Full text of Working Papers (Federal Reserve Bank of Atlanta) : International Money and Common Currencies in Historical Perspective, Working Paper 2002-7

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

International Money and Common Currencies in
Historical Perspective
Gerald P. Dwyer Jr. and James R. Lothian
Working Paper 2002-7
June 2002

Working Paper Series

Federal Reserve Bank of Atlanta
Working Paper 2002-7
June 2002

International Money and Common Currencies in Historical Perspective
Gerald P. Dwyer Jr., Federal Reserve Bank of Atlanta
James R. Lothian, Fordham University

Abstract: The authors review the history of international monies and the theory related to their adoption and use.
There are four key characteristics of these currencies: high unitary value; relatively low inflation rates for long
periods; issuance by major economic and trading powers; and spontaneous, as opposed to planned, adoption
internationally. The economic theory of the demand for money provides support for the importance of these
characteristics. The value of a unit is arbitrary for a fiat money, but the other characteristics are likely to be
important for determining any fiat money that will be the international money in the future. If the euro continues
to exist for the next half century or so and has a relatively stable value, the authors conclude that the euro is likely
to be serious competition for the dollar as the international money.
JEL classification: E42, F33, N10
Key words: euro, international money, fiat money, dollar, search theory

The authors presented an earlier version of this paper at the Conference on Euro and Dollarization: Forms of Monetary
Union in Integrating Regions. They thank Sven Arndt, Benjamin J. Cohen, Cesare Robotti, and George von Furstenberg
for comments on an earlier draft of this paper. The views expressed here are the authors’ and not necessarily those of the
Federal Reserve Bank of Atlanta or the Federal Reserve System. Any remaining errors are the authors’ responsibility.
Please address questions regarding content to Gerald P. Dwyer Jr., Research Department, Federal Reserve Bank of Atlanta,
1000 Peachtree Street, N.E., Atlanta, Georgia 30309-4470, 404-498-7095, 404-498-8810 (fax), gdwyer@dwyerecon.com, or
James R. Lothian, Distinguished Professor of Finance, School of Business, Fordham University, 113 West 60th Street, New
York, New York 10023, 212-636-6147, 212-765-5573 (fax), lothian@fordham.edu.
The full text of Federal Reserve Bank of Atlanta working papers, including revised versions, is available on the Atlanta
Fed’s Web site at http://www.frbatlanta.org. Click on the “Publications” tab and then “Working Papers” in the navigation
bar. To receive notification about new papers, please use the on-line publications order form, or contact the Public Affairs
Department, Federal Reserve Bank of Atlanta, 1000 Peachtree Street, N.E., Atlanta, Georgia 30309-4470, 404-498-8020.

International Money and Common Currencies in Historical Perspective

1. Introduction
Is the euro likely to supplant the dollar as an international money? Such a question might seem
premature, especially since the euro physically has been in existence only for a few months. We are inclined
to think that it is not too early to start to think about this issue seriously since the implications are large for
the world’s monetary landscape.
The move to a circulating euro on January 1, 2002 has been followed by little or no difficulty in the
twelve countries involved. By most prognostications, moreover, the future for the new currency looks very
good. That was not the case at all a decade ago, as devaluations and widened currency bands shook the
Exchange Rate Mechanism (ERM). Skepticism with regard to the future of a single currency abounded
among both economists and financial professionals.
A major reason for this shift in opinion has doubtless been the change in the underlying economic
environment in the countries involved, in particular the greater convergence in monetary policies that took
place during the intervening period.
The purpose of this paper is to review the historical evidence on the question of international
monies. In contrast to other studies, we adopt a very broad temporal perspective, beginning with a
discussion of the Byzantine gold solidus or bezant that was introduced by the Emperor Constantine and that
served as a world currency for the next seven centuries. We continue with a review of medieval monetary

1

history and the international currencies of the Italian city states.1 We then turn to a discussion of the various
international monies that have existed from the early modern period to the present.
Using this review of the historical evidence as a background, we go on to discuss the theory
surrounding this issue. The underlying questions of interest are with the factors affecting the establishment
of common currencies and, more important, their longevity. Here we return to the earlier analyses of Carl
Menger (1892) and a century later Friedrich A. Hayek (1978). In both analyses, the key distinction using
some terminology borrowed from Hayek (1967) is between monetary institutions that are the result of
human action but not human design and institutions that are planned and orchestrated from on high.
2. Historical Overview
2.1. The Early Middle Ages
International monies, monies that circulate for use in transactions across national boundaries, begin
with the silver drachma first coined in ancient Athens in the fifth century B.C. (Chown, 1994). Judged on
the basis of the hoards that have been uncovered, not just within the Mediterranean region but throughout
Europe and well into Asia, the coinage of Rome – first the gold aureus and the silver denarius after the
currency reforms of Augustus – became the drachma’s successor.2 Beginning with Nero (A.D. 54-68) and
continuing into the early fourth century, currency debasement and inflation became the rule. An unsurprising
effect of this continual depreciation was a decline in the acceptability of the Roman coinage outside the
narrower confines of the Roman empire. The denarius, which had its specie content reduced entirely,
1

On these issues see Cipolla (1967) and Lothian (2002 forthcoming).

2

This discussion of Roman currency draws on Einzig’s (1962) monograph on exchange-rate
history. As Cohen (1998) points out, there was a substantial overlap period after Rome’s ascendancy
in which the silver drachma continued to circulate along with the Roman coinage.
2

ceased to function as money internationally. The aureus kept its fineness in terms of specie but was issued
as a lighter coin and became more commodity than money, being valued by weight rather than by its face
value.
After earlier attempts at reform failed, the Emperor Constantine introduced a new currency, the
solidus. Called the nomisma by the Greeks and the bezant by Western Europeans, the solidus continued
to be minted in Byzantium until that city fell to the crusaders in 1203, and for a while thereafter was still
minted in Nicea, the new seat of the then much diminished Byzantine Empire.
The solidus was a relatively heavy, full-bodied gold coin meant for use in large transactions. It soon
became an international currency. The Greek monk Cosmas Indicopleustes, a contemporary observer –
writing during the reign of the Emperor JustinianI (A.D. 527-565) – reports that the solidus was “accepted
everywhere from end to end of the earth.”3 He goes on to say that it is “admired by all men and in all
kingdoms, because no kingdom has a currency that can be compared to it.” The economic historian and
medievalist R.S. Lopez (1951) in reviewing the solidus’ history uses the term “Dollar of the Middle Ages”
to describe it. Based on the hoards of coins that subsequent research has uncovered, he traces its sphere
of influence from England to India, an area in which he claims (p. 211) it was accepted “as an instrument
of payment as good as gold itself.”
The solidus contained the equivalent of 4.5 grams of pure gold. The British gold sovereign and the
ten-dollar gold U.S. Eagle by way of comparison contained roughly 8.0 and 8.8 grams, respectively.

3

The phrase “dollar of the middle ages” is that of R.S. Lopez (1951). Both Lopez (1951, p.
209) and Cipolla (1967, p. 15-16) cite the statement by Indicopleustes. Cipolla goes on to discuss
corroborative evidence reported by earlier historians of the subject.
3

Valued at current market prices the gold in the solidus would be worth roughly $42 and the gold in the
eagle and the sovereign $73 and $68 respectively. Numismatic evidence suggests, moreover, that the
solidus’ gold content varied little from the fourth century through the mid-tenth century.
The result was an international money with an historically unparalleled life span. The solidus did
not, however, enjoy a monopoly over this full period. From the end of the seventh century, it shared its
position with the dinar, an almost identical coin minted in various parts of the Moslem world. Introduced
in the last decade of the seventh century by Abd el Malek, the fifth caliph (A.D. 685-705) of the Syrian
Umayyad dynasty, the dinar like the solidus kept a stable metallic content for centuries.
The fall from grace of the solidus and the dinar began at roughly the same time and for very much
the same reasons. Fiscal strains led to increased money creation which for commodity monies means
debasement in one or both of two forms: 1. a decrease in the weight of the coins: 2. an alteration of their
relative specie content, their fineness. From the standpoint of the usefulness of the currency in trade and
hence confidence in it, the first is the less harmful of the two. Coins can be valued by weight. Even without
debasement, coins often become worn through use and were weighed to determine their value.
For the solidus, debasement by reducing the weight started in the late tenth century under the
Emperors Nicephorus Phocas (963-969) and John Tzimiskes (969-976) and continued into the eleventh
century. As so often has been the case, financing high levels of government expenditures was the impetus
for debasing the money.4 The death knell for the solidus’ status as an international currency was sounded

4

Sussman (1993) presents a sound analysis of later debasements which illustrates the general
principles and raises the right questions. Rolnick et al. (1996) present some stylized observations about
debasements and Velde et al. (1999) present a simple analysis that highlights the general asymmetric
information that can rationalize the profitability of debasements.
4

in the late eleventh century when the fineness was reduced. Nevertheless, a solidus of close to the old
weight was still minted (along with inferior coins) in the early thirteenth century in Nicea following the fall
of Constantinople to the Crusaders in 1203. Debasement of the dinar also began in the late tenth century.
The explanation again appears to be largely fiscal.
2.2. The Thirteenth Century Commercial Revolution and the European Return to Gold
Although the reach of the solidus extended well into Northern Europe, it was used most extensively
in the Mediterranean region. Northern Europe as well as Western Europe had no counterpart to the solidus
and was otherwise less of a money economy than either the Byzantine Empire or the Moslem nations.
That situation changed in the thirteenth century. The thirteenth century was extraordinary in a
number of ways. It was a time of great learning and considerable scholarly interchange by all accounts.5
It was the start of a European commercial revolution, the chief manifestation of which was substantially
increased trade not only within Europe itself but between Europe and the rest of the world. It also was a
time of considerable financial innovation including the return to gold coinage in Western Europe.
International trade was centered around the fairs that were held regularly throughout Europe, the
fairs of Champagne being the most important. These fairs also were foreign-exchange centers. Initially the
money changers at the fairs, who also were merchants and gave rise to the term “merchant banker”,
confined their activities to manual exchange, the changing of one type of coin for another. Then as bills of

5

See Spufford (1984) on the commercial revolution and the financial developments that
accompanied it. On broader cultural issues – philosophical and economic thought and university life –
see Gilson (1991) and Schumpeter (1954).
5

exchange increasingly came into use, the money changers increased their roles and became intermediaries
in this market.
The reintroduction of a Western European gold coinage took place in 1252 with the striking of two
full-bodied gold coins in Northern Italy – the genoin of Genoa and then a few months later the fiorino (or
florin) of Florence. These two coins, but particularly the florin, were the world currencies. In the fifteenth
century, however, their place was taken by the ducato of Venice. All three coins circulated side by side
with coins of two other sorts – token coins used in small transactions and silver coins used in somewhat
larger transactions.
Gresham’s Law did not come into play for any of these three coins because exchange rates among
the various coinages remained flexible. The data that exist for this period suggest that there was a
surprisingly active market in foreign exchange. Exchange-rate data, in fact, indicate why the two coins, the
florin and the genoin, were attractive and became international currencies. We can see this in Table 10.1
which lists exchange rates relative to the florin at fifty-year intervals for a variety of currencies.
What stands out is the depreciation of all currencies relative to the florin, shown by the upward
trend in the exchange rates. A major cause of these movements was the series of debasements that took
place in all of the European countries throughout this era. A second influence was the discovery and
subsequent mining of silver in several European countries during the fifteenth century. These data are not
consistent with the modern notion that a metallic money necessarily is a stable money.
A second point to notice in Table 10.1 is the difference in the pace at which the exchange-rate
depreciation occurred. The English pound sterling, for example, showed relatively little movement, a drop

6

in value of 0.2 per cent per year over these two and a half centuries. The Castilian marivedi, in contrast,
registered a decline of 2.0 per cent per year over the 200 years for which data on it are available.
The debasement of these currencies does not seem to have been matched by increases in the
various countries’ price levels. The likely reason is that increases in money supplies were themselves
matched at least to some degree by increases in real output and in desired quantities of money demanded.
These currencies, therefore, also served as units of account for bills of exchange, credit instruments
which began in the thirteenth century to evolve into the principal form of international settlement.6 The
primary purpose of these bills of exchange was to eliminate shipment of specie when goods were bought
and sold. The mechanism was simple and evidently also quite effective, since the bill of exchange remained
the major instrument used in foreign transactions until the end of the nineteenth century. Discountable bills
still survive in modified form today, with payment now commonly being ensured by letters of credit.
A standard scenario involving the use of bills of exchange was something of the following sort: An
English merchant wants to buy cloth from an Italian exporter. To pay for this transaction and to avoid
having to ship specie, the English merchant purchases a bill of exchange from a London merchant banker.
The bill of exchange is then sent as payment to the Italian exporter who ships the cloth and remits the bill
to a merchant banker in Florence. The Florentine banker in turn pays the Italian exporter in florins and then
settles with the London banker. This settlement could be simply a bookkeeping transaction, the Florentine
banker canceling offsetting obligations to the London banker, or settlement could involve a shipment of

6

De Roover (1974, pp. 210-212) trace the development of bills of exchange to the letters of
exchange and cambium contracts used a century or so before. See McCusker (1978, pp. 19-23) and
Neal (1990, pp. 6-9) for discussions of the mechanics of the transaction involved in the use of bills of
exchange, and Einzig (1962) for a broader historical treatment.
7

specie at some future time at which the two have agreed to reconcile their books. The use of bills of
exchange became increasingly widespread from the thirteenth century on.
2.3. Common Characteristics of Early International Monies
Cipolla (1967) reviews monetary developments in both the early and later medieval period and
points to three common characteristics of the various international monies. The first is high unitary value.
We have already discussed this in connection with the solidus. It was also true of the dinar and the three
Italian coins. The dinar contained approximately 4.25 grams and the three Italian coins contained
approximately 3.5 grams of gold. Valuing these at today’s price of gold of roughly $290 per troy ounce,
this works out to prices of roughly $40 and $32 respectively. In the United States today by way of
comparison, most cash transactions are still carried out using twenty-dollar bills.
The second characteristic that the medieval monies shared during their respective hey-days was
intrinsic stability. They all kept the same weight and fineness during these periods though their values in
terms of goods and services did change over time. Correspondingly, in each instance after debasement
set in, the currency eventually ceased to serve as an international money.
The third characteristic is that the various currencies were all issued by strong economic powers
that were active in international trade. Since we will go on to consider international monies in today’s
context of fiat monies, it is useful to consider how well these characteristics apply to fiat monies.
Intrinsic stability, the second of these requisite characteristics, would seem to be the most relevant
for the success of fiat monies as well as metallic monies. In current terminology, it provided a nominal
anchor. It applies with at least equal force to fiat monies as to commodity monies since the real value of fiat

8

money can be easily affected by the issuer, something not as directly manipulable by issuers of commodity
monies.
A large economy with a substantial presence in world trade, the third observed characteristic, is
somewhat harder to pin down. Recently developed search-theoretic models of money, which we review
below, suggest however that such scale effects do indeed matter. An alternative view is that a strong tradeoriented economy is simply an indicator of brand-name capital. Like sterling during the period in which
Britain was the world’s major power and the dollar today, such currencies could be viewed as subject to
less risk of political upheaval. In either case, this generalization again applies with equal or greater force to
fiat monies.
High unitary value, the first characteristic, is another matter entirely. The value of a unit of money
is certainly not much of a restriction on fiat monies, because fiat monies weigh relatively little and it is cheap
to change the units to whatever value makes them useful in international transactions. Cipolla, however,
suggests that it might in fact be a mark of prestige. Lopez (1951) in his discussion of early medieval monies
makes a similar observation. He says (p. 214-215 ):
Clearly, then, the bezant was more than a lump of gold. It was a symbol and a faith, the messenger
of the divine emperor to his people and the ambassador of the chosen people to the other nations
of the world. We cannot make fair appraisal of the monetary policies of the empire in strict
economic terms since moral and psychological values also were involved and since these values
by affecting the internal stability and international prestige of the state had, in their turn, a bearing
on economic conditions.
It is perhaps more obvious now than when Cipolla wrote that there is a fourth characteristic of these
various world currencies: All of them were simply adopted. None of these international monies arose due
to anyone’s intention to create a new international money. Rather, their roles as international monies

9

evolved due to their becoming generally acceptable to others over time. They thus achieved their status
without any laws being passed, without any official monetary conferences being held, and without any
foreign ministers issuing joint communiques. They became world monies, to use Hayek’s (1967)
terminology in his article of that name, as “the result of human action and not of human design.”
3. International Monies from the Seventeenth Century to the Present
International monies have continued to be used in more recent times.
3.1 Later International Monies
As the sixteenth century wore on, the center of international trade shifted from the Mediterranean region
and the Italian city states to the northwest corner of Europe, first to Antwerp and then rather abruptly near
the close of the sixteenth century to the Dutch Republic. In the century that followed, the Dutch economy
experienced rapid growth. The United Provinces as a result became the world leader in shipping and in
trade as well as in international finance. The Dutch currency also became the key currency in international
transfers. It remained so, moreover, for most of the seventeenth and eighteenth centuries (see Dehing and
‘t Hart, 1997). Its strong suit, in addition to the preeminent Dutch economic and financial position, was its
intrinsic stability.
The rise of the guilder was an evolutionary process. At the end of the sixteenth century, close to
800 different foreign coins were accepted in the Dutch Republic as money, and by 1810 the number had
increased to nearly 1,000 ( Dehing and ‘t Hart, 1997). Out of this bewildering array, the guilder emerged
as the unit of account and the primary transactions currency. These two functions were, however, split
between two versions of the currency. The guilder banco, the deposit entries on the books of the Bank of
Amsterdam, became the unit of account. The circulating silver guilder was the medium of exchange. The
10

two guilders – banco and coin – were closely but not rigidly linked. In 1638 the Amsterdam magistrates
set a par value for the bank money in terms of the circulating medium, with the guilder banco at a slight
premium. This premium, the agio, remained generally small and quite stable. From1638 to 1775, the agio
was 4.1 per cent, implying an average internal rate of exchange between the two of 104.1 in guilder coinage
per 100 guilder banco. The standard deviation of the agio over this period was only 1.1 per cent.7 The
debasements that had been characteristic of the guilder until roughly 1630 was a thing of the past. As John
McCusker (1978, p. 42) summarizes the situation, “The unchanging metallic content ... during the
seventeenth and eighteenth centuries made Dutch money – with sterling – one of the soundest, most stable
currencies in the world.”
During the course of the seventeenth century, Dutch economic growth slowed appreciably and
Englishgrowth began to accelerate (Israel, 1995). Near the end of that century, England started to develop
into a financial power. The initial event here was the chartering of the Bank of England in 1694. Rapid
development of Englishcommercialbanking followed. In the latter half of the eighteenth century, the number
of London banks increased close to threefold and the number of banks outside London – the country banks
– grew even faster, from a dozen or so in 1750 to 334 in 1797 to double that number in 1810 (Ashton,
1955, pp. 179-183). During this period, the London Stock Exchange also was formed and an active
market in foreign exchange developed.
By the last quarter of the eighteenth century England had replaced the Dutch Republic as the
world’s leading trading nation and London had replaced Amsterdam as its chief financial center (Jonker,

7

These data are taken from McCusker (1978, Table 2.6, pp.46-51).
11

1997). Throughout, sterling remained a relatively stable currency and, although this stability was temporarily
interrupted in the decade or two surrounding the Napoleonic Wars, it was reestablished soon afterwards
and maintained until the start of the First World War. Sterling as a result, became the new international
currency, benefitting not only from its inherent stability but also, and we think more importantly, from the
new British primacy in international trade and finance.8
The United States became the world’s largest economy in the late nineteenth century, with U.S.
real Gross Domestic Product (GDP) surpassing U.K. real GDP in the last third of that century. British
financial and monetary dominance, however, continued until the start of the First World War. The outbreak
of war was associated with substantial gold flows to the United States from Europe including the United
Kingdom. New York, which had already become a major financial center, benefitted and London lost.
After the Armistice in November of 1918, the world tried to get back to normal. Many of the belligerent
countries returned to gold in the early 1920s. The United Kingdom only did so in 1925, after first
experiencing half a decade of wrenching deflation. Sterling retained some of its status as an international
money and reserve currency for a time but the position of the U.S. dollar in both regards strengthened.
Then the Great Depression and World War II intervened and financial and trade links between countries,
as much else, were disrupted.
The U.S. dollar was designated the official reserve currency of the fixed-exchange-rate system
designed at the Bretton Woods Conference in 1944. After the war ended, the dollar not only took on this

8

See Benjamin J. Cohen (1998, p.31-32) for a discussion of the links between Britain’s
increasing importance in international finance after 1815 and the rise of sterling as an international
currency.
12

role but became a widely used international money as well as a currency used in domestic transactions in
various countries. A good idea of what went on during this period and its resemblance to times past is
indicated by Cipolla’s discussion (1967, p.13) in his monograph on medieval monies in which he, like
Lopez (1951), compares the solidus to the dollar.
These pieces of paper [i.e. dollars], I knew, were more generally acceptable than any European
currency. In any part of Europe there could be no difficulty in finding people who would take them
as money . I could spend them everywhere, asking for everything. They were, they are, the
international currency par excellence.
Fixed exchange rates among the world’s currencies broke down in 1971 under the pressures
unleashed by inflationary U.S. monetary policy (Darby, Lothian, et al., 1982: Bordo, 1993). Many
believed at the time that the subsequent move to floating exchange rates would reduce the dollar’s role. This
was wrong on several counts. Official holdings of dollars as reserve assets remained high and greatly
increased. Private demands for dollars increased, particularly once the United States got inflation under
control and as other countries’ inflation worsened. One indication of the extensiveness of this phenomenon
is provided by the considerable number of papers in this volume that discuss dollarization.
3.2. Further Empirical Generalizations
This later experience is consistent with earlier experience: stability of value as well as country size
and prominence in trade clearly matter. What appears important also, however, is financial development.
All three of the later international currencies – the guilder, sterling and the dollar – were issued by countries
with dominant positions in international finance and relatively unfettered markets. Cohen (1971, 1998)
discusses this last characteristic under the headings of “exchange convenience” and “capital certainty.”
George S. Tavlas (1998) makes a similar point. He states that: “the issuing country should possess financial

13

markets that are substantially free of controls, broad (that is, containing a wide variety of financial
instruments), and deep (that is, having well-developed secondary markets).” Tavlas also argues that the
issuing country must be politically stable and militarily powerful. This was certainly true of the three
countries issuing these later international monies. It was also the case for the issuers of the earlier monies.
4. Theoretical Considerations
On one level, it is trivial to say that international fiat monies have arisen due to their being generally
acceptable. A domestic fiat money arises due to its being generally acceptable in transactions and for no
other reason. Why should an international fiat money be any different? Possibly the level of indeterminacy
is more obvious for international fiat money. Perhaps it seems obvious that Argentine people would prefer
to use the Argentine government’s fiat money if other things are the same. It is much less obvious whether
an Argentine will pick American, Brazilian, British or Swiss money when using non-Argentine money.
If legal restrictions are ignored, there is no particular reason to expect people in Argentina to bear
substantial costs to use money issued by the Argentine government rather than by some other government.
While people may be willing to bear some costs to use domestically produced money due to national pride
or other nonpecuniary considerations, there is no reason to think that they will bear substantial costs.
Programs to encourage people to buy domestically seldom, if ever, succeed. People buy the clothing that
has the highest value to them personally, even if it is produced by foreigners. There is an important
difference, though, between money and other domestic goods such as clothes.
The money that is useful depends on what money is used by others, whereas a pair of shoes will
be equally useful whether everyone else wears domestic or foreign shoes. This implies that there is a
coordination problem that must be solved for choosing money, a problem that is largely irrelevant for
14

clothes. It would be easy, though, to overemphasize the greater importance of coordination for money
compared to other goods. Driving a foreign car is more useful if some other people drive one too because
repair facilities will be nonexistent otherwise. The producer of the good, whether a car or a money, has an
incentive to assist users.
The choice of a fiat money is a radical indeterminacy. Why did people in the United States in the
1790s use United States dollars rather than pounds sterling or French francs? It might seem that legaltender laws resolve the question to some extent. Legal tender laws require that a money be acceptable in
payment. Everything else the same, if one money is legal tender and the other is not, the one that is legal
tender seems likely to be the money used. This observation has not, however, been demonstrated in a wellarticulated theory. In any case, alternative monies rarely are perfect substitutes.
4.1. More Stable Money
One reason that alternative monies are not perfect substitutes is that likely inflation rates for different
monies can differ substantially. Expected inflation and variable unexpected inflation impose costs on holders
of the money, and both expected inflation and variable unexpected inflation can generate revenue for an
issuer of money. Expected inflation is costly to a holder of money because the real value of the money
depreciates while it is being held. More variable unexpected inflation can generate costs because the
environment is less predictable and the risk of engaging in various activities is higher. Expected inflation
generates revenue directly to the issuer because printing more money creates higher inflation. More variable
unexpected inflation can generate more revenue because the unexpected higher inflation can be supplied
by printing more money, and the unexpected lower inflation raises the real quantity of cash balances
demanded, which can be met by printing more money (Auernheimer 1974). The literature on time
15

consistency and related issues, starting with Barro and Gordon (1983), shows that alternating periods of
high inflation fueled by printing money and low inflation with money printed to provide the increased
demand for money are difficult to sustain as an equilibrium. Even so, the incentive to produce such an
outcome means that it can affect even a rational expectations equilibrium. Moreover, this incentive can
affect the transition to a rational expectations equilibrium when agents are learning the preferences of the
money issuer.
In models with learning, governments can acquire a reputation for producing low, predictable
inflation, but they also can acquire reputations for producing high inflation (Sargent 1999). Even if the
choice of local currency is affected by legal tender laws, the choice of an international money is not affected
directly by such laws. Hence, a money’s reputation is all the more important for determining which money
will be used as international money. As a result, it is not surprising that Cipolla (1967) places substantial
weight on the stability of the money for determining what will be used as international money. It surely is
important.
4.2. International Trade
Why might it also matter how much international trade is done by a country, as Cipolla (1967)
suggests? At first glance, it is not obvious why this should matter at all. Even though Switzerland is a small
country, people might well find Swiss francs to be the most useful currency and the emphasis above on
inflation would suggest that as a real possibility. Even if a high-inflation country engages in a lot of trade,
why would anyone want to use its money as an international medium of exchange?
Reflection on the functions of money, though, suggests that the quantity of trade will matter, and
recent theoretical research provides support for this notion. The functions of money directly related to trade
16

include its being a unit of account and a medium of exchange. Considering costs of changing prices – menu
costs – suggests that reputation for low inflation is likely to be the important attribute for what currency is
used as a unit of account because it affects the frequency of price changes. On the other hand, if the money
used for a trade is different from the money used for posting the price, there is little obvious gain from
posting the price in a stable currency since the actual transaction price in the unstable money has to be
computed from the posted price in the stable money and the current exchange rate between the stable
money and the unstable money. Menu costs interpreted narrowly would have to be very important in order
to justify posting prices in a money different from the one used in the transaction. In domestic economies,
prices commonly are posted in currencies other than the one used in trades only in the case of extraordinary
inflations. Hence, it seems likely that the money used as an international unit of account will tend to be the
dominant international medium of exchange.
Search-theoretic models of money show why the amount of international trade will matter for the
choice of the money used as a medium of exchange. The indeterminacy of the choice of a particular fiat
money as an international medium of exchange is very clear in the context of the search theories of money
introduced by Kiyotaki and Wright (1989) and elaborated in a series of papers culminating in Trejos and
Wright (1995). These search-theoretic models are particularly well suited to examining the general issue
of international money because they do not impose the use of any particular money, or of money at all.
Cash-in-advance models presuppose that a money must be used, and models with money in the utility
function presuppose that a money is useful. Search theories of money allow for the endogenous choice of
a money to solve the general problem of finding a double coincidence of wants.

17

The generality of allowing for an endogenous choice of money is bought at the price of the
theoretical analysis being highly stylized. The choice of a particular international money is examined in
Matsuyama, Kiyotaki and Matsui (1993) and Trejos and Wright (1996). In these search-theoretic models
of international money, countries are defined effectively by the probability that people will meet, and people
can trade when they meet. People within a country have a probability of meeting another person from their
country and a separate probability of meeting a foreigner. The probability of meeting a person within the
country is assumed to be higher than the probability of meeting a foreigner. Trade is effected by an
exchange of a money for a good when both sides find such a trade advantageous. This trade can be an
exchange for domestic or foreign money.9 Matsuyama, Kiyotaki and Matsui (1993) show that the
likelihood of using a foreign money is higher in equilibrium if the other country is larger and if the probability
of meeting a foreigner is higher.
In short, a country that is larger and engages in more international trade is more likely to have its
money used as an international money. This conclusion is derived holding constant other characteristics of
the money, including the stability of the money.
These models have multiple steady-state equilibria for any sizes of economies and degrees of
integration, even holding constant the reputation of the money for stability. While it is possible at this level
of generality just to say that there are multiple equilibria and leave it at that, it is neither desirable nor
necessary to do so. Additional factors will determine the actual equilibrium observed. As Menger (1892,
p. 250) summarizes the point, even the use of money domestically is

9

Barter will not occur because demand and production are structured to rule out a double
coincidence of wants.
18

the spontaneous outcome, the unpremeditated resultant, of particular, individual efforts of the
members of a society, who have little by little worked their way [to a particular solution].
Actions that, in themselves, do not have any obvious implications for what money or monies will be
international money can affect the actual money used.
Even if there are many possible equilibria, it is plausible that the likely inflation rates generated by
a currency are of central importance among the variables affecting the actual equilibrium.
Other factors may matter too. On one level, there are curious things about the euro coins and
notes.10 The euro coins differ by country of issue, with different backs on each. The notes are the same in
all countries. These notes depict arches and other monuments, but the monuments are not real. The
supposition apparently is that Germans would not be impressed by their currency having pictures of the Arc
de Triomphe, and similarly the French would not be impressed by their currency having pictures of the
Brandenburg Gate. Perhaps this will not matter; then again, it may turn out to matter.
5. Conclusions
A useful set of stylized facts emerge from our historical review of international monies spanning the
fourth to the twentieth century. A sequence of international monies has existed historically, each occupying
center stage sometimes for several centuries and eventually being replaced by the next. The only exception
is the dollar, which is the current international money and, therefore, has not been replaced. These
currencies have four key characteristics: they had high unitary value; they had relatively low inflation rates

10

The euro was introduced with a substantial period to overcome some of the issues raised by
Selgin (1994).
19

for long time periods; they were issued by major economic and trading powers; and they became
international monies by human action rather than by design.
Economic theory explains the roles of these factors. The size of a country matters because it affects
the likelihood that someone in another country will trade with someone from that country and therefore the
usefulness of a common money in trading. On the other hand, there is no necessary reason that people from
two different countries will pick any particular money – especially a particular fiat money. Historically, this
coordination problem was solved partly by the relative values of metals, and therefore the usefulness of their
denominations, and partly by the historical evolution of the choice of money. With fiat monies, the historical
evolution of the use of any particular money assumes prime importance because the size of a monetary unit
is relatively easier to change. Relative inflation rates affect the cost of holding different monies and therefore
will affect the actual money used.
In recent centuries, international monies have been issued by countries with important financial
markets to which foreigners have relatively free access. The relative importance of this factor is an
important question for future research.
What does this analysis suggest for the euro? The euro area is large enough in terms of trade to be
a serious competitor to the dollar as an international money. Indeed, some European countries that are not
part of the European System of Central Banks (ESCB) have adopted the euro. Whether the euro will
replace the dollar in other geographic areas depends on two factors. The cost of holding euros relative to
the dollar – largely determined by the relative inflation rate of the euro relative to the dollar – will be very
important.

20

A more important factor for the euro than for other international monies, though, is the permanence
of the European Central Bank (ECB) and ESCB itself. The European Common Market is an international
organization and, while the single new money – the euro – has been created in ways that raise the cost of
leaving the ESCB, it is not impossible for a country to quit the system. The ESCB is more nearly analogous
to a currency union than to an international money in some respects, and currency unions’ history has been
one of formation and disintegration (Bordo and Jonung, 2000). If the European area evolves more nearly
into a common government, then such disintegration becomes impossible short of civil war. Otherwise, it
is an issue to be decided by participants in an individual country. In sum, the future of the euro itself is in
doubt.
Conditional on the euro persisting for, say, fifty or a hundred years and the European Common
Market not evolving into a single government, the euro may well supplant the dollar as international money
if the euro’s inflation rate is low relative to the dollar’s inflation rate. If it does, then the euro will be the first
international money that arose from planning rather than evolution.

21

References
Ashton, T.S. 1955. An Economic History of England: The 18th Century, New York: Barnes and Noble.
Auernheimer, Leonardo. 1974. The Honest Government’s Guide to the Revenue from Money
Creation, Journal of Political Economy 82, May-June: 598-606.
Barro, Robert J. and David B. Gordon. 1983. A Positive Theory of Monetary Policy in a Natural Rate
Model, Journal of Political Economy 91, August: 589-610.
Bordo, Michael D. 1993. The Bretton Woods International Monetary System: A Historical Overview,
in Michael D. Bordo and Barry Eichengreen (eds.) A Retrospective on the Bretton Woods
System, Chicago: University of Chicago Press for the NBER, 3-104.
Bordo, Michael D. and Lars Jonung. 2000. Lessons for EMU from the History of Monetary Unions,
London: Institute of Economic Affairs.
Chown, John F. 1994. The History of Money from AD 800, London: Routledge.
Cipolla, C.M. 1967. Money, Prices, and Civilization in the Mediterranean World, Fifth to
Seventeenth Century, New York: Gordian Press.
Cohen, Benjamin J. 1971. The Future of Sterling as an International Currency. London, Macmillan;
New York, St. Martin's Press.
Cohen, Benjamin J. 1998. The Geography of Money. Ithaca, N.Y.: Cornell University Press.
Darby, Michael, James R. Lothian and Arthur E. Gandolfi, Alan C. Stockman and Anna J. Schwartz.
1983. The International Transmission of Inflation. Chicago: University of Chicago Press for the
NBER.
Dehing, Pit and Marjolein ‘t Hart. 1997. “Linking the Fortunes: Currency and Banking, 1550-1800,” in
Marjolein 't Hart, Joost Jonker, and Jan Luiten van Zanden (eds.) A financial history of the
Netherlands, Cambridge and New York: Cambridge University Press, 37-63.
De Roover, Raymond. 1974. Business, Banking, and Economic Thought in Late Medieval and Early
Modern Europe (edited by Julius Kirshner) Chicago : University of Chicago Press, 1974.
Einzig, Paul. 1970. The History of Foreign Exchange. 2nd ed. London: Macmillan.
Gilson, Etienne. 1991. The Spirit of Medieval Philosophy. Notre Dame, Ind.: University of Notre Dame
22

Press.
Hayek, Friedrich A. 1978. The Denationalization of Money: An Analysis of the Theory and Practice
of Current Currencies. London: Institute of Economic Affairs.
Hayek, Friedrich A. 1967. The Result of Human Action not of Human Design, in Friedrich A. Hayek
(ed.) Studies in Philosophy, Politics, Economics, and the History of Ideas. Chicago: University of
Chicago Press.
Israel, Jonathan I. 1989. Dutch Primacy in World Trade. Oxford: Clarendon.
Israel, Jonathan I. 1995. The Dutch Republic: Its Rise, Greatness and Fall, 1477-1806. Oxford:
Clarendon Press.
Jonker, Joost, The Alternative Road to Modernity: Banking and Currency, 1814-1914, in Marjolein
't Hart, Joost Jonker, and Jan Luiten van Zanden(eds.) A Financial History of the Netherlands,
Cambridge and New York: Cambridge University Press, 1997, 94-123.
Kiyotaki, Nobuhiro and Randall Wright. 1989. On Money As A Medium of Exchange. Journal of
Political Economy 97, August: 927-54.
Lopez, Robert S. 1951. The Dollar of the Middle Ages, Journal of Economic History 11, Summer: 209234.
Lothian, J. R. 2002. Exchange Rates. Oxford Encyclopedia of Economic History, forthcoming.
Matsuyama, Kiminori; Nobuhiro Kiyotaki and Akihiko Matsui. 1993. Toward A Theory of International
Currency. Review of Economic Studies 60, April: 283-307.
McCusker, John J. 1978. Money and Exchange in Europe and America, 1600-1775. A Handbook.
Chapel Hill, N.C.: University of North Carolina Press for the Institute of Early American History
and Culture.
Menger, Carl. 1892. On the Origins of Money, Economic Journal 2, June: 239-55.
Neal, Larry. 1990. The Rise of Financial Capitalism, Cambridge: Cambridge University Press.
Rolnick, Arthur J., François R. Velde and Warren E. Weber. 1996. The Debasement Puzzle: An Essay
in Medieval Monetary History. Journal of Economic History 56, December: 789-808.
Sargent, Thomas J. 1999. The Conquest of American Inflation. Princeton: Princeton University Press.
23

Schumpeter, Joseph A. 1954. History of Economic Analysis. New York: Oxford University Press.
Selgin, George. 1994. On Ensuring the Acceptability of a New Fiat Money. Journal of Money, Credit
and Banking 26, November: 808-26.
Spufford, Peter. 1998. Money and its Use in Medieval Europe. Cambridge and New York: Cambridge
University Press..
Spufford, Peter. 1984. Le rôle de la monnaie dans le révolution commericiale du xiiie siècle, in Etudes
d'histoire monétaire / textes réunis par John Day, Lille: Presses universitaires de Lille.
Sussman, Nathan. 1993. Debasements, Royal Revenues, and Inflation in France During the Hundred
Years’ War, 1415-1422. Journal of Economic History 53, March: 44-70.
Tavlas, George S. 1998. The International Use of Currencies: The U.S. Dollar and the Euro, Finance
&
Development
35,
June:
46-49.
Available
at
http://www.imf.org/external/pubs/ft/fandd/1998/06/tavlas.htm
Trejos, Alberto and Randall Wright. 1996. Search-Theoretic Models of International Currency. Federal
Reserve Bank of St. Louis Review 78, May/June1996: 117-132.
Trejos, Alberto and Randall Wright. 1995. Search, Bargaining, Money, and Prices. Journal of Political
Economy 103, February: 118-41.
Velde, François R., Warren E. Weber and Randall Wright. A Model of Commodity Money, with
Applications to Gresham’s Law and the Debasement Puzzle. Review of Economic Dynamics
2, January: 291-323.

24

Table 10.1. Indices of exchange rates relative to the Florentine florin.
1252
1300
1350
1400
1450
1500
Austria
Castile
Cologne
England
Flanders
France
Rome
Venice

90
37.5
80
80
58.8
75

100
100
100
100
100
100
100
100

141
431
336.3
100
128.3
250
138.2
100

225
1137.9
630
96
255.2
220
214.7
145.3

333
2586.2
915
121.3
373.3
312.5
290.2
181.3

Source: Lothian (2002 forthcoming), based on data from Spufford (1988).

25

495
6465.5
1680
146.7
609.5
387.5
382.4
193.8