The Triumph of the Standard Formula
This chapter describes how the standard formula spread beyond Britain in the course of the nineteenth century. While Britain adopted the standard formula along with the gold standard, in 1838 the German Monetary Union implemented the standard formula but remained on a silver standard. When Germany unified in 1871, it switched to a gold standard. Meanwhile, the United States and a group of countries led by France put in place important features of the standard formula while maintaining bimetallism. All eventually followed Britain onto the gold standard.
By the early nineteenth century, shortages of small change of the type we have studied in medieval Europe were less frequent.1 However, various problems with small change persisted. They occurred in France and the United States as both countries moved from a “double standard” (bimetallism) toward the gold standard.
In this chapter we shall describe how the final triumph of the standard formula was intimately related to the “battle of the standards” and ultimately coincided with the end of bimetallism. First, however, we describe how Germany formally adopted the standard formula.
Germany’s monetary union of 1838
Given the ambiguity that long surrounded aspects of Britain’s implementation of the standared formula, Germany should be considered the first country to implement it completely. Germany did so explicitly in 1838.2 It was on the silver standard then, had no single monetary authority, but participated in a currency union. The Dresden treaty established a currency union among sovereign states. It represents the first pure implementation of the standard formula. Aside from the unquantified commitment to limit issues, the only restrictions on token coinage were limited legal tender and convertibility.
Germany’s earlier long-standing problems with gluts of small change had come from the hundred rival issuers of currency and their inability to commit to a common standard, as we saw in chapter 15. After the Congress of Vienna in 1815, the number of sovereign states in Germany had been reduced to 39. They formed a customs union in 1834. Coalescing into a currency union was helped by the fact that most German states were on one of three silver standards (the convention thaler in Bavaria and Austria, the gulden in southern and western Germany, and the Prussian thaler in northern and eastern Germany).3 Also, most German mints had adopted the steam-driven presses devised by the engineer Uhlhorn in 1817.
The first step toward monetary union was taken when the states in southern and western Germany formally defined their common standard in August 1837, the silver gulden of 60 kreutzer. The terms of the union concerning denominations below the gulden are remarkable. The union defined the weight and fineness of the 6-kreutzer and 3-kreutzer pieces to be struck by the individual states as 9.3% below full weight, and required that the issuing state redeem its small coins in gulden pieces, when presented in amounts of 100 gulden or more. Mindful of their history, members of the union established a rotating monitoring system whereby member i monitored the quality of member i + k’s coins in year k of the union. Coins of 1 kreutzer and smaller were left unregulated.
The currency union was considerably enlarged in July 1838 at Dresden. The gulden and the Prussian thaler were established as concurrent units at a fixed ratio. The terms of the Dresden union concerning small change were slightly different. Small denomination underweight coins could be minted by individual states subject to a convertibility requirement, and demonetization could only be carried out with prior notice and with redemption of outstanding coinage. The states also committed themselves not to issue more than necessary for the needs of trade, and small coins could not be legal tender for more than the denomination of the smallest full-bodied coin.
The currency union of Dresden covered 85% of the German population. By 1857 it had been extended to 95% of the population, and it even included the Austro-Hungarian empire until 1866. It remained in force until 1871, when the newly formed German empire decided to abandon the silver standard for the gold standard. The monetary law of 1873 established the same system for its subsidiary coinage (with per-capita limits and redemption). The existing large silver thalers, however, remained legal tender but were not made convertible.4
Bimetallism versus the gold standard
In countries adhering to the “double” or bimetallic standard, such as France and the United States, full adoption of the standard formula did not come until after bimetallism had collapsed. While bimetallism was not incompatible with adopting the main features of the standard formula for some denominations, the nearly simultaneous decisions of Germany, France, and the United States to leave silver in 1873 had important ramifications for how France and the United States would implement the standard formula.
Bimetallism is a monetary system in which the mint stands ready to convert either of two metals (gold and silver) into coins, and those coins are unlimited legal tender at given rates.5 The original purpose of bimetallism was to allow silver to be used for small denominations and gold for large ones, and to maintain a constant exchange rate between the two.6
Because a given weight of gold is so valuable, small denominations in a gold standard cannot be full-bodied. This fact urgently recommended the standard formula under the gold standard. By comparison, maintaining bimetallism made adopting the standard formula less pressing. Large denominations could consist of full-bodied gold coins, small denominations of full-bodied silver coins, and the exchange between them could under the proper circumstances7 be fixed by the rules of bimetallism. But in 1873, France and the United States both ended the free coinage of silver. The ensuing fall in the market value of silver promptly made their existing silver coins into overvalued tokens, for which convertibility was ultimately provided.
To understand why this happened, we briefly review two theories of bimetallism, the first of which assumes that relative prices of silver and gold are exogenous (a “small country” assumption), the second of which makes relative prices of gold and silver respond to the choice of monetary regime of a government or a coalition of governments.
Bimetallism in a small country
The model of chapter 11 can be used to study bimetallism for a small country that takes relative prices of gold and silver to be exogenous. Bimetallism adds to the chapter 11 model a government rule that makes gold and silver coins legal tender at a fixed ratio.
However, if there are frequent and substantial changes in the relative prices of the metals in terms of consumption goods (φg, φs), it is difficult to sustain a fixed exchange rate for very long in the model of chapter 11. To maintain concurrent circulation of coins of both metals in the face of such changes, the government has to adjust coin specifications (i.e., debase or reenforce coins of one of the metals) to realign the intervals.
Appreciations in φs and φg have very different ramifications for potential shortages of small change. An increase in the price of goods in terms of silver, φs, (a cheapening of silver) shifts the silver interval to the right and ultimately leads to minting of silver and melting of gold. At worst, all currency becomes silver, which leads to no shortages of the small denomination silver coins. But an increase in the gold price of goods, φg, makes silver disappear; and in our model, large gold coins cannot take the place of small silver coins.
Thus, even if a bimetallic system could potentially provide several tiers of denominations within the old medieval system of coin production, it remained vulnerable to a cheapening of gold.8 That would force the monetary authorities to contemplate a debasement of the silver coin or to think of alternate ways of providing smaller denominations.
Monetary authorities confronted just that problem after the gold discoveries in California and Australia in 1849 sparked an increase in φg. Within a few years, the United States resorted to token coinage for their smallest denominations, albeit without convertibility. Several European countries reacted similarly, until a desire to coordinate their responses led them to create the Latin Monetary Union, centered in France. But both the United States and the Latin Union maintained free coinage of silver and unlimited legal tender for their main silver coins, the 5F and the $1 pieces. They remained committed to bimetallism for parts of their denomination structures. Their attachment to bimetallism has been the subject of much debate. To understand that debate, we use a different model, which we describe next.
Worldwide bimetallism
When a large country or a large coalition of countries adopts bimetallism at the same rate, it is no longer appropriate to take the relative prices of gold and silver (φg, φs) as exogenous because they can be affected by the demands for monetary stocks of gold and silver, which in turn depend on the choice of standard. The endogeneity of relative prices means that for a large country or coalition of countries, bimetallism can be less susceptible to the problems mentioned in the previous section. This point was recognized and formalized by Walras ([1900] 1977), Fisher (1911, ch. 7), and Velde and Weber (2000).
To make gold and silver prices endogenous, Velde and Weber’s model focuses on the substitution between monetary and nonmonetary uses of gold and silver. Velde and Weber set our seigniorage parameters σi to zero for each type of coin, thereby collapsing the minting and melting intervals to zero. They impose a single cash-in-advance constraint cast in terms of the nominal value of all gold and silver coins. The cash-in-advance constraint applies to consumption of three consumption goods, one a nondurable consumption good, the remaining two being durables, namely, gold and silver jewelry, to capture the nonmonetary uses for these metals. Velde and Weber study a pure endowment economy, so that the supplies of three types of consumption good are taken to be exogenous. The consumers have preferences over the nonmonetary stocks of gold and silver (jewelry) that are represented by a concave utility function. The single cash-in-advance constraint applies to the sum of gold and silver coins, weighted by the exchange rate. This means that at any constant exchange rate between gold and silver coins, the money holder is indifferent between holding gold and silver coins.9
There are potentially six endogenous variables: stocks of gold and silver coins, stocks of gold and silver jewelry, the price level, and the exchange rate. Alternatively, the “legal ratio” of gold to silver, rather than the exchange rate, can be taken to be endogenous. While there might be six endogenous variables, there are only five independent restrictions on these six variables: the resource constraints for gold and silver, the first-order conditions for gold and silver jewelry, and the cash-in-advance constraint. Accordingly, there are multiple equilibria. Within limits, this allows the government to choose the legal price ratio of gold to silver; given that government-set price, competitive forces will then cause the stocks of gold and silver coins and also the stocks of gold and silver jewelry to adjust.10 Provided that they are not too large, disturbances to supplies and demands of gold and silver are consistent with a fixed government-set relative price of gold for silver because there is room for the monetary and nonmonetary stocks of the metals to adjust to assure equilibrium. Thus, Velde and Weber show that for a large country bimetallism can be much more robust than the analysis of the previous section based on the model of chapter 11 would suggest. Bimetallism is feasible for a range of exchange rates: given a “legal” ratio between gold and silver, quantities of metal can move between monetary and nonmonetary uses so that the relative value of the metals in nonmonetary uses, the market ratio, matches the legal ratio.
Velde and Weber model the world as a single economy, and consider a single legal ratio holding for all countries on a bimetallic standard, a state of affairs that presupposes some form of coordination among countries. Velde and Weber also show that changes in the legal price ratio can be necessary with more or less frequency, depending on the sizes of disturbances and on parameters that are influenced by the size of the bimetallic coalition. Such adjustments in the legal price ratio would require ongoing coordination among countries in a bimetallic coalition.11 A lack of coordination bedeviled European countries in the sixteenth and seventeenth centuries, when big movements in the gold/silver ratio prompted haphazard responses from a multitude of political units. Isaac Newton’s reports to the Treasury are filled with detailed comparisons of the gold/silver ratio in various European countries. But coordination gradually became easier. One of Napoleon’s legacies was a gold/silver ratio of 15.5, on which a number of European countries converged. The market price of gold in terms of silver remained remarkably stable in the vicinity of that ratio, even in the face of such large events as the California gold discovery. Consequently, for a long time in the nineteenth century, international bimetallism was viable with very few adjustments in the legal ratio. And it seems that it would have been viable for many years after that. Nevertheless, for reasons apparently unconnected to the prospects then prevailing for sustaining a stable legal ratio, bimetallism suddenly collapsed in 1873.12
Passage to gold
The United States and France adopted the standard formula when they reassembled coherent monetary policies after the wreck of bimetallism. In the United States, political forces striving for a return to bimetallism long delayed the extension of the standard formula to the outstanding stock of silver coinage. In the Latin Union, member states had to negotiate an arrangement suitable to all.
The United States
In the United States, the process of adopting the standard formula was long and arduous.13 The Mint Act of 1792 (1 Statutes at Large 246) established a bimetallic system with gold and silver coins freely minted on demand and unlimited legal tender. Full-bodied copper coins were minted on government account. The ratio between gold and silver was initially set at 15:1, but in the 1820s it came under reconsideration. Congress explicitly decided to maintain bimetallism, but debased the gold coins and changed the ratio to 16:1.
When the gold discoveries in 1849 pushed the intrinsic value of the silver coinage above its legal tender value, Congress sought to remedy the situation without forsaking bimetallism. The result was an Act of 1853 (10 Stat. 160). It modified the coinage of 5¢, 10¢, 25¢, and 50¢ silver pieces by ending their free coinage, making them 6.9% below weight with respect to the $1 silver coin, and making them legal tender for private debts for up to $5. The Secretary of the Treasury was to regulate the quantities minted, and the coins were to be sold at par by the mint in exchange for gold coins or silver dollars. At the same time, Congress made explicit its intention to maintain bimetallism, and declined to follow the English model.14 The measures were considered temporary, the diminution in weight was minimal, and there was no mechanism for redemption. According to Carothers (1930, 126–27), Congress expressed concerns over the perceived risks in government-issued token coinage: not only the possibility of private duplication, but also over-issue on the part of government and consequent depreciation. The ambiguities of the British model led to serious misunderstandings: “There was an almost universal belief that a coin legally rated at a value above its bullion value was a debased coin of doubtful honesty.”
After the Civil War ended, and as the greenback episode drew to a close, the United States prepared to return to a metallic standard. Instead of returning to the bimetallic standard, it instead adopted a gold standard in 1873. Silver was no longer freely minted. The silver dollar, still legal tender, was henceforth minted on government account only (albeit in quantities set by law from 1878 to 1893). Coins of 1¢, 3¢, and 5¢ in copper and nickel were created, with legal tender limited to 25¢. Finally, by an act of 1879 (21 Statutes at Large 7) the legal tender limit was also raised to $10, and, more importantly, it was provided that “the holder of any of the silver coins of the United States of smaller denominations than one dollar, may, on presentation of the same in sums of twenty dollars, or any multiple thereof, at the office of the Treasurer or any assistant Treasurer of the United States, receive therefor lawful money of the United States.”
Thus, for the first time, the standard formula was adopted in the United States for subsidiary coinage smaller than $1. The silver dollar remained inconvertible. Its metallic content started to lose its value as events in Europe triggered a collapse of the price of silver.
The Latin Monetary Union
The Latin Monetary Union bound France, Belgium, Italy, and Switzerland from 1866 to 1926 (Greece joined in 1868).15 It was formed to coordinate policies regarding subsidiary coinage, and became the framework within which the member countries conducted their passage to the gold standard.
The center of the union was France, its largest member. France’s monetary unit, the franc, had been defined as 4.5g of silver by a law of 17 Germinal XI (in the Revolutionary calendar, or April 7, 1803). The same law provided for the coinage of gold francs at a 15.5 ratio to silver francs, and coins of both metals were unlimited legal tender. Full-bodied silver coins ranged from 0.25F to 5F, and a few bronze coins of 0.05F and 0.10F were occasionally minted at cost on government account. As new countries formed nearby, they adopted an identical system, with an identical content of the unit of account: Belgium in 1832, Switzerland in 1850, Italy in 1862.
In the early 1850s, the relative price changes caused by the gold discoveries in California made silver coins disappear from circulation. Several countries reacted as the United States did in 1853, by making the smallest coins token. Switzerland reduced the silver content of its subsidiary coins (2F and lower) by 11% in 1860. Italy made its subsidiary coinage 7% lighter in 1862. France reduced by the same amount the content of its smallest coins only (0.50F and lower) in 1864. Belgium then called for a coordinated approach to subsidiary coinage; a monetary convention in 1865 resulted in the Latin Monetary Union, which went into effect in August 1866 for fifteen years, renewable by tacit consent.
The union adopted a surfeit of protective measures: a common standard of token coinage was set for all coins smaller than 5F; the subsidiary coins had limited legal tender; their issue was limited on a per capita basis country by country; they were accepted by all governments up to 100F and redeemable on demand into full-bodied silver or gold coins. Switzerland had asked for further precautions, in particular that each country keep the profit made on token issues in a reserve fund, as it had done since 1860. This 100 percent reserve requirement for token coinage was rejected by the other countries.
The union also specified the silver and gold contents of the full-bodied coins. Because it did not require that those coins be freely minted, and made them unlimited legal tender only for debts to governments, the arrangement did not embody an explicit bimetallic standard. The other members had wanted the gold standard, but France wished to maintain its existing system intact. The treaty did make the 5F piece legal tender throughout the union, and the members continued to allow silver be freely minted, as in the United States (a 5F piece was very close in size to a $1 piece). France’s position that bimetallism was sustainable and that the world gold/silver ratio could be stabilized so long as enough countries adhered to a common ratio makes sense in the model of worldwide bimetallism that we described earlier.
Free riders in monetary unions
The treaty had fully accepted token coinage but not fiat money. The union was soon tested.16 On April 30, 1866, Italy declared war on Austria. The next day, Italy suspended convertibility of its banks’ notes and required Italian banks to buy government bonds. During the war, Italy acquired Venice and also inflation. Small denomination notes soon appeared, privately issued at first (Fratianni and Spinelli 1997, 76). Italy’s token coinage was legal tender everywhere in the union and soon flooded it. When the union came up for renewal in 1878, the sum of Italian small denomination notes and coins exceeded Italy’s legal quota by 100%. The negotiations for renewal of the union centered on depriving Italian subsidiary coins of their legal tender status in the rest of the union and on forcing Italy to reacquire them. Italy grudgingly agreed to redeem its subsidiary coins circulating in other member countries for full-bodied coins, and to use those subsidiary coins to redeem its small notes. Soon Italy was able to restore convertibility of all its notes, and its coins were legal tender again throughout the union. But another crisis in 1893 prompted another suspension, followed by another agreement with other members of the union for Italy to repatriate its subsidiary coinage. Greece’s subsidiary coinage was likewise “nationalized” in 1908.
The accident of 1873
The end of bimetallism in Europe, as in the United States, posed the same problem of convertibility for what had previously been a full-bodied silver coinage. In 1871, Germany adopted the gold standard and embarked on a program designed to replace its silver coins with gold coins. To do so, it sold silver and bought gold at the prevailing exchange rate of 15.5. For reasons that remain unclear,17 the Latin Union proved unwilling to absorb Germany’s stocks of silver, and its members suspended free coinage of silver in quick succession in 1873, at the same time as the United States. Bimetallism disappeared abruptly. The market price of gold in terms of silver increased substantially, devaluing the stocks of silver of the formerly bimetallic countries and formerly silver-standard countries like Germany. These silver coins became effectively tokens in terms of the countries’ units now based on gold.
Members of the Latin Monetary Union were quick to impose on themselves annual caps on silver minting. Then in 1878, they agreed to suspend minting of silver and to resume it only by unanimous decision. France had already suspended free coinage of silver by a law of August 5, 1876. The preamble of the law presented the suspension not as the collapse of a system, but as the final step in an evolution: “The theory of the double standard, on which our monetary law of the year XI is based, has been called into question ever since its origin. It is, in our view, less a theory than the result of the primitive inability of the legislators to combine together the two precious metals otherwise than by way of an unlimited concurrence—metals, both of which are destined to enter into the monetary system, but which recent legislators have learned to coordinate by leaving the unlimited function to gold alone and reducing silver to the role of divisional money.”
The standard formula limps into place
By 1900, Britain, France, Germany, and the United States all had implemented the standard formula in practice, though curiously enough, the promise of convertibility had not been universally formalized. Thus, Britain remained officially silent about convertibility, but had made convertibility a longstanding practice. To be sure, Cannan (1935, 40) saw this practice as having built a set of expectations: “In this country, there is little doubt that in case of a considerable falling off of demand [for subsidiary coins] the Government would be compelled to take back enough of the coin to keep up its value, and the obligation might just as well be acknowledged at once.” Cannan had written this in 1918, and a footnote added in later editions cited withdrawal of silver coin from 1921 to 1924 as proof that the expectations were valid.
The formerly bimetallic countries were confronted with the issue of convertibility to a greater degree. Their belated adoption of the gold standard had saddled Germany, France, and the United States with large stocks of previously standard silver coins, creating a situation known as the “limping standard.”
Those silver coins (thalers in Germany, 5-franc pieces or “écus” in France and the Latin Monetary Union, dollars in the United States) were legal tender but de facto token, as the value of their silver content had plummeted along with the world price of silver after 1873.
In the United States, as a result of political tussles, considerable quantities of silver dollars were produced after the suspension of silver coinage and until 1893. Only by an act of March 14, 1900 (31 Statutes at Large 45) did Congress make it the Treasury’s duty to maintain parity to the gold dollar of “all forms of money coined and issued by the United States.” This firmly established the gold standard, and set in law a practice to which the Treasury had held fast until then, sometimes at considerable cost (Dewey 1918, 444–55).
Members of the Latin Monetary Union, alert to the free-rider problems inherent in their union, dealt more quickly with the status of their silver coinage. In 1878, they bound themselves to keep the stock of silver coins constant. But the coins remained legal tender for their face value throughout the union, and circulated widely outside their countries of origin. Nothing had been said about the status of the coins should the union dissolve or a country secede, which could happen with a year’s notice. In 1884, France made clear that it wanted the issue addressed, and a new treaty was negotiated the following year. The negotiations were difficult: expecting to be a net debtor, Belgium initially opted out and won some last-minute concessions. Ultimately, the members committed themselves to redeeming their net balances of 5F pieces in gold or gold-denominated instruments within five years of the dissolution of the union (Belgium and Italy committed to redeem only half of their balances). Meanwhile, as long as the union remained in force, the Bank of France promised to accept all 5F pieces at par.18 As in Britain, the commitments were ultimately honored.19
1 France suffered a severe shortage in the early days of the Revolution, around 1790 (Marion 1919, 270–78). Private coins such as the one shown in figure 4.7 were produced to meet this shortage.
2 See the overview in Sprenger (1993).
3 Gold coins were minted in some parts of Germany, but they had no fixed legal tender value.
4 See the discussion of limping standards at the end of this chapter.
5 See Redish (2000) for an extensive study of bimetallism.
6 Its proponents in the late nineteenth century emphasized other properties such as its superior ability to stabilize the price level in the face of separate fluctuations in the price of gold and silver. This argument was made by French officials as early as 1867 (Willis 1901, 76).
7 See the discussion in the section on page 310 entitled “Worldwide bimetallism.”
8 We have seen such an event take place in the 1340s, in chapters 9 and 10.
9 Velde and Weber’s formal model thus abstracts from the differing denominations that gold and silver coins took in practice, and does not include a penny-inadvance constraint. Gold and silver coins are treated symmetrically.
10 See Velde and Weber (2000, 1216–17), in particular the quotation ascribed to Walras.
11 The need for international coordination was keenly felt by proponents of bimetallism in the nineteenth century.
12 Friedman (1990), Flandreau (1996), Velde and Weber (2000), and Redish (2000, 202–6) discuss whether that collapse was inevitable. Velde and Weber discuss the welfare implications of various choices of the legal ratio, and find that in general they cannot argue that bimetallism was welfare-improving relative to a uni-metal standard.
13 See Carothers (1930) and Redish (2000, 209–39).
14 Congress was aware of the English model, described upon request by the U.S. ambassador to Britain (Carothers 1930, 114).
15 We largely follow Willis (1901) on the history of the union.
16 Zarazaga (1995) presents an equilibrium model of the coordination problems inherent in a system with multiple issuers of fiat money.
17 Flandreau (1996) has speculated on the relevance of the 1870–71 war between France and Germany.
18 A similar policy on the part of the German Reichsbank kept the silver thalers at par until they were retired in 1907 (Mises 1953, 55).
19 When World War I came, all belligerent countries suspended convertibility, and Switzerland became the unwilling recipient of their subsidiary coinage. France and Belgium made good on their obligations when the union came to an end in January 1927.