I

Philadelphia
I Q E> A R Y

2

relied upon for longer-term reform.

But the understanding gained in

working out the general principles of the Reformed System, in clarifying
the issues and in appreciating more precisely the interests and intentions
of the participants, will play an important role in the evolutionary
process.

The future shape of the system remains uncertain.

But the

major alternatives will almost certainly be found to have been
implicit if not precisely spelled out in the agreed and the unagreed
portions of the "Reformed System."
/

It would be wrong to regard failure to reach an agreement
on long-run reform as revealing major conflicts among national
interests.

The existing disagreements are deep but narrow.

Nobody

is debating basic issues of the kind posed during the 1930's, such
as bilateralism versus multilateralism, autarchy versus trade liberaliza­
tion, tight control over payments against full exchange market converti­
bility.

The disagreements that separated the members of the CXX can be

overcome by an effort of political will.

The world has lived so well

on the capital generated by past efforts of this kind made following
World War II, that it seems to have become too complacent to repeat
them.

It is not a hopeful sign for the future if calamity rather

than comfort must be looked to as the mainspring of action.

But

evolution may very well lead us where cooperation failed to take us.
On that trip the "Reformed System" may be a useful guide.




3

Antecedents
When the Bretton Woods system finally broke down in the
summer of 1971, many observers looked to the U.S. to come forward
with some plan of action for repair and reconstruction.
was hesitant.

The U.S.

It was widely felt on the American side that the

U.S. role in the world had diminished, and that the U.S. could no
longer afford nor effectively assert the kind of leadership that
helped

to create the Bretton Woods institutions, the Marshall

Plan, and the GATT.

It took time before the U.S. was prepared,

beginning with the IMF meeting of September 1972, to put forward
some components of a plan.
The American approach to international monetary reform,
as well as the response on the part of other countries, reflected
the experience of the final years of the Bretton Woods system.
Everybody understood that exchange rates, in conditions of mounting
inflation and differential rates of real growth, could no longer be
as stable as they had been during the 1950fs and early 1960's.

It

was clear also that the world wanted a more symmetrical system than
that into which the Bretton Woods blueprint had evolved.

The

original design of that blueprint had been almost entirely symmetrical.
The same rules, rights, and obligations applied to Paraguay and to the
U.S.

But because reality was asymmetrical, the system evolved to a

dollar-gold exchange standard and eventually, when the dollar had




4

become convertible only in a very limited sense, into something close
to a pure dollar standard.
In its early stages, that system suited both the U.S. and
most other countries.

The U.S. received easy financing for its

perennial payments deficits.
reserves0

Other countries thereby acquired dollar

The U.S. was unable, in a system where all currencies were

pegged to the dollar, to modify its own exchange rate, but for many
years it felt no desire to do so.

Other countries could change their

rate and could thereby regulate the degree to which the system
required them to finance American deficits.
As time went on, both sides increasingly found the benefits
of the system less attractive while its costs seemed to become more
onorous.

The world became tired of financing American deficits and

inflating national currencies and price levels in the process.

The

U.S. became increasingly troubled by its inability to devalue as the
overvaluation of the dollar became more and more evident.
From this experience there developed a universal desire for
a more symmetrical system.

In such a system, the U.S. felt, it would

have the same ability as others to modify its exchange rate.

The

U.S., so other countries seemed to feel, should be required to make
the dollar convertible and settle its deficits in reserve assets like
every other country.




The role of the dollar would be greatly reduced.

5

Convertibility
In the fall of 1972 it seemed apparent -- which is no longer
obvious today -- that the decision-makers of the world wanted to
return to a fixed exchange rate system.

Academics might prefer

flexible rates, but finance ministers, central bankers, the private
financial sector, and internationally oriented businessmen saw fixed
rates as the rule and floating rates as a temporary makeshift.

Yet

a system of rates fixed by a peg to the dollar, in which every
currency would be convertible into reserve assets except the dollar,
was not negotiable.

Thus the American reform plan, as it evolved

and was gradually presented, had as its keystone the promise of
dollar convertibility.
It was clear that convertibility raised severe problems
for the United States.

But it also held one advantage.

A dollar

convertible into exchange assets would give the U.S. an effective
opportunity to control the dollar's exchange rate.

There might have

been ways other than by making the dollar convertible to accomplish
this major American objective.

Convertibility, moreover, was not a

foolproof means of assuring control over the dollar rate.

Under the

Bretton Woods system, even during the years when the dollar remained
convertible into gold d£ facto as well as die jure, other countries
could and very probably largely would have frustrated an American
change in the gold value of the dollar by changing their own gold




6
price accordingly while retaining their dollar peg.

But convertibility

into reserve assets probably was the cleanest way of regaining control
over the dollar rate.
Under convertibility, the U.S. could shift from one fixed
exchange rate to another while continuing to convert.

Alternatively,

by ceasing to convert and allowing the dollar to float, the U.S.
could achieve a rate change in accordance with market forces to the
extent that other countries did not intervene in exchange markets to
prevent this.
Convertibility at a fixed rate, nevertheless, raises severe
problems for the U.S.

A large economy with a small foreign sector

finds balance-of-payments adjustment at such a rate more difficult
than one structured inversely.

The number of dollars by which GNP

must be reduced in order to eliminate one dollar1s worth of trade
deficit is greater in a nearly closed economy than in an open one.
Payments adjustment via the income mechanism is costly.

Adjustment

via the price mechanism, to be sure, is easier for the nearly closed
than for the open economy, because the range of possible substitutions
of domestic production for imports and the elasticity of supply of
exports both are greater in the nearly closed economy.

But to activate

the price mechanism, given the usual rigidity of prices, would require
freedom to alter the exchange rate.




7

The greater difficulty experienced by a nearly closed
economy, such as the American, in adjusting its balance of payments
at a fixed exchange rate is not universally accepted.
economy, moreover, is not all that closed.

Since the 1950*s, its

average propensity to import has about doubled —
to six per cent.

The American

from about three

But there can be very little doubt that a large

country with a small foreign sector has less of an incentive to
subject its domestic economy to discipline in order to achieve a
balance-of-payments objective than has an economy where the foreign
sector is of major importance.

The U.S. is close to being an optimum

currency area, i.e., one that finds it preferable to adjust its
balance of payments by exchange rate movements rather than by deflating
or inflating the domestic economy.
Convertibility is difficult for the U.S. also because its
currency is used so widely in official and private international
balances and transactions.

Capital movements running into many

billions of dollars, in response to interest rate differentials
or speculative incentives, would require very large reserves.

The

fact that the Eurodollar market can create dollars adds to potential
demands upon the U.S. for conversion of dollars, even though the
dollars presented for conversion necessarily must be dollars in the
U.S. rather than Eurodollars.

Dollar flows among third countries

also can give rise to conversion demands if the country losing dollars




8

is in the habit of holding dollar balances while the country gaining
dollars is in the habit of demanding conversion.
Finally, planning for convertibility raised problems in
the light of the low level of American reserves and the large holdings
of dollars by foreign monetary authorities.

Resistance to U.S.

devaluation in 1971 had shown that it would not be easy for the U.S.
to achieve current account surpluses sufficient to permit accumula­
tion of substantial reserves.

So long as dollars were used in inter­

national settlements, moreover, the current account surplus might
merely cause the U.S. to earn back its dollars and reduce liabilities,
instead of acquiring assets.

Arrangements would clearly have to be

made for dealing with these problems, but they were not spelled out
in much detail in the early stages of the CXX discussions.

The

subsequent unsettlement of all balances of payments resulting from
the rise in the price of oil has for the time being materially
altered this aspect of the monetary reform problem, as it has so
many others.
Given these difficulties, the American plan might have
opted for some qualified form of convertibility.

For instance, the

U.S. might have offered to pay in reserve assets for the amount of
its trade deficit, or current account deficit, or some part thereof.
This would have avoided the need to convert dollar balances that
foreign monetary authorities acquired as a result of capital movements,




9

as well as the need to convert the existing official balances.

A

serious difficulty, under such a scheme, would have been the alloca­
tion of these reserve assets to countries which might have acquired
dollars through trade surpluses both with the U.S. and with other
countries as well as through capital movements.

Conceivably the IMF

could have acted to allocate reserve assets made available by the
U.S. in accordance with some key, such as members' total current
account surpluses, or surpluses with the U.S., or in accordance
with quotas.

Problems of computation as well as of equity would

have been serious but perhaps not impossible to overcome.

This,

however, was not the road chosen in the American plan.

The Defense of Dollar Convertibility
Instead, the American plan sought to make convertibility
livable for the American economy principally by two devices.

One

was a semi-automatic or presumptive system of balance-of-payments
adjustment activated by a reserve trigger.

The other was an option

to float when exchange losses -- or possibly gains -- became unmanageable.
Particular stress was placed on the need for symmetry -- surplus
countries were presumptively expected to adjust in response to the
same reserve indicators.

The method of adjustment was not specified.

It might take the form of domestic contraction or expansion at a
fixed exchange rate, or, more likely perhaps, adjustment of the
exchange rate itself.




10

Foreign countries who viewed themselves as potential surplus
countries saw many objections.
symmetry in adjustment.

One objection related to the postulated

Why, it was asked, should the countries that

had succeeded in getting into surplus positions and therefore supposedly
were doing things right be required to adjust along with those who had
deficits and therefore must have been following bad policies?
try to cure the quick as well as the sick?

Why

This had been an issue

that Keynes confronted in designing his Clearing Union, and at that
time the U.S. had taken a dim view of his proposition that surplus
countries as well as deficit countries should adjust.
tion was reversed.

Now the situa­

The U.S. had to overcome considerable resistance

before it became generally accepted in the CXX that adjustment should
be symmetrical.
A leaning toward automaticity or, as the U.S. preferred to
call it, presumption
trigger

in a country’s response to the indicator or

and the choice of a reserve indicator for triggering this

response were harder to defend.
automaticity was intended.

The U.S. made clear that no absolute

Triggering of the indicator was to create

a presumption of need for adjustment only.

But more than a mere signal

that it was time for possibly inconclusive, consultation and assessment
clearly was needed if reserve movements under convertibility were not
to become excessive.




11

In continuing informal discussions of monetary arrangements
during former years, it had become evident that few governments were
willing to surrender power over their exchange rate to an automatic
mechanism.

The automatic version of the "crawling peg," for instance,

had been widely rejected in the ongoing dialog between government
4/
officials and academics."”

For politicians, exchange rates are too

important a part of the economy to be left to economists and their
contrivances.
The reserve indicator had been evaluated during the public
debate over the crawling peg.

There seemed to be a good deal of

support for the view that, in terms of the likelihood that the
signals thrown off would be the correct ones, reserves would perform
as effectively as would spot or forward exchange rates or the current
or basic balance of payments.

But in a plan that sought to make the

dollar convertible, a reserve indicator rather than something else
was needed for a reason other than the timeliness of its signals.
This simple reason was that, when U.S. reserves ran low while those
of other countries ran high, the U.S. was in imminent danger of having
to suspend convertibility.

There could be no better signal for urgently

needed action.

4/

A proposal popular in academic circles during the late 1960’s,
involving frequent small exchange rate changes in response to
some indicator which would provide flexibility while making
speculation on future moves relatively expensive.




12

Fundamentally, the reserve indicator was a replica of the
textbook version of the pre-World War I gold standard.

The rules of

that legendary game -- which even in 1914 was not what it used to
be, and perhaps never was -- told central banks to contract when gold .
reserves were low and to expand when they were high.

The mechanism

of adjustment was the discount rate, rather than the exchange rate.
The problem with the gold standard, as with the U.S. plan, had been
that central banks were more ready to act when reserves were low
than when they were high.

Reserve indicators, however, involved

several additional difficulties.
One problem relates to the nature of these reserves.

There

is agreement in the CXX that the SDR should become the principal
reserve asset and that the role of gold and of reserve currencies
should be reduced.

In the U.S. view, however, immediate and total

elimination of the dollar as a reserve asset would deprive the system
of flexibility and would also be inconvenient for those numerous
countries, especially among the developing countries, that prefer to hold
their reserves in reserve currencies.

Total elimination of the dollar

from reserves, except perhaps working balances, would also, of course,
increase the difficulties that the U.S. might at times experience
under convertibility and increase the frequency with which adjustment
action may have to be taken by the U.S.




13
Another not fully resolved question concerning reserves
is whether they should be interpreted gross or net.

For a reserve

currency country, liabilities to official holders are an important
determinant of its net reserve position.

If holding of official

dollar balances were permitted, a net reserve indicator might be
activated by the ups and downs of liabilities even though the gross
reserve assets of the reserve currency country remained unchanged.
Still another reserve problem relates to the relative
advantages of using, respectively, reserve stock or reserve flow
indicators.

Situations could be visualized in which rapid loss of

reserves, even when their absolute level is still high, would require
adjustment action.

The same could be true, vice versa, for reserves

increasing rapidly from a low level.

On the other hand, the absolute

level of reserves obviously cannot be ignored and in many cases may
be the more relevant concern, especially in a system of convertibility.
The choice of a reserve norm, departures from which up and
down to certain levels would constitute warning or action signals,
presents another set of problems.

Countries presumably would enter

the plan with the reserves they happened to have at the time.

Gradual

movement up or down to the country’s norm, which then would be on a
rising trend over time, seems appropriate.

Norms in the aggregate

should add up to a desirable, i.e., noninflationary and nondeflationary,
level of world liquidity.




But countries* view of their appropriate

14

shares in world reserves and of the appropriate rates of growth of
these reserves may lead to results inconsistent with existing or
desirable aggregate international liquidity.
Even a detailed working out of the foregoing problems would
not necessarily guarantee symmetrical functioning of the adjustment
mechanism under all conditions.

An American deficit or surplus,

for instance, might have as its counterpart the surpluses or deficits
of a large number of countries.

This diffusion would mean that no

single country, other than the U.S., would necessarily experience a
reserve movement sufficient to carry it to a trigger point.
the U.S. would be required to adjust in that case.

Only

Alternatively,

some smaller country or group of countries might have a large surplus
or deficit mainly with the U.S. while the U.S. was in balance except
for these particular relationships.

Such imbalance might be sufficient

to trigger off the smaller participants but would probably not do so
for the U.S., nor of course for any third country, with asymmetry in
adjustment against the result.
These are highly technical problems that nevertheless
contain very marked elements of national interest.

The question

is not only which solution may be technically superior, but also how
these interests can best be balanced.




15

The second major device by which the U.S. sought to
make dollar convertibility livable, in addition to the reserve
indicator structure, was the option to float.

Thus the question

whether and in what circumstances the International Monetary Fund
might be authorized to allow or disallow a float acquired major
significance.

This, too, remains on the agenda for the future.

In addition, the right to float and its limitations involve the
question whether a reserve currency country, when it wants to float,
could request other countries not to peg their currencies to that of
the reserve

currency country.

be impeding

the free float

If they peg, they would, of course,

of the reserve currency.

On the other

hand, if the reserve currency country can deprive others of the right
to intervene by the use of its currency, -- pegging is an extreme case
of intervention -- difficulties may arise for these other countries in
managing their exchange rates.

Evaluation
Some observers suspected the American plan to have been a
thinly disguised prescription for a floating rate system.

In support

of this interpretation it was argued that the plan made no provision
for dealing with the so-called "dollar overhang,” did not concern
itself with

how

the UnitedStateswas to acquire additional reserves,

and ignored

the problem of making discrete changes in fixed rates,

especially when these were signalled ahead to the market by an




16
approaching trigger point.

It was argued that the plan thus failed

to provide the conditions that would be prerequisite for a serious
effort to achieve and maintain convertibility.
be rejected.

Such criticism can

The plan was presented piecemeal, as it was evolved

by a small group working within the U.S. Government.

Its gestation

period, and the attendant staff work, was far more limited than the
White Plan and possibly the Keynes Plan had enjoyed in pre-Bretton
Woods days.

In contrast to those discussions of 30 years ago, the

American plan and the work of the CXX as a whole had had a large
input of ideas from the academic community and from continuing
discussions of international monetary problems within various
groups bringing together officials, academics, and often businessmen.
The intellectual basis of the effort therefore was as broad and as
solid as the world's idea-generating processes could have made it.
But the final molding of these diverse and often conflicting inputs
into a tightly organized plan of notable internal consistency
necessarily had to be the work of a few people.
In the discussion of the CXX, the American plan has been
modified and combined with many other elements.
which the plan was most specific, however, —
structure and the exchange rate regime —

In the areas in

the reserve indicator

it has set a distinct stamp

on the "Reformed System" to the extent it was agreed by the CXX.
The way in which the resulting system would work, assuming its




17

unsettled portions to be sensibly compromised, would depend very
much on the behavior of national economies.

It could be a system

of very stable rates, if inflation were avoided, real growth rates
were not too dissimilar, and structural factors in national economies
did not change too much.

Such an outcome undoubtedly would be pleasing

all around.
The system could also, however, turn out to be one approxi­
mating floating rates.

To at least some of the participants, this

would probably be a disappointment.

The system is the result of an

effort to satisfy a widespread demand for dollar convertibility
without yielding to the desire, possibly implicit in this demand,
of subjecting the American economy to the frdiscipline of the balance
of payments."

The need to protect the American economy against

deflation, one that, with respect to the British economy, Keynes had
stressed very strongly at Bretton Woods, has given rise to a

blue­

print with a much higher degree of potential exchange rate flexibility
than many of the participants probably intended when the negotiations
began.

Meanwhile the world has moved to an improvised system of

total flexibility.
system.

To date it has not fared too badly with that

If this experience should continue into the future, a shift

to something like the long-term blueprint of the CXX should not
constitute as much of a change as might have been the case had it
been completed and adopted in 1974.