The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
The Role of the Dollar as an International Currency
Remarks
Alan S. Blinder
Vice Chairman
Board of Governors of the Federal Reserve System
Washington, D.C.
before
Conference sponsored by
Federal Reserve Bank of Dallas
"Exchange Rates, Capital Flows and Monetary Policy
in a Changing World Economy"
Dallas, Texas
September 14, 1995
The Role of the Dollar as an International Currency
I would like to talk tonight about a subject of interest to
the Federal Reserve, to other central banks, and, I hope, to you:
The role of the dollar as an international currency. There is
much talk these days about the current or prospective decline of
the dollar as a reserve currency. But much of this talk is
remarkably undisciplined by facts. So I'd like to explore with
you what we do and do not know about the dollar's role as the
linchpin of international transactions. Is it really declining?
And what are the implications of such a decline for the United
States?
Before you measure a phenomenon, you must define it. But
there is no clear definition of what it means to be the world's
centreil international currency. I would suggest that the tacit
working definition is a blend of four characteristics which,
among them, encompass the three classic roles of money as a store
of value, medium of exchange, and unit of account:
First, an international currency should constitute a
preponderant share of the official reserves of central banks.
This, of course,
is the narrow definition of an international
reserve currency.
Second, the currency should be used extensively as hand-tohand currency in foreign countries.
Third,
it should be used to denominate a disproportionate
share of international trade.
Fourth,
it should have a. dominant role as the currency-of-
choice in international financial markets.
Note that the last three characteristics are the results of
millions of private economic decisions— about what currency to
hold, about how to invoice trade, and about the currency
denomination of securities. The first characteristic results from
government decisions. But these are, in turn, probably guided by
market realities.
On all four counts, there is really only one serious
contender for the title of principal international currency
today: the U.S. dollar. But is the dollar's role declining? As we
shall see, the answer is: probably yes, but slowly.
But before getting enmeshed in the facts, a bit of
historical perspective may be useful. A glance back over the
centuries suggests that the international use of a country's
currency is strongly linked to the centrality of that country in
world trade. International trade in Europe initially was centered
on regional fairs where merchants and local buyers gathered. By
the 12th century, Italian bankers were playing a pivotal role in
serving these merchants with coins, credit, and what we would now
call a payments mechanism. Later, especially after the discovery
of the New World, the Spanish joined in.
However,
it would be stretch to claim that Italian or
Spanish cities provided the first international currency. That
honor probably belongs to the Dutch guilder. As commercial
activity and trade shifted northward in Europe in the 17th
century, deposits at and notes issued by the Bank of Amsterdam—
2
denominated in guilders— became the means of payment for much
trade in the western world. This dominance lasted until the end
of the 18th century, when Napoleon's army occupied Amsterdam— an
event, which, I imagine, caused some severe disruptions to the
guilder payments system!
The action then shifted to Paris and, especially, to London
as Britain (a) went on the gold standard in 1816 and (b) became
the world's biggest exporter of manufactured goods and biggest
importer of raw materials. British banks and financial
institutions came to dominate the financing not only of intraEuropean trade, but also of trade with the Orient. Soon foreign
banks began settling transactions in pounds.
The outbreak of World War I knocked Britain off the gold
standard and somewhat diminished London's role as the world's
preeminent financial center. Its upstart rival, of course, was
New York. The period from 1914 to 1931 saw the dollar gradually
gain on sterling as the principal international currency— a
process that was greatly accelerated when Britain suspended
convertibility in 1931 and then war ravaged Europe. But sterling
continued to play a large international role for years after
Bretton Woods, and London remains a major international financial
center to this day. There is a good deal of inertia in these
matters.
The period of dollar dominance probably began with World War
II. The question, of course,
suggests,
is when and how it will end. History
first, that the honor of being the world's premier
3
currency moves around the globe as patterns of trade change. But,
second, that these things evolve slowly, although major events
like wars can certainly hurry things along or even change their
direction. The guilder dominated world markets for almost two
centuries; the pound for about a century. The dollar has now had
a run of about a half-century. Are our days numbered?
Don't count the dollar out too soon. Part of the reason is
the institutional inertia that I have just mentioned. Another
part is the fact that the successor currency is far from clear.
By most measures, the Deutsche mark is in second place now and is
gaining on us. But with monetary union somewhere on the European
horizon, the long-run future of the mark itself is in question.
Will the EMUs currency, whatever it is called, become the next
international reserve currency? The only other rival at present
appears to be the Japanese yen; and the nexus of world trade does
indeed seem to be shifting toward Asia. But, as we shall see, the
yen's international role is quite limited now; and, if we peer
into the distant future, one may legitimately wonder whether
Asia's dominant trading nation will be Japan or China.
The Facts: The Dollar as Official Reserve currency
Let me now turn to some facts, beginning with the dollar's
role in official reserve holdings.
The number of dollars held as official reserves has grown
rapidly, and with minimal interruptions, since the collapse of
the Bretton Woods system. Since 1986 alone, dollar reserves have
roughly tripled.
(See Chart 1.) But, of course, official reserves
4
Chart 1
International Reserves
(Foreign Exchange)
Billions of U.S. Dollars
Source: International Monetary Fund
in other currencies have also risen. The dollar's share of total
reserves rose irregularly from about 55% in the mid 1960s to
nearly 80% in the mid-to-late 1970s, and then declined in stages
back to about 56% in 1990. Since then, however,
again to about 63%.
it has risen
(Chart 2.) Focusing on the last 15 years, for
which we have the best data, the total net decline in the
dollar's share is only about 7 percentage points. That's
something, but not much.
Which countries gained at the dollar's expense? The
unsurprising part of the answer is: the yen, which rose from 4.3
percent of reserves in 1980 to 8.5 percent in 1994. The
surprising part of the answer is that the share of the Deutsche
mark in world reserves has not increased. It has instead
fluctuated in the 12-19 percent range since 1980 without much
trend.
Earlier this year, there were scare stories in the media
about central banks, especially Asian central banks, diversifying
away from the dollar. This was alleged to be one factor weighing
on the value of the dollar. As best we can tell, these rumors
were mostly unfounded.
For the region as a whole, the share of
the dollar in official foreign exchange holdings has been quite
stable at around 60%. It is true that some nations,
like Taiwan,
have reduced their dollar share in recent years; but others,
like
Hong Kong, have raised theirs.
To the extent that the dollar's share in central bank
portfolios has declined, what is the reason? One argument is that
5
Chart 2
Currency Composition of Official Foreign Exchange Reserves
Percent
Note: The figures through 1974 are not comparable with those since 1975,since currency specification is available for a much larger
proportion of the total in the later period. If rough allowance is made for the difference in coverage, the percentage in the U.S. dollar
would be close to 80 starting in 1970, when there was a large replacement of sterling by dollars in reserves, and remain that level until 1977.
From 1979, dollar portion of ECUs are added to the U.S. dollar. End of period.
Source: International Monetary Fund
the increased volatility of our exchange rate has diminished the
dollar's attractiveness as a reserve currency. Here we should
start and can finish the debate with George Stigler's old
question: Is this fact in fact a fact? The answer is no:
volatility has not increased in recent years.
(Chart 3.) I am
more inclined toward a simpler hypothesis: the share of world
trade and finance denominated in dollars has declined
(more on
this below), and central banks are just adjusting to a changing
reality.
Facts: The Dollar as Hand-to-Hand Currency
The statistical basis becomes a good deal thinner when we
turn to the next aspect of the dollar's international role: as a
medium of exchange for cash transactions in foreign countries. An
almost overwhelming collection of casual observations and
anecdotes tells us that the dollar is the currency of choice in
Latin America, Asia, Russia, the Middle East, and elsewhere. But
accurate, or even systematic, data are not available.
Federal Reserve staff estimate that 50-70 percent of U.S.
currency— a total now of $185-$260 billion--is held outside the
United States. The methods we use are, of necessity,
indirect;
for example, one technique is based on shipments of $100 bills by
the Federal Reserve Bank of New York. But the fact that different
methods, using different data, all point toward the same range
gives us some confidence in the estimate— which is, of course, a
pretty rough one.
Seignorage earnings on these foreign-held dollars are not
6
Chart 3
Volatility in the Exchange Rate*
(Absolute Values of the Weekly Percent Changes)
1980
1982
* G -10 multilateral trade weights
1984
1986
1988
1990
1992
1994
trivial. Using the current interest rate on Treasury securities
of maturity equal to the average of the Fed's portfolio to impute
interest, they amount to a hefty $11-15 billion per year. That is
a considerable fraction of our total earnings.
Is this highly-profitable role of the dollar diminishing? We
know that currency shipments abroad are growing faster in the
1990s than they did in the 1980s. And we believe that the share
of new currency flowing abroad exceeds the share of the stock,
meaning that the share of our currency held abroad appears to be
rising. But what about the dollar's share in the burgeoning
market for paper money circulating in foreign countries?
Anecdotal evidence says that the yen's share is quite small,
but the Deutsche mark's share is not. One recent German study
estimates that 30-40% of German currency is held outside Germany.
If correct, that would amount to roughly $3 5-$55 billion— a large
sum, but less than 20% of the volume of dollars circulating
abroad. Overall, it seems safe to assume that the dollar
comprises the lion's share of foreign currencies in hand-to-hand
circulation outside their native lands. But we know literally
nothing about whether the dollar's market share is rising or
falling.
Facts: The Dollar's Role in World Trade
The third aspect of the dollar's role as an international
currency is its use as the currency for pricing goods and
services in world trade— as the international unit of account,
to speak. It is well-known that a great deal of international
7
so
trade is conducted in dollars. The latest available data pertain
to 1992, so I will use that year for comparisons and hope that
things have not changed much since.
Taking U.S. trade first, one of the more amazing statistics
to me is that roughly 80% of American imports— largely the
products of foreign companies— are invoiced in dollars. Such a
large share is quite atypical, presumably because our domestic
market is so large.
(Chart 4.) Among the other major industrial
countries, the fraction of imports invoiced in destination
currency ranges from a low of 17% (Japan) to a high of 56%
(Germany). On the export side, of course,
invoicing in your own
currency is much more common— ranging from 4 0% to 77% for other
countries, and 92% for the United States.
But dollar prices are also common in trade that does not
involve the United States as either buyer or seller. Overall,
almost 50 percent of total world trade is denominated in dollars.
Since U.S. exports and imports denominated in dollars account for
only about 23% of trade, roughly 25% of world trade does not
involve the U.S. but is nonetheless priced in dollars. We have no
real competitor in this regard. The mark is the currency for
invoicing about 15% of world trade. But all of this is either
German exports or imports.
Has the dollar's role as the international unit of account
been declining? Superficially,
it appears to have fallen— from
56% of world trade in 1980 to 48% in 1992. But this decline was
entirely due to the declining relative importance of oil, which
Chart 4
Share of Imports Invoiced in Own Currency
(1992)
Percent
I
|WMV.VX
w .w .v .v .
ssssssss
SfffffSflt
WKV.W
Xv.vXv.v
rtWftvXv
•Xv.v.vX'
yX vX vX ',
iW.vXv
v.v.v*\\v«
w XXt
wv
w
woXvX
wX
vX
United
States
Germany
Source: European Commission
ww.vX
SrWftWS
$*$$$!:
%%%%#
v.v.v.v.v.
S«Xw»
•>:•»»»>:
France
Netherlands
w
XyX'X'
’X'X'X'X*!'
S'X^X'X'X
nrm'i'nTi
•
x.¥vS.:vW>:.¥
:>
v:
x
%••%%%%%%%%
?/Xv»vXv
Italy
.v.wXO.O
a
Sv .v .v .v
•vIvXvIw
XwAvIw
AV.W>IW
Japan
is of course priced in dollars, in world trade. In non-oil trade,
the dollar's share has been fairly stable for years.
The Facts: The Dollar's Role in International Finance
Historically, as I noted at the outset, the market's choice
of an international reserve currency has followed patterns of
trade— with Italy, Holland, Britain, and the United States
dominant in their turn. But, in the modern world, financial
considerations may overwhelm trade considerations. For example,
the vast majority of foreign exchange transactions today stem
from trade in assets rather than trade in goods. So we need to
consider the dollar's role in international financial markets.
Let us start with foreign exchange transactions. According
to the 1992 BIS survey, fully 83% of reported foreign exchange
turnover involved the dollar on one side of the transaction. The
two closest competitors were the mark and the yen, at 3 8% and 24%
respectively.
(In case you are wondering why these shares already
exceed 100%, there are two sides to every forex transaction; so
shares measured this way must add up to 200%.) A similar survey
in 1989 found that the dollar was involved in 90% of foreign
exchange transactions. So the dollar's role in this regard, while
still dominant, appears to be declining.
(The 1995 survey results
will be available soon.)
Turning to other financial markets, we find that in 1994
about 43% of banks' cross border claims were denominated in
dollars. The closest competitors were the mark and the yen, each
with about 14-15%. Similarly, about 47% of Eurocurrency deposits
9
were Eurodollars, versus about 17% for Euromarks and only about
5.5% for Euroyen. In both of these cases, however, the dollar's
market share has declined rapidly and substantially from peaks
that were 20-25 percentage points higher about 15 years ago.
In international bond markets (defined roughly as bonds
floated outside the borrower's country), about 35% of the
outstanding issues are presently denominated in dollars— versus
16% in yen and 11% in Deutsche marks. But here, too, the dollar's
franchise has been eroded substantially in the last 10-15 years.
Its market share was about 53% in 1981.
Thus,
it is in world capital markets that the dollar's
dominance has eroded most. Surprisingly, at least to me, none of
this erosion can be attributed to the diminishing share of the
U.S.
in world GDP. That share has not in fact declined since
1980. So we must look elsewhere for an explanation.
One candidate is the fact that, while the U.S. went first
and probably farthest, many foreign countries have followed our
lead in liberalizing and deregulating their financial systems.
Financial innovation naturally follows in the wake of
liberalization. And some of this deregulation and innovation,
such as easing restrictions on foreign participation in domestic
markets and domestic participation in foreign markets,
particularly impacts cross-border transactions. We believe that
these developments have, on the whole, been salutary for the
world financial system and for the global economy more generally.
But, as a side effect, they have diminished the dollar's
10
preeminence in international finance.
Implications: What Does It All Mean?
In sum, wherever we have data the message seems to be more
or less the same: The dollar is still unquestionably the world's
dominant international currency by any conceivable definition.
But its preeminent position is eroding slowly, as mainly the
Deutsche mark but secondarily the yen move up on the pecking
order. What, then, are the implications for the United States?
Let me go back over the list of four characteristics,
focusing
now on possible implications rather than bare facts.
Central bank use of the U.S. dollar as official reserves
presumably raises the demand for the assets that serve as
reserves— mostly, Treasury securities. That,
in turn, should lead
to lower interest rates on those securities relative to
comparable private securities. To look for such evidence, Board
staff compared yield spreads between government and private
securities in the U.S., U.K., Canada, Germany and Japan. It turns
out that the yield spread in the U.S. is not unusually large.
(See Chart 5.) Similarly, a decline in the demand for dollars as
central bank reserves should reduce this spread; but,
in fact,
the spread seems to have widened slightly in recent years.
In brief,
it is not clear that the dollar's role as an
official reserve currency gives the United States Treasury any
bonus in the form of lower borrowing costs. But,
if it does,
there is no evidence that this bonus is diminishing.
The chief implication of the use of paper dollars as a hand-
11
Chart 5
Long-Term Bond Spreads: Private Less Government
(1985-1995 Average)
Percent
M w Xs
iv Iv
K
v
I v Iv
AVAi
wXvX'X*!*
>v»v
.•XwXw.w
m
m
Jv.vXvXv.Vi
A'i'A'KWK*;
w.vXv.vX*.
!;X»
v.v.vIvXvIv
.v.v.w.vXy
{wMwHvK
W iv I v K v I v i
¥:%W&8S
uvXvAw.*
•Vi>Y*ViV»y
Canada
Germany
SSSS2&W8
vKw.vv.v
iV»V#ViV<V«V
•w.v.v.w*
WAV.V.W.*
Japan
United
Kingdom
'.'.V.V.V.V.V.
United
States
to-hand medium of exchange in foreign countries is that the U.S.
government earns considerable seignorage profits. In essence, the
Fed borrows interest-free from foreign holders of our currency,
invests the proceeds in U.S. government securities, and turns
over the profits to the Treasury. As noted earlier, this
seignorage revenue amounts to perhaps $11-$15 billion per year.
Should demand for U.S. currency decline, this revenue source
would, of course, decline proportionately. I mentioned earlier
that we do not know whether the dollar's share of the world
currency market is increasing or decreasing. But we are fairly
sure that the absolute volume of foreign holdings of U.S.
currency is growing faster than our GDP, the federal budget, or
almost any other indicator of the size of our domestic economy.
Thus seignorage is becoming a relatively more important source of
financing for the U.S. government. For example,
it is now roughly
the same size as federal receipts from estate and gift taxes.
The third aspect of the dollar's role as an international
currency is its use in invoicing trade. The interesting question
here is whether the fact that most of our imports are denominated
in dollars affects the pass-through of exchange-rate changes into
domestic prices— either in the short run or the long run.
The short-run case is easier to make. Suppose the dollar
depreciates.
If foreign goods sold to American importers are
priced in foreign currency, their dollar prices rise immediately
and automatically— unless foreign suppliers decide to cut their
home-currency prices. But there is no such automaticity if these
12
same goods are invoiced in U.S. dollars, as most of them are.
Instead, dollar prices rise only if foreign exporters make
affirmative decisions to raise them. On a strictly neoclassical
view of the world, currency denomination should be an
institutional detail of no importance to pricing. But in a world
with both nominal and bureaucratic rigidities,
it might matter.
I suspect that dollar invoicing may help explain the
relative immunity of U.S. domestic prices from exchange-rate
influences. But this is just a conjecture, neither supported nor
refuted by research, despite the best efforts of Board staff. The
data we need to test it are simply not readily available.
Last, but certainly not least, comes the dollar's role in
world financial markets. Here, two plausible implications spring
to mind.
First, just as was the case for official reserves, a
worldwide portfolio preference to hold dollar-based assets should
lead to lower dollar interest rates, other things equal. And, of
course, any decline in this preference should erode that
interest-rate advantage. But many other factors— especially
expected changes in exchange rates— influence interest rate
differentials between dollar and non-dollar assets. So looking
for this (presumably small) effect in the data seems a bit like
looking for a needle in a haystack.
Second, it is at least believable, though by no means
certain, that the dominant role of the dollar in world financial
markets gives U.S. financial institutions a competitive edge over
13
their foreign rivals. For example, U.S. banks' have access to the
Federal Reserve as lender of last resort. And U.S. financial
institutions may have a comparative advantage in dollar-based
financing. After all, the dollar markets are our home turf. If
hysteresis matters in these markets, as it probably does,
American firms acquired a durable competitive advantage simply by
being there first.
Do American firms really have such an advantage? We simply
do not know. I can offer only one suggestive piece of evidence.
In the Eurobond market, American companies lead in underwriting
dollar-denominated debt issues, Japanese companies lead in
Euroyen financing, German banks lead in Euromarks, the French
lead in Eurofrancs, and so on. That's hardly a proof, but it does
point toward a home-court advantage.
Conclusion
One final, and very general,
implication is worth
mentioning. A shift away from dollars for any reason— including
international diversification— should, other things equal,
to a decline in the dollar's value.
lead
Indeed, worries that the
shrinking international role of the dollar is putting downward
pressure on the exchange rate seem to underlie recent concerns
about the dollar's preeminence in world markets.
But what is declining is our international market share, not
the absolute demand for dollars in world markets, which is
growing rapidly.
Furthermore, any influence of these
developments on the exchange rate must be swamped by other more
14
fundamental factors like actual and prospective monetary and
fiscal policy. So it seems unlikely that the dollar is being
weighed down by its loss of market share.
In sum, it would be going too far to say that the recent
concern over the declining role of the dollar as the world's
preeminent international currency is much ado about nothing.
Rather, it is a bit too much ado about relatively little. The
dollar's dominance has in fact declined, but only a bit— and
rather slowly. Furthermore, the costs of this decline to the
United States are hard to assess and look to be fairly minor. Of
course, I might have a different assessment if the dollar was
being rapidly dethroned and relegated to secondary status in
world markets. But that, fortunately,
the case.
15
is most emphatically not
@FedFRASER