See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/5040934

Lessons from the history of money

Article · February 1998

Source: RePEc

CITATIONS READS
7 9,731

1 author:

Francois Velde

74 PUBLICATIONS   1,114 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Comédie française View project

1918 pandemic View project

All content following this page was uploaded by Francois Velde on 17 February 2016.

The user has requested enhancement of the downloaded file.

Lessons from the history of money

François R. Velde

Introduction and summary in particular in providing a natural anchor for

The use of money began in the sixth century the price level. But they also have certain disad-
B.C. in what is now western Turkey, when vantages, manifested in particular in the difficulty
lumps of gold found in rivers were melted and of providing multiple denominations concur-
turned into pieces of uniform size imprinted rently. These problems arose early on, in the
with a stamp. For almost all of the time since fourteenth century, in the form of money short-
then, the common monetary system has been ages. Societies tried to overcome these disadvan-
commodity money, whereby a valuable commodity tages, and this led them progressively closer to
(typically a metal) is used as a widely accepted fiat money, not only in terms of the actual value
medium of exchange. Furthermore, the quantity of the object used as currency, but also in terms of
of money was not under anyone’s control; private the theoretical understanding of what fiat money
agents, following price incentives, took actions is and how to manage it properly.
that determined the money supply. In the process, societies came to envisage
Today, the prevalent monetary system is that the use of coins that were worth less than their
of fiat money, in which the medium of exchange market value to replace the smaller denominations
consists of unbacked government liabilities, which that were often in short supply. These coins are
are claims to nothing at all. Moreover, govern- very similar to bank notes; they are printed on
ments have usually established a monopoly on base metal, rather than paper, but the economics
the provision of fiat money, and control, or poten- behind their value is the same. What governments
tially control, its quantity. Fiat money is a very learned over time about the provision of small
recent development in monetary history; it has change is thus directly applicable to our modern
only been in use for a few decades at most. system of currency.
Why did this evolution from commodity In his A Program for Monetary Stability (1960),
money to fiat money take place? Is fiat money Milton Friedman begins with the question: Why
better suited to the modern economy or was it should government intervene in monetary and
desirable but impractical in earlier times? Were banking questions? He answers by providing a
there forces that naturally and inevitably led to quick history of money, which he describes as
the present system? a process inevitably leading to a system of fiat
Fiat money did not appear spontaneously, money monopolized by the government (p. 8):
since government plays a central role in the These, then, are the features of
management of fiat currency. How did govern- money that justify government inter-
ments learn about the possibility and desirability vention: the resource cost of a pure
of a fiat currency? Did monetary theorizing play commodity currency and hence its
any role in this evolution?
In this article, I will argue that the evolution François R. Velde is an economist at the Federal Reserve Bank
from commodity to fiat money was the result of of Chicago. This article draws on joint work with Thomas J.
a long process of evolution and learning. Com- Sargent, University of Chicago and Hoover Institution.
modity money systems have certain advantages,

2 Economic Perspectives
 tendency to become partly fiduciary; The commodity money system delivers a
the peculiar difficulty of enforcing nominal anchor for the price level. The mecha-
contracts involving promises to pay nism by which this takes place can be described
that serve as medium of exchange in the context of a profit-maximizing mint, which
and of preventing fraud in respect to was how coins were produced in the Middle
them; the technical monopoly character Ages and later.2 Suppose there is a way to convert
of a pure fiduciary currency which goods into silver and silver into goods at a con-
makes essential the setting of some stant cost (in ounces of silver per unit of goods),
external limit on its amount; and finally, which can be thought of as either the extraction
the pervasive character of money which cost of silver and the industrial uses of the metal
means that the issuance of money has or the “world price” of silver in a small country
important effect on parties other than interpretation. Silver is turned into coins by the
those directly involved and gives spe- mint; the mint (which really represents the pri-
cial importance to the preceding fea- vate sector) also decides when to melt down
tures. ... The central tasks for govern- existing coins.
ment are also clear: to set an external The government’s role is limited to two
limit to the amount of money and to pre- actions. It specifies how much silver goes into
vent counterfeiting, broadly conceived. a coin, and it collects a seigniorage tax3 on all
new minting.
This article will find much to validate this When the mint is minting new coins, its costs
view. It turns out that the problem of counter- are the cost of the silver content, the seigniorage
feiting, identified as central by Friedman, pro- tax, and the production cost;4 its revenues are the
vided obstacles that were overcome only when market value of the coins, which is the inverse of
the appropriate technology became available. the price level. Similarly, when the mint is melt-
As technology changed and offered the possibility ing down coins, its costs are the market value of
of implementing a form of fiduciary currency, the coins, and its revenues are the value of the silver
various incomplete forms of currency systems contained in them.
were tried, with significant effects on the price Whether the mint will produce new coins
level. These experiments led to the recognition or melt down existing coins will thus depend
that quantity limitation was crucial to maintaining on how the price level relates to the parameters:
the value of the currency. The need for a govern- silver content of the coins, production costs, and
ment monopoly, however, does not emerge from seigniorage rate. The price level cannot be too
our reading of the historical record, and we will low (or the purchasing power of the coins too
see that the private sector also came up with its high) or the mint could make unbounded profits
own solutions to the problem of small change, by minting new coins and spending them. Simi-
thereby presenting alternatives to the monetary larly, the price level cannot be too high (or the
arrangements we have adopted. 1 purchasing power of the coins too low), or the
mint would make profits by melting down the
Commodity money and price stability coins. The absence of arbitrage for the mint places
Among the desirable features of a monetary restrictions on the price level, which is contained
system, price stability has long been a priority, as in an interval determined by the minting point
far back as Aristotle’s discussion of money in Ethics. and the melting point (figure 1).
In the words of the seventeenth century Italian This system, which prevailed until the late
monetary theorist Gasparo Antonio Tesauro (1609), nineteenth century, has some noteworthy fea-
money must be “the measure of all things” (rerum tures. The quantity of money is not controlled
omnium mensura) (p. 633). Aristotle also noted that directly by the government; rather, additions to
commodity money, specifically money made of or subtractions from the money stock are made
precious metals, was well suited to reach that goal: by the private sector, on the basis of incentives
“Money, it is true, is liable to the same fluctuation given by the price level. The incentives operate
of demand as other commodities, for its purchas- so as to make the system self-regulating. If coins
ing power varies at different times; but it tends become too scarce, their value increases and the
to be comparatively constant” (Aristotle, Ethics, price level falls until it reaches the minting point,
1943 translation). when more coins are added to the stock. If coins

Federal Reserve Bank of Chicago 3

 FIGURE 1 Although governments considered minting
a fiscal prerogative, they were constrained in their
Constraints on the price level caused choice of the seigniorage rate. High rates, a form
by arbitrage
of monopoly rent, were possible only if the gov-
Minting Melting
ernment could effectively prevent competition.
point point But in medieval Europe, all manner of coins circu-
lated in all places and individuals were quite
Price level willing to take their metal to the mint of a nearby
lord or king, subject to transportation costs, if
they found the local seigniorage rate too high.
Also, the technology for making coins was rather
become too numerous, on the other hand, their
crude and available to any jeweler or goldsmith,
market value reaches their intrinsic value and it
so that counterfeiters would also be tempted by
becomes worthwhile for the mint to melt them
high seigniorage rates. In practice, then, the width
down. The commodity nature of the currency
of the interval was rather small, and production
places bounds on the price level, but does not
costs with seigniorage were on the order of 1
determine the price level within that interval.
percent to 2 percent for gold and 5 percent to 10
Within the interval, the price level depends
percent for silver (the latter being ten times less
on how the quantity of money relates to the vol-
valuable, transport costs were higher).
ume of transactions, according to Irving Fisher’s
famous quantity theory equation.5 As long as the Multiple denominations and token coinage
price level is inside the interval, the stock of coins,
This simple commodity system lacks one
or quantity of money, is fixed. Variations in the
feature: multiple denominations. Although it is
volume of transactions or in income would shift
always possible to express any price in pennies,
the price level up or down, unless such variations
in practice it is necessary to have a range of coins
were so severe as to push the price level up to
of various denominations.6
the melting point or down to the minting point.
In its last incarnation (the so-called classical
In that case, the mint would enter into action
gold standard), the commodity money system
and modify the quantity of money in the appro-
handled multiple denominations in a straight-
priate way.
forward way, which is described in textbooks,
Consider now the interval in figure l. Its
for example, John Stuart Mill (1857).
position on the real line is determined by the
world price of silver and the silver content of a The standard formula
dollar coin. Any reduction in the number of The method that Cipolla (1956) calls the stan-
ounces of silver per dollar, that is, any debase- dard formula, consists of choosing a principal
ment of the currency, shifts the interval to the (large) denomination, which continues to be pro-
right; the price level is therefore higher. But the vided as before at the initiative of the private sec-
width of the interval is determined by produc- tor, thus continuing to provide a nominal anchor
tion costs and the seigniorage tax. We may take for the price level. The provision of lower or sub-
production costs as a technological given, but sidiary denominations relies on three key elements:
the seigniorage tax is chosen by the government. 1) monopolization of coinage by the government,
In principle, the government could make the tax 2) issue of token coins, and 3) peg of the token
a subsidy; it could even subsidize the production coins by having the government convert them
costs completely. In that case, the interval in fig- on demand into the larger denominations. The
ure 1 would be reduced to a point, the minting intrinsic content of token coins was somewhat
point and melting point would coincide, and the or much smaller than the face value at which
price level would be completely tied to the world they circulated. Some authors call such coins
price of silver. This would eliminate any fluctua- partly fiduciary. The opposite of a token coin is
tions in the price level due to the quantity theo- a full-bodied coin.7
retic effects described above. The only variations In the case of the gold standard, the larger
would be due to fluctuations in the world price denominations were gold coins, and currencies
of silver. In western European practice, howev- (the U.S. dollar, the British pound, and the French
er, the seigniorage rate was positive in almost franc) were defined by the number of ounces of
all countries. gold per currency unit. The subsidiary coinage

4 Economic Perspectives
consisted of silver and bronze coins, which were of counterfeiting.10 While nowadays counterfeiting
token. The government’s willingness to peg, say, may seem to be a significant but not overwhelm-
the silver quarter at 1/40 of a gold eagle was ing nuisance, which suitable technology can
implemented by the U.S. Treasury.8 always remedy (such as that embodied in the
Thus, in the standard formula, tokens play recently issued $100 and $50 bills), in the past it
the same role as convertible notes issued by the presented an insuperable obstacle to the develop-
central bank. As with notes, a mechanism serves ment of the standard formula.
to regulate the quantity outstanding: Excess quan- One way to prevent counterfeiting is to impose
tities of token quarters are turned in at the treasury high costs of entry to counterfeiters. Law enforce-
in exchange for gold eagles, while needed tokens ment provides a second method; as the Italian
are sold by the mint.9 economist Montanari wrote in 1683, “A die which
The advantages of a token coinage are the costs the prince 3 to make, will cost a counterfeiter
same as the advantages of a representative money 8 or 12; because he who works at the mint does
system, as pointed out by a long line of writers, not risk his life, and receives only the wage com-
including Adam Smith, John M. Keynes, and mensurate to his activity; but if a goldsmith has
Milton Friedman. Resources that had been spent to make a coin at the risk of his whole being, he
forming and maintaining that part of the stock of will not be persuaded if not with a lot of gold.”
metallic currency were freed up for other purposes. The death penalty11 for counterfeiters adds a risk
To quote the French monetary official Henri premium to the counterfeiters’ wage costs, which
Poullain, writing in 1612: “In a card game, where may or may not be sufficient to wipe out their
various individuals play, one avails oneself of potential profits. A third method is to make the
tokens, to which a certain value is assigned, and government currency difficult to imitate, for ex-
they are used by the winners to receive, and by ample, if it is produced with a technology that is
the losers to pay what they owe. Whether instead not accessible to the private sector in some way;
of coins one were to use dried beans and give either the government can make better coins or
them the same value, the game would be no less the same coins more cheaply.
enjoyable or perfect” (Poullain, 1709, p. 68). If such a cost or technology advantage is not
Another advantage, from the point of view available to the government, then attempts at
of the government, is that the issue of tokens is issuing token coinage will be plagued by coun-
quite profitable. To the extent that tokens circulate terfeiting or competition from neighboring curren-
for more than their intrinsic value plus the costs cies. Ultimately, the gross seigniorage rate will
of minting, they represent a pure profit, the be driven down to the production costs (common
seigniorage in the medieval and modern sense to both government and counterfeiters). Thus,
of the word. without the appropriate technology, only full-
These two advantages (social savings and bodied coins can be used for small denominations.
government revenues) have been understood
for centuries, and, as Friedman points out, have The big problem of small change
provided impetus for the development of money This seemingly trifling aspect of the monetary
away from a strict, full-bodied commodity version. system turns out to have bedeviled Western soci-
However, these two motivations do not determine eties for centuries. Nowadays, the only problem
clearly in which direction money will develop; most people see with small change is that we
perhaps, in fact, each pushes in a different direc- have too many pennies around, but for students
tion. The tension will be illustrated in the historical of monetary history, the “big problem of small
process I describe. change” (a phrase coined by Carlo Cipolla in 1956)
refers to recurrent coin shortages that were preva-
Prerequisites of a token coinage lent before the adoption of token coinage. The last
Whatever its advantages, the implementation time the U.S. experienced a shortage of small
of the standard formula depended on some pre- change was in 1965–66, when quarters and dimes
requisites. With a token coinage, the profits to the still contained silver; the Coinage Act of 1965
issuer are large, and, as Friedman says (1960, p. 6), made them completely token (Spengler, 1966).
“In fraud as in other activities, opportunities for
profit are not likely to go unexploited.” The gov- Full-bodied small change
ernment’s ability to maintain its monopoly on The medieval technology for making coins
token issue is thus dependent on the prevention was very simple. Metal was melted and beaten

Federal Reserve Bank of Chicago 5

into sheets, the sheets were cut with shears into FIGURE 2
blanks, and the blanks were placed between two
hand-held dies. The upper die was struck with a Constraints on the price level
of two coins
hammer and the blank imprinted. Dies were made
by goldsmiths using ordinary tools, and the design Minting Melting
on coins could easily be copied by any goldsmith. point point
Dollars
Thus, the government and the private sector had
access to the same technology.
Around 800 A.D., Charlemagne unified most Minting Melting
of Western Europe and created a uniform currency. point point

Until the twelfth century, Europe only had one Pennies

Price level
coin, the silver penny, initially minted identical-
ly across Charlemagne’s empire. Thus the com-
modity money system was in its simple, one-coin
form. Around the year 1200, large improvements of the intervals do not coincide, one type of coin
in the European economy, improved safety, and is never minted. Equating the lower ends of
economic expansion led to greater volumes of the intervals (by the government’s choice of the
trade and the need for larger denominations than intrinsic contents and the seigniorage rates)
the penny. This led to the appearance of silver makes the mint stand ready to buy silver for the
coins of about five to ten times the content of a same price, whether it pays in pennies or dollars.
penny, called grossi. Over time, the denomina- On the other hand, if the upper ends of the
tion structure became richer, with the addition intervals do not coincide, one coin might be
of gold coins in the mid-thirteenth century. melted, but the price level could still rise further
Coins throughout the denomination structure and the other coin remain in use. Equating the up-
remained close to full-bodied. per ends of the intervals makes the ratio of metal
However, the commodity money system contents in the two coins equal the exchange rate,
acquires unexpected complications when multiple in which case pennies are strictly full-bodied.
denominations are introduced. To see this, let If the melting point for pennies is higher than
us return to the mint’s problem, and suppose the melting point for dollars, pennies are rela-
we have two currencies, dollars and pennies. tively light.
The same reasoning as before will apply to both Thus, if pennies are not full-bodied, a suffi-
coins separately. As a result, the requirement cient rise in the price level will make large coins
that there be no arbitrage left for the mint will disappear. If the mint prices differ, a sufficient
now place two sets of restrictions on the price fall in prices will prompt minting of only one of
level, which we can represent by two intervals, the two coins. The perpetual coexistence of both
as in figure 2. coins in the face of price fluctuations requires
In order to make the two intervals compa- that pennies be full-bodied and that equal mint
rable, the lower one (which corresponds to pen- prices prevail for both coins; that is, the intervals
nies) is scaled by the market exchange rate between must coincide and the sum of the seigniorage
the two coins (expressed in dollars per pennies). rates and the production cost must be equal for
This simply means that the mint’s calculations the two coins.
about minting or melting pennies are computed The state of the technology creates yet anoth-
in dollars. er difficulty. We have seen that government had
The intervals must overlap, of course. Recall little freedom to choose the seigniorage rate: It had
that the position of a coin’s interval on the real to be positive and could not be large. But making
line is linked to the intrinsic content of that coin, small coins was much more expensive than mak-
so that a smaller intrinsic content of the dollar ing large coins, because making a small coin or a
corresponds to a higher price level. With two large coin involves essentially the same process,
coins, the ratio of intrinsic contents must be rea- independent of the size or content of the coin.
sonably close to the intended parity between In the extreme, if it costs the same to make a pen-
denominations, although it need not coincide ny or a dollar, then the production costs for 100
with that parity. But that is not enough: A coin pennies is 100 times the cost per dollar for the
is produced only when the price level reaches same value of output (the coins). Historical data
the minting point. Therefore, if the lower ends shows that the cost of making a coin fell with the

6 Economic Perspectives
denomination, but not fast enough. Figure 3 plots the total money stock M = M1+ eM2 (where M is
the production costs as a function of coin size for the total stock in dollars, M1 is the number of
various European countries. dollar coins, e is the market exchange rate, and
This technological constraint presented the M2 is the number of pennies). As long as there
mints with a dilemma: provide only full-bodied are enough pennies to carry out small transac-
coins and see pennies never minted, or offer the tions (not just in the physical sense M2 but in
same price for bullion in pennies or in dollars and terms of their total value eM2), there can be more
face the risk of seeing the price level increase and or fewer pennies or they can be worth more or
large coins disappear. Thus, the commodity less. If, for some reason, the relative share of
money system with full-bodied denominations small transactions changes and more pennies
has the potential for either shortages or gluts of are needed, more pennies will be provided only
small change. if the minting points are lined up correctly and
In fact, shortages of small change were a the price level falls enough. But for the general
common complaint, running through centuries price level to fall, the shock must affect the volume
of monetary history all over Europe and also of all transactions, and it is not hard to imagine
(in the early nineteenth century) in the U.S. The situations where the existing stock of pennies is
above argument, although limited to the supply insufficient, yet no new pennies are minted.
side, shows how vulnerable the commodity These shortages of small change have a curi-
money system was to such shortages, given the ous feature: In a decentralized economy, agents
technology available. An analysis of the demand choose how many pennies to hold. In order for
side reveals even more trouble. them to hold too few pennies, there needs to be
If we think of pennies and dollars as required a price incentive for them to economize on pen-
for consumption purchases (a feature called a nies. This occurs through a rate of return domi-
cash-in-advance constraint), but we assume that nance, that is, the return on holding pennies is
large coins cannot be used in small transactions, lower than the return on holding dollars. In other
whereas small coins can be used in large trans- words, the market exchange rate, e (in dollars
actions, it emerges that, within the overlapping per pennies), falls. But this means that the share
intervals of figure 2, there is a certain indetermi- of pennies in the total stock of coins shrinks fur-
nacy of the exchange rate between dollars and ther, accentuating the shortage of small change.
pennies or the ratio at which pennies enter into Furthermore, a fall in the exchange rate shifts
the lower interval of figure 2 to the left,
making it likelier that the price level
FIGURE 3 will hit the upper bound of the interval
Production costs of coins in late medieval for pennies, the melting point.
Europe, 1350–1500 Thus, shortages of small change
costs/content (percent of face value)
push the economy in a vicious cycle,
50 by making the shortage even more
severe through a depreciation of the
smaller denominations, and ultimately
40
bringing about a melting down of
pennies, once they have depreciated
30 to the value of their intrinsic content.
Within the confines of the available
20 technology, one partial remedy is for
the government to counteract the left-
ward shift of the interval due to the fall
10
in e by reducing the intrinsic content of
pennies, which shifts the interval to the
0 right. Figure 4 plots the evolution of the
10 1 10 2 10 3 10 4 10 5
mg silver
mint equivalent (the inverse of the intrin-
Note: Production costs were calculated by the percent of the face value
sic content) for two medieval Florentine
of the coin as a function of the coin size in milligrams of silver. silver coins, the picciolo (a penny, or 1d)
Source: T. Sargent and F. Velde, 1997, “The evolution of small change,”
Federal Reserve Bank of Chicago, working paper, No. WP-97-13. and the grosso (worth 4d), during the

Federal Reserve Bank of Chicago 7

 FIGURE 4 The evolution of monetary doctrine
The intrinsic content of two Florentine coins, These shortcomings of the commodity
1250–1530 money system were a result of the state
lire/Florentine pound of minting technology until 1550 or so.
Moving toward the standard formula,
or toward fiduciary coinage, required a
Picciolo (4d) better technology. However, the tech-
10 2
Index of the price nology would have gone unexploited
in pennies of had monetary doctrine not weakened its
gold florin
attachment to the concept of full-bodied
Grosso
(60-80d) coinage. This evolution of monetary
doctrine can be traced in the writings of
medieval jurists.12 This doctrine arose
from their efforts to understand observed
price patterns and devise ways to deal
101
with the legal consequences for private
contracts (the problem of the standard
1250 1300 1350 1400 1450 1500 1550
Notes: The data have been plotted inversely. The coins, picciolo and grosso,
of deferred payments).
are measured in Italian denaro (d), the equivalent of a penny. Because medieval Europe had begun
Source: T. Sargent and F. Velde, 1997, “The big problem of small change,”
Federal Reserve Bank of Chicago, working paper, No. WP-97-8. with the penny and later added larger
coins, the tradition was that prices were
denominated in pennies, dozens of pen-
nies (shillings), and scores of shillings
Middle Ages (the gold florin’s intrinsic content
(pounds).13 Many nominal debts and contracts
remained constant). A pattern of recurrent debase-
were thus expressed in pounds of the small coin,
ments is apparent. The graph also displays the
whose constant debasement led to the long-term
price of the gold florin in terms of silver pen-
inflation that is apparent in figure 4.
nies. This corresponds to the exchange rate of
When the penny was the only coin, monetary
pennies per dollar or 1/e (the florin ranged from
doctrine was straightforward. In modern terms,
240d in 1250 to 1,680d in 1530). One way to inter-
it applied standard price theory to money, treat-
pret this graph is that the periodic debasements,
ing it as a commodity like any other. When a loan
evident as upward steps, occurred to remedy
of 100 pennies came due, 100 pennies were owed,
the upward drift in the price of the florin, as our
irrespective of any fluctuations in the purchasing
model predicts.
power of pennies. The Neapolitan jurist Andrea
This version of the model takes the price of
d’Isernia (1220–1316) wrote: “If I lend you a mea-
silver in terms of real resources as constant. In
sure of wheat in May when it is expensive and
fact, this cost could be taken as variable over time,
is worth perhaps 3 tarini, and I reclaim it in July
embodying a variety of shocks (changes in the
after the harvest when it is worth perhaps 1 tarino,
technology to mine silver, including new discov-
it is enough to return the measure of the same
eries, and changes in the demand for silver in
wheat in kind, even though it is worth less; like-
industrial activities). Furthermore, the model
wise if it is worth more, for example if I lent it in
assumes that large and small coins are made of
July and demanded it in the following May ... the
the same metal; but small and medium coins being
same reasoning applies for money as it does for
made of silver and large coins being made of gold,
wheat and wine” (d’Isernia, 1541).
the intervals of figure 2 shift around due to changes
From Charlemagne’s reform around 800 A.D.
in the relative price of gold and silver. Depending
(which restored a uniform currency in Western
on the width of the intervals, small shocks might
Europe) to the twelfth century, the penny changed
be accommodated, but larger variations lead to the
content at various rates, through the action of wear
same problems outlined above, unless e is allowed
and tear and debasements. Such changes in the
to change. The difficulties in providing multiple
intrinsic content of a penny were also treated by
denominations render bimetallism (the simulta-
jurists in a similar way. The jurist Azo (d. 1220)
neous use of two metals in legal tender curren-
formulated a simple rule: “The same money or
cies with a fixed exchange rate) a fragile system.
measure is owed that existed at the time of the
contract” (in Stampe, 1928, p. 36).

8 Economic Perspectives
 With the appearance of larger denomination with leather money that he redeemed into gold
coins and the existence of time-varying rates of after the successful conclusion of the siege. D’Isernia
exchange between denominations, the legal argued that, under the specific circumstances al-
problems grew more challenging, and jurists ready identified by the current doctrine, money
began to diverge in their answers. A distinction could be made of worthless material, like lead or
was made between the “intrinsic quality” of a leather, as long as it was redeemed after the end
coin (its metal content) and the “extrinsic quality,” of the emergency into good money. This was the
taken to mean either its purchasing power (the basis for the concept of deficit financing, which
inverse of the price level) or its rate of exchange would play an important role in the development
with other coins. The general consensus prevail- of fiat money. By the late sixteenth century, these
ing in the fourteenth and fifteenth centuries called notions were commonly held. The widely cited
for adjusting debt repayments for variations in René Budel (1591) held it “to be indubitable that
the intrinsic quality, but ignoring variations in a Prince in the midst of costly wars, and therefore
extrinsic quality; and small coins were considered in great necessity, can order that money be made
legal tender to the degree that they were full-bodied out of leather, bark, salt, or any material he wants,
and interchangeable with large coins. if he is careful to repair the loss inflicted thereby
However, jurists also observed the existence on the community with good and better money”
of positive seigniorage rates (the width of the (Budel, 1591, chapter 1, paragraph 31).
interval in figure 1), and realized that money’s In other words, the intrinsic content could
purchasing power could be greater than its intrinsic be set to 0, as long as some measure of convert-
value. In other words, they discovered that the ibility, either immediate or in the near future,
price level could move above the minting point. was implied. In 1481, a small town in Catalonia
One strand of the legal literature insisted that carried out an experiment to solve its problem
seigniorage should be set close or equal to 0. of small change: it was authorized by the king
Others, who argued that precious metals as bullion of Aragon to issue pure copper coins, 14 whose
and in the form of coins should afford the same intrinsic value was about 25 percent of their face
utility, recommended that the state subsidize the value, as long as “the city be known to pledge,
mint completely (in particular, the jurist Bartolo and effectively pledge to receive said small money
da Sassoferrato, 1313–1357). As jurists, they tried from those who might hold it, and to convert it
to define rules for repayment of monetary debts. and return for it good money of gold or silver,
They correctly perceived that their proposal would whenever and however much they be asked”
eliminate some fluctuations in the standard of value. (in Botet y Sisó, 1911, p. 328). This experiment
In practice, the jurists realized that govern- was imitated by a number of other Catalonian
ments were unwilling to subsidize mints and cities, although they were plagued by counter-
were tempted to increase seigniorage revenues feiting, which the state of technology made rela-
as much as they could. A small tax was considered tively easy.
acceptable and a larger tax under very specific
circumstances, such as a fiscal emergency (paying Technological change and
for a sudden war or the king’s ransom). Some policy experiments
even argued that, in the words of Gabriel Biel These developments in monetary doctrine,
(d. 1495) a large seigniorage rate “is the easier and the early Catalonia experiment, show that
way to collect quickly the required funds with- technology remained the real barrier to the imple-
out fraud and undue exactions from the subjects. mentation of a standard formula for small change.
It is, moreover, felt less and for this reason more The technology did change, in two major waves;
easily borne without protest and without the and each wave opened up new possibilities that
danger of a rebellion on the part of the people. governments exploited.
It is the most general form of taxation embracing Recall that the standard formula incorporates
all classes, clergy, laity, nobility, plebeians, rich several ingredients: monopolization of coinage,
and poor alike” (Biel, 1930 translation, p. 35). issue of tokens, and convertibility of the tokens.
Some jurists like d’Isernia even went further. The ingredients are logically distinct. The period
D’Isernia probably observed episodes such as between the first and the second wave of techno-
the siege of Faenza in 1241, when the Emperor logical change (1550 to 1800) saw a wide variety
Frederic II ran out of money and paid his troops of experiments, in which some but not always

Federal Reserve Bank of Chicago 9

all ingredients were proposed or implemented. Until then, copper, silver, and minting costs each
The variety of outcomes offered a rich mine of represented a third of the face value of billon
lessons in monetary doctrine. coinage. With Philip II’s decree, the silver was
withheld and the copper content reduced.
Mechanization and the Age of Copper Philip II had efficiency in mind. He ordered
The first major shift in minting technology that the new copper coins be issued only to retire
took place around 1550. In southern Germany, existing small denomination coins (M2) with
two processes were independently developed token coinage and that the mechanism with its
to mechanize the minting process, using ma- melting and minting points be preserved for
chines rather than tools to cut uniform blanks providing large denomination silver coins (M1).
and impress them with a design. One technology Retaining the mechanism for supplying M1
(the screw-press) proved to be better than the would keep the price level within the appropri-
other (the cylinder-press), but also more expen- ate melting and minting points so long as some
sive, and only prevailed in the late seventeenth large denomination coins continued to circulate.
century. Until then, the other proved popular in But Philip II’s successors, Philip III (1598–1621)
a number of countries, including the various and Philip IV (1621–64), saw that the cylinder
German states and Spain. press offered opportunities to enhance revenues.
The king of Spain heard about the cylinder- A first experiment in 1602, whereby the copper
press technology from his cousin the count of content of coins was reduced by 50 percent with
Tirol, who had been the first to install the new no resulting effect on the price level, convinced
machines in his state mint. The machines were the government that the intrinsic value of the coins
imported and set up in Segovia in 1582, and could be made much lower and the seigniorage
applied to the silver coinage of pieces of eight. rate much more lucrative. Another experiment,
The coins produced in Segovia were much more carried out in 1603, further reinforced the point
uniform and round, and more sharply imprinted, that individuals did not care about the composi-
than anything done using the old hand tools. tion of their money balances. After the 1602 reduc-
The Spanish government soon realized the poten- tion, two kinds of pennies circulated, one twice
tial in this technology, and decided in 1596 to pro- as heavy as the other; it was decided that all old
duce all small denominations in pure copper (heavy) pennies were to be brought to the mint,
with the new machines. King Philip II explained stamped with a “2" and one two-cent coin returned
his reasons in an edict: for every two old pennies presented. The opera-
We have been advised by people tion was successful and all old pennies were pre-
of great experience, that the silver sented, affording the government 50 percent
which is put in those billon coins15 is seigniorage on the stock of pennies.
lost forever and no profit can be drawn From that point on, the Castilian govern-
from it, except in their use as money, ment knew no restraint, and enormous quanti-
and that the quantity of silver which ties of vellón (as these copper pennies were called)
is put to that use for the necessities of were minted and used to finance government
ordinary trade and commerce in this consumption. Figure 5 shows the path of nomi-
kingdom is large. We have also been nal and real balances of vellón in that period;
advised that, since we have established note that the total money stock before 1600 was
a new machine in the city of Segovia around 20 to 30 million ducats.
to mint coins, if we could mint the Recall that we express the total quantity of
billon coinage in it, we would have money as M = M1 + eM2, where M1 represents
the assurance that it could not be the stock of large denomination (silver) coins,
counterfeited, because only a small and M2 represents small denomination (copper)
quantity could be imitated and not coins. The exchange rate between the two types
without great cost if not by the use of of coins is e, and M2 is expressed in dollars. The
a similar engine, of which there are policy followed by the Castilian government
none other in this kingdom or the consisted in increasing M2 to the point at which
neighboring ones. And it would thus it completely replaced M1, all the while with no
be possible to avoid adding the silver inflation (real and nominal balances coincide).
(in Rivero, 1919, p. 150). In terms of the total money stock, M1 + eM2, a

10 Economic Perspectives
 FIGURE 5 the manipulations, which were less suc-
cessful as balances of vellón fell over time.
Nominal and real values of vellón, The Spanish experience unleashed
1595–1680
unprecedented “man-made” inflation,
million ducats
40 which made the Price Revolution of the
sixteenth century (price level increases
due to the inflow of American gold and
30 silver) look tame. It was among the first
large-scale experiments in inconvertible
Nominal value
fiat currency (although the coins were
20 accepted at face value in payment of taxes).
It demonstrated the ease with which
token coinage could overtake the money
10 stock, the workings of the quantity theory,
Real value the need for the issuer of inconvertible
(in silver equivalent) token coinage to restrain issues, and the
0 strength of the temptation created by high
1600 1620 1640 1660 1680
seigniorage rates for a government un-
Source: F. Velde and W. Weber, 1997, “Fiat money inflation in 17th century
Castile,” Federal Reserve Bank of Chicago, manuscript. willing or unable to raise other taxes.
The Spanish experiment was not the
only one at the time. During the Thirty
Years War, which started in 1618, many German
progressive displacement of M1 by vellón is con-
states concurrently debased their small denomi-
sistent with no change in e, and, other things
nations (all the while maintaining silver coinage
being equal, an unchanged money stock will
intact) and issued large amounts of copper coin-
correspond to a constant price level. However,
age to raise revenues through seigniorage. The
once M1 has disappeared, the money stock con-
results are shown in figure 6, which tracks the
sists only of copper coins M2, and all further
exchange rate between large denomination
increases in M2 result in increases in the price
coins and small denomination coins and makes
level, as is apparent in figure 5. Once the figure
it clear why the Germans called this die große
of about 20 million ducats was reached, nominal
Inflation (the great inflation), at least until a simi-
and real balances diverged, and inflation set in
lar experiment exactly 300 years later (the famous
with a vengeance. The disappearance of
silver released the price level from the
constraints imposed by the melting/mint- FIGURE 6
ing points for the dollar interval, and un-
Prices of the gold florin and silver thaler,
leashed the quantity theory with copper in Bavarian kreutzers
as the determinant of the price level.
index, 1608=1
The only way to return the price level 8
to its bounds was to engineer a reappear- Thaler
ance of the silver coins, either by decreasing Florin
M2 or by decreasing e. The Castilian gov- 6
ernment toyed with the idea of decreasing
M2 by an open-market operation (selling
bonds to buy back the copper coinage), 4
but in the end decided to halve e over-
night, in 1628.
The rest of the movements in vellón 2
balances are due to repetitions of the ear-
lier operations of vellón issue, restamp-
ing (multiplying the face value of existing 0
1610 1615 1620 1625
coins by N and extracting a seigniorage of
Source: H. Altmann, 1976, Die Kipper- und Wipperinflation in Bayern, München:
(N–1)/N) and overnight devaluations. As Neue Schriftenreihe des Stadtarchivs München, pp. 272–273.
figure 5 shows, Castilians grew weary of

Federal Reserve Bank of Chicago 11

German hyperinflation of 1922–23 under the Weimar operations as immoral, because in his view the
Republic). Poland and Russia also underwent king has no right to tax his subjects without their
copper inflations in the 1650s, as did the Ottoman explicit consent. Mariana noted that such high
empire in the 1690s. This is why the seventeenth tax rates would never be tolerated on any other
century has earned the name the Age of Copper. tax base. The worst consequence he predicted
was general hatred of the government. Quoting
Lessons from the Castilian inflation Tacitus, he recalled that “everyone claims pros-
The lessons were not lost on contemporary perity for himself, but adversity is blamed on
observers. The Spanish episode was discussed the leader” (1994, p. 104).
not only by writers in Spain, but also in Italy, The Frenchman Poullain, quoted earlier,
France, and elsewhere, leading to a consensus concluded that token coins could replace other
on quantity limitations and limited legal tender coins for domestic transactions and that this was
for small coins. precisely why their quantity should be limited.
One of the more famous commentators was Poullain, as a monetary official, successfully
the Jesuit Juan de Mariana (1536–1624), who wrote fought back various plans to issue copper on
a treatise on the vellón coinage between 1603 a large scale. Only twice, in 1640 and 1653, did
and 1606, as the experiment was beginning and France come close to embarking on a Spanish-
inflation had not yet taken off. He lays out argu- style inflation, in both cases at times of fiscal
ments pro and con, and thus provides a window emergency.
on the debates among policymakers around the The Italian Montanari, also quoted above,
Spanish king. wrote: “It is clear enough that it is not necessary
The advantages vaunted by proponents of for a prince to strike petty coins having metallic
the copper coinage are not limited to the social content equal to their face value, provided he
savings mentioned by Philip II in his edict. Pro- does not strike more of them than is sufficient
ponents claimed that without a stock of silver for the use of his people, sooner striking too few
coins as a potential reserve to settle trade deficits, than striking too many. If the prince strikes only
Spain would be forced to maintain surpluses as many as the people need, he may strike of
and resort to import substitution, thereby stimu- whatever metallic content he wishes” (Montanari,
lating Spanish industry; they also claimed that 1804, p. 109). Various other writers stressed quan-
the copper money was lighter and easier to tity limitations, as well as limited legal tender for
transport, and that its cheap provision would small coins. The latter measure uncouples the
lower the rate of interest and stimulate agricul- two stocks of money in the equation M1 + eM2,
ture and industry. In other words, arguments which was critical in the Spanish experience.
were made that, beyond the social savings from
forsaking commodity money, increases in the Monopoly versus laisser-faire
quantity of money could stimulate output.16 English coins had always been made of ster-
Mariana was conscious that incentives for ling silver, and shortages of small change became
counterfeiting created by the overvaluation of particularly acute when pennies and farthings
copper coins could be resolved by the new ma- ceased to be minted altogether in the sixteenth
chines in the Segovia mills. He was doubtful of century. From that point until 1817, English policy
the arguments on balance of trade and stimulus alternated between three regimes for the supply
of the economy, which could be made to go the of small change: private monopolies of incon-
other way through an anticipated inflation effect. vertible token coinage, government monopoly
He predicted that copper coinage would drive of full-bodied coinage, and laisser-faire (that is,
out silver, lead to an increase in prices, and in- the absence of government intervention).
duce the government to set price controls that Private monopolies (1613–44) were created
would either be ignored or counterproductive, by royal charter, which granted various individuals
at which point the government would be forced in turn (usually well-connected aristocrats) the
to reduce the face value of the coins, as indeed exclusive right to issue token coinage, although
happened in 1628. Mariana saw the projected these were never made legal tender and their
sequence of inflation and deflation as disruptive quantities were limited by the terms of the charter.
to trade and contracts and, therefore, to the A government monopoly was asserted in 1672,
king’s tax revenues. He also viewed the high making private tokens illegal, and the Royal
seigniorage rates of 50 percent in the restamping mint issued copper coins, intermittently and

12 Economic Perspectives
insufficiently, until 1754. Although mechaniza- and the nineteenth century U.S. One of the ingre-
tion had been adopted in 1660, England remained dients of the standard formula is monopolization
committed to full-bodied copper coinage.17 of coinage; it is not clear, on theoretical or histori-
The laisser-faire regime (mid-sixteenth cen- cal grounds, that this ingredient is needed if the
tury to 1613, 1644 to 1672, and the late eighteenth other two (issue and convertibility of tokens) are
century) was characterized by the absence of present. Of the two advantages of the token coinage
government-issued small denominations and system, social savings and government revenues,
by the issue of tokens by private parties or local the latter clearly provides an impulse toward
governments. In the late sixteenth century, up to monopolization that the former does not.
3,000 London merchants issued tokens. In the
period from 1644 to 1672, over 12,700 different The steam engine and the gold standard
types of tokens have been catalogued, issued in The second major technological innovation
1,700 different English towns. From the 1740s following the mechanization of minting around
on, trade tokens took over when official coinage 1550 was the adaptation of the steam engine to
ceased. Some of these issues were authorized by minting. In 1787, Matthew Boulton, partner of
government. The city of Bristol sought and secured James Watt, produced trade tokens for the Angle-
permission to issue farthings in 1652, and went sey Copper Company. A few years later he was
through three different issues over the next 20 producing copper coins for private issuers across
years. The Bristol farthings, furthermore, were England and even in France. In Paris, the most
officially convertible into large denominations. popular token coins in 1790–92, issued by the firm
They are also known to have been counterfeited. of Monneron, were minted in Birmingham by
The government put an end to the laisser-faire Boulton’s steam presses. The British government
regime twice, in 1672 and by the Act of Suppression contracted with him to produce official copper
in 1817; each time, it did so immediately after coinage in 1797, then bought the technology, and
adopting a new technology. in 1817 eliminated its competitors by making
France’s experiences were somewhat parallel. private coins illegal. The new steam-driven
In the early seventeenth century, private monop- presses were used to mint the new silver coins,
olies were instituted for brief periods of time. which, under the Coinage Act of 1816, were for
France also had a brief experience with free to- the first time issued as partly fiduciary coins,
ken issue in 1790–92. The government had decided whose intrinsic value was significantly lower
in September 1790 to issue substantial amounts than their face value. It took a decade and a half
of large-denomination paper currency backed before an implicit agreement was reached between
by a land sales scheme. Soon, thousands of private the Bank of England and the Treasury for the
and municipal banks emerged to intermediate convertibility of the silver coinage into gold
the government’s notes with their own small upon demand. By 1830, the standard formula
denomination notes and, in some cases, coins. had been fully implemented.
Initially, the government abstained from regulat- England’s implementation of the standard
ing the industry, which operated on fractional formula in 1816 applied both to bronze or copper
or 100 percent reserves, depending on the insti- coinage and to silver coins, leaving gold as the
tution. But soon the government moved to elim- single anchor for the price level (the gold stan-
inate its competitors in the business of issuing dard) and officially abandoning bimetallism.
currency. The government decided to issue It took other countries some time to follow suit:
medium-sized notes (equivalent to silver coins) Germany in 1871, France and the Latin Monetary
in June 1791, followed in December 1791 by small- Union (Belgium, Switzerland, Spain, and Italy)
denomination notes. Technical difficulties post- in 1873, the Netherlands in 1875, and the U.S.
poned the first issue of small notes to August 1792. in 1873–79 (the so-called Crime of 1873). Recently,
The government could now impose a monopoly. researchers (Friedman 1990 and Flandreau
Within a few weeks, all private banks were for- 1997) have argued that this abandonment was
bidden to issue their notes and private coins a mistake, and bimetallism was better suited to
were outlawed, amid unproven allegations of stabilizing the price level than the gold standard.
wildcatting and fraud. Nevertheless, there was no substantial differ-
These episodes present parallels with the ence between applying the standard formula to sil-
Free Banking Eras of eighteenth century Scotland ver and applying it to copper coinage, and the

Federal Reserve Bank of Chicago 13

 FIGURE 7
coinage. Monopolization is less obvi-
ous an outcome, especially given the
Market and intrinsic values of a vellón cuarto coin, prolonged Free Token Eras of England.
1597–1659 Friedman argues that government
silver maravedis needs a monopoly on fiduciary cur-
10 rency because free entry into the
issue of irredeemable paper would
8 drive down currency to its intrinsic
Market
value (namely, 0). As figure 7 shows,
this is what happened in seventeenth
6
century Spain, as the market value of
copper coinage was driven down to
4 its intrinsic value. Arguably, counter-
feiting was widespread, but judging
by figure 5, government issues are
2
Intrinsic enough to account for the phenome-
non. Surely, experiences with fiat
0 money in the twentieth century (a
1600 1610 1620 1630 1640 1650 1660
century replete with hyperinflation)
Source: F. Velde and W. Weber, 1997, “Fiat money inflation in 17th century Castile,” show that governments can drive the
Federal Reserve Bank of Chicago, manuscript.
value of a paper currency they mo-
nopolize to its intrinsic value with
great efficacy.
forces identified by Friedman (1960) and those Perhaps it is not surprising that seventeenth
leading to coin shortages seemed to lead to the century Spain was under an autocratic regime,
outcome effectively adopted by most countries. as was contemporary France (which came close
to the same outcome). England, where counter-
Conclusion
weights to the executive were at least apparent
The questions raised by Friedman (1960) at the time and constitutionally set in 1688, main-
about the necessary ingredients for an efficient tained a different policy. Nor perhaps is it a
and well-managed currency are old questions surprise that the standard formula was first im-
indeed. The big problem of small change led plemented in Britain, the most advanced democracy
monetary thinking on the path to fiduciary cur- in Europe at the time. The policy was implemented
rency, at least in the form of intrinsically trifling in 1816, just as Britain was emerging from a suc-
but convertible tokens; policy followed only af- cessful use of inconvertible paper money to finance
ter the right technology became available. As 20 years’ worth of wartime expenditures (in
technology changed and experience accumulated, contrast to France’s similar attempt in 1790–97,
various elements of the standard formula were which proved less durable). Irredeemable currency
tried separately, including irredeemable copper for deficit financing was already a centuries-old
money. The resulting inflation led to the recogni- idea; the Catalonian town of Gerona used a coin
tion that a form of quantity theory was at play, issued as siege money to start a convertible-token
and led governments to formulate various ways system in 1481. Success with deficit financing was
of limiting the quantity—through convertibility probably a good predictor of success with subsid-
and through monopoly. iary coinage; both may have something to do with
Of the main ingredients of the standard for- the degree of accountability of policymakers.
mula, the historical trend points clearly to token

14 Economic Perspectives
NOTES
1
Much of the material presented here derives from the work in 10
Note that the ability to maintain a monopoly on full-bodied
Sargent and Velde (1997a, 1997b). coinage is dependent on the same.

2
The model sketched here is developed fully in Sargent and 11
The punishment for counterfeiters was particularly severe. In
Velde (1997a). medieval France, they were boiled alive (not poached). A doc-
ument from 1311 details the costs of executing two counterfeit-
3
Seigniorage is literally the lord’s right to collect a tax, and is ers, including the price of a large cauldron and the cost of
derived from the French term for lord,seigneur. adding iron bars to the cauldron, a detail that suggests a rather
long process (Saulcy, 1879–92, Vol. 1, p. 180).
4
This cost is exclusive of the coin’s content. It represents the
costs of transforming metal into coins, and is to some degree in- 12
An anthology of their writings is in Velde (1997).
dependent of the content.
13
This did not preclude the denomination of many prices in
5
The equation is pY = vM, where p is the price level,Y is in- terms of the gold coin and fictitious subdivisions thereof.
come or the volume of transactions, v is velocity, and M is the
quantity of money. 14
Interestingly, the coins were modeled on a currency issued
some years earlier as emergency money during a siege and lat-
6
The optimal denomination structure is an unstudied problem; er left in circulation.
however, see Telser (1995).
15
A mixture of silver, to give value, and copper, to give bulk,
7
In the numismatic sense, token means something that is not of- commonly used for small denominations.
ficially money, but used as money; numismatists will speak of
full-bodied tokens. From an economic viewpoint, the distinc- 16
Another famous proponent of similar arguments was the Scot
tion between official and unofficial money is somewhat arbi- John Law (1671–1729), whose experiment in setting up a paper
trary. currency in France went spectacularly awry in 1720, during the
Mississippi Bubble.
8
Act of June 9, 1879: “Be it enacted ... that the holder of any of
the silver coins of the United States of smaller denominations 17
The proclamation of 1672 stated that small coins “cannot well
than one dollar, may, on presentation of the same in sums of be done in silver, nor safely in any other metal, unless the in-
$20, or any multiple thereof, at the office of the Treasurer or any trinsic value of the coin be equal, or near to that value for
assistant treasurer of the United States, receive therefor lawful which it is made current.” Sir Isaac Newton, master of the mint,
money of the United States” (Statutes at Large 21 [1879]: 7). wrote in 1720: “Halfpence and farthings (like other money)
should be made of a metal whose price among Merchants is
9
The status of silver dollars remained uncertain between the known, and should be coined as near as can be to that price, in-
Bland–Allison Act of 1878 and the final defeat of the pro-sil- cluding the charge of coinage. ... All which reasons incline us
ver forces after 1896. Only after 1900 did the silver dollar be- to prefer a coinage of good copper according to the intrinsic
come no different in nature from other subsidiary coins. value of the metal” (Shaw, 1896, pp. 164–165).

REFERENCES

Altmann, Hans Christian, 1976, Die Kipper- und Cipolla, Carlo M., 1956, Money, Prices and Civili-
Wipperinflation in Bayern, München: Neue zation in the Mediterranean World, Fifth to Seven-
Schriftenreihe des Stadtarchivs München. teenth Century, New York: Gordian Press.

Aristotle, 1943, Ethics, H. Racham (trans.), Cam- D’Isernia, Andrea, 1541, Commentaria in Usus Li-
bridge, MA: Harvard University Press. brum Feudorum, Lyon, France: J. Giunta.

Biel, Gabriel, 1935, Treatise on the Power and Utili- Flandreau, Marc, 1997, “The gradient of a river:
ty of Money, Robert B. Burke (trans.), Philadel- Bimetallism as an implicit fluctuation band,”
phia, PA: University of Philadelphia Press. Paris: l’Observatoire Français des Conjonctures
Economiques, manuscript.
Botet y Sisó, Joaquim, 1911, Les Monedes Catalans,
Vol. 2, Barcelona: Institut d’Estudis Catalans. Friedman, Milton, 1990, “The crime of 1873,”
Journal of Political Economy, Vol. 98, December,
Budel, René, 1591, De Monetis et Re Numaria, pp. 1159-1194.
Book 1, Cologne: Schann Nackel.

Federal Reserve Bank of Chicago 15

 , 1960, A Program for Monetary Stabili- Shaw, William Arthur, 1896, The History of Cur-
ty, New York: Fordham University Press. rency, 1252 to 1894, London: Wilsons & Milne.

Mariana, Juan de, 1994, De Monetae Mutatione, Jo- Spengler, Joseph J., 1966, “Coin shortage: Mod-
sef Falzberger (ed.), Heidelberg: Manutius Verlag. ern and premodern,” National Banking Review,
Vol. 3, pp. 201–216.
Mill, John Stuart, 1857, Principles of Political Econ-
omy, Fourth Edition, London: John W. Parker Stampe, Ernst, 1928, “Das Zahlkraftrecht der
and Son. Postglossatorenzeit,”Abhandlungen der Preußis-
chen Akademie der Wissenschaften, philosophisch-
Montanari, Geminiano, 1804, “Della moneta: historische, Klasse 1, Berlin: Verlag der
Trattato mercantile,” in Scrittori Classici Italiani di Akademie.
Economia Politica, parte antica, Vol. 3, Milan:
G.G. Destefanis. Telser, Lester G., 1995, “Optimal denominations
for coins and currency,” Economic Letters, Vol. 49,
Poullain, Henri, 1709, Traité des Monnoies, Paris: pp. 425–427.
Frédéric Léonard.
Tesauro, 1609, “Tractatus de monetarum augmen-
Rivero, Casto Maria del, 1919, El Ingenio de la to et variatione,” in De Monetarum Augmento Vari-
Moneda, Revista de Archivos, Bibliotecas y Mu- atione et Diminutione Tractatus Varii, Turin, Italy.
seos, Vol. 40.
Velde, François R., 1997, “An anthology of writ-
Sargent, Thomas J., and François R. Velde, ings on money, 12th–17th century,” Federal Reserve
1997a, “The big problem of small change,” Fed- Bank of Chicago, manuscript.
eral Reserve Bank of Chicago, working paper,
No. WP-97-8. Velde, François R., and Warren E. Weber, 1997,
“Fiat money inflation in 17th century Castile,”
, 1997b, “The evolution of small Federal Reserve Bank of Chicago, manuscript.
change,” Federal Reserve Bank of Chicago,
working paper, No. WP-97-13.

Saulcy, Ferdinand de, 1879–92, Recueil de Docu-

ments Relatifs à l'Histoire des Monnaies, 4 vols.,
Paris: Imprimerie Nationale.

16 Economic Perspectives

View publication stats