Pre-industrial Bimetallism: The Index Coin Hypothesis
Ernst Juerg Weber1
University of Western Australia
Business School – Economics Program
In early monetary systems the unit of account was separate from the medium of exchange. Commodity prices and prices of coins were quoted in terms of a fixed quantity of metal that was embodied by an 'index coin'. Coins circulated at their metal value because coinage was imperfect and fixed exchange rates would have interfered with the operation of bimetallism. An indication that the exchange rates of coins were market determined is the absence of value marks on coins. During the Industrial Revolution, improvements in the quality of coinage led to the fusion of the unit of account and medium of exchange function of money. As a consequence, pre-industrial bimetallism gave way to nineteenth century bimetallism, in which the make of currencies alternated between silver and gold.
Bimetallism prevailed for almost two and a half millennia, from the origins of coinage in antiquity until the nineteenth century. In early bimetallism full-valued gold coins circulated side by side with silver coins, occasionally supplemented by token coins, which were made of an alloy of base metals. During the Industrial Revolution, the historic bimetallic standard degenerated, giving way to a bimodal monetary standard, in which, depending on market conditions, the make of currencies alternated between silver and gold. Unlike pre-industrial bimetallism, nineteenth century bimetallism was essentially monometallic, with each metal intermittently serving as medium of exchange. By the end of the nineteenth century, all major countries were on monometallic standards: Europe and America on gold and Asian countries on silver. Compared with the endurance of bimetallism, the gold-standard era was short, providing a mere transition period between historic bimetallism and the modern paper standard that eventually arose during the twentieth century.
Monetary economists tend to deplore the demise of bimetallism.2 Friedman (1990a,b) argues that the abandonment of bimetallism by the United States was a costly policy mistake that went largely unnoticed when Congress debated the Coinage Act of 1873. Gallarotti (1994) suggests that policy makers succumbed to a "growing ideological attraction to gold" in both Europe and the United States. Oppers (1996) takes this as evidence that Germany and the Latin Monetary Union capriciously switched to gold in the 1870s. Flandreau (1996) calls the emergence of the international gold standard a "blatant failure of international cooperation." However, the fact that bimetallism was abandoned universally and permanently makes it unlikely that the adoption of monometallism was an inadvertent policy mistake. It is also unclear why this alleged policy mistake should have occurred in the 1870s and not at another time, maybe two hundred years earlier? The main flaw of the view that the switch to monometallism was avoidable is that it compares the gold standard, which took hold in the second half of the nineteenth century, with the degenerate nineteenth century bimetallism that preceded it. De facto monometallism had emerged in most industrial countries already decades before the official change to gold, which simply recognized the prevailing state of monetary affairs. Thus, the historic dividing line between bimetallism and monometallism lies not in the 1870s, but decades earlier (in Great Britain more than a century earlier), when pre-industrial bimetallism, in which gold and silver coins circulated side by side, gave way to alternating nineteenth century bimetallism.
This paper reviews the history of bimetallism from antiquity until the nineteenth century. The analysis focuses on the fundamentals of bimetallism, namely the technological and institutional conditions that determined the process of coinage. The main argument is (1) that pre-industrial bimetallism differed from nineteenth century bimetallism and (2) that technical advances in the process of coinage made bimetallism unworkable during the Industrial Revolution. In pre-industrial times the unit of account was separate from the medium of exchange because the quality of coins was poor. The crux of the argument is that there is no obvious unit of account in a metallic monetary system with heterogeneous coinage. As there was no high quality coin that could have served as unit of account, how did merchants quote prices and in what unit of account did they conduct commercial calculations? The hypothesis in this paper is that pre-industrial merchants used the official mint weight of a leading coin of their hometown for accounting purposes. Thus, both commodity prices and prices of coins (exchange rates) were quoted in terms of a fixed quantity of metal, embodied by an 'index coin'.3 Even specimens of the index coin had a market price that deviated from parity because of wear and tear and deliberate abuse. Therefore, the index coin did not enjoy a decisive advantage as a medium of exchange, and it circulated side by side with other coins, even though its official metal weight served as unit of account. The index coin stayed on an equal footing with other coins as a medium of exchange until the Industrial Revolution. Then, the quality improvement in coinage gave rise to de facto monometallism by fusing the unit of account function of the index coin with the medium of exchange function. The ability to circulate at par gave the index coin (and its multiples and fractions) a decisive advantage as medium of exchange. At the same time, the operation of Gresham's Law frustrated attempts to retain a bimetallic coinage by defining the unit of account in both silver and gold. Thus, the quality improvement in coinage lies at the root of the transformation of pre-industrial bimetallism to de facto monometallism during the nineteenth century.4
The index coin hypothesis complements standard price theory in order to explain the purchasing power of money in a monetary system with heterogeneous coinage. Economic forces determine the exchange value of a unit of monetary metal and the official mint weight of the index coin anchors the purchasing power of the unit of account. The mint weight of the index coin, the inverse of the official price of the monetary metal, is a purely nominal concept, playing the same role as the money stock in a paper standard. Deliberate changes in the mint weight of the index coin produced proportional changes in the purchasing power of the unit of account. This paper applies the index coin hypothesis to the monetary systems of ancient Greece (Section 1), Rome (Section 2) and the Middle Ages (Section 3), and to the transition of pre-industrial bimetallism to de facto monometallism during the Industrial Revolution (Sections 4 and 5). The conclusion (Section 6) relates the index coin hypothesis to issues in modern monetary economics.
1. The Origins of Bimetallism in the Greek World
Coinage was invented in western Turkey in the late seventh century BC.5 The region was inhabited by Greeks, who lived in coastal cities, and by the Lydians, whose capital Sardes lay further inland. The earliest coins consisted of electrum, a natural alloy of gold and silver (white gold), which was found locally in alluvial deposits. Geological processes determined the composition of natural electrum, although mints quickly mastered the technique of alloying electrum. The analysis of two coins yielded 47.8% and 50.0% of gold, and 47.0% and 43.3% of silver, the remainder being accounted for by impurities (Cooper 1988, p. 8). Electrum coins played a significant role in early monetary history, circulating in western Turkey and the Black Sea area from the seventh until the fourth centuries BC. Remarkably, the history of coinage had started with a symmetallic standard, as envisaged by Marshall (1887).6
The legendary Croesus (561-546 BC), the last king of Lydia, is often credited with the creation of pure silver and gold coins. But silver coins emerged on the island of Aegina, across the Aegean Sea, at about the same time (and possibly somewhat earlier). In the sixth century BC, coinage spread from western Turkey to the Greek mainland and beyond to Greek cities in Sicily and southern Italy. Apart from the early use of electrum, Greek cities almost exclusively struck silver coins until well into the classical period (the time from the Persian wars (492-479 BC) until Alexander the Great (356-323 BC)). The first significant amount of gold coinage was issued by Athens when Sparta occupied the silver mines at Laurium during the Peloponnesian War (431-404 BC). Philip II (382-336 BC), who established Macedonian hegemony in Greece, minted both silver and gold coins, although most gold coins may have been issued posthumously by his son Alexander (Carradice and Price 1988, ch. 7).7 On the Persian campaign, Alexander founded a large number of mints, introducing the practice of coinage as far eastward as the Indus valley. Earlier, coins had been struck only in the western satrapies (provinces) of the Persian Empire, in the former kingdom of Lydia.
Almost every Greek city and island minted coins, displaying a variety of designs, including animals, mythological figures and gods. Some coins were more ubiquitous than others (for example the Athenian silver 'owls' in the second half of the fifth century BC), but no mint ever established a monopoly in the Greek world. The 'owls' of Athens competed with the 'turtles' of Aegina, the 'colts' (named after Pegasus) of Corinth, and other coins. At the pinnacle of Hellenistic influence, during the fifth and fourth centuries BC, Greek coins circulated freely in the Mediterranean and the East, from Spain to the Indus valley. Alexander put his portrait on imperial coins, but his rule was too short to secure a lasting dominance of imperial coinage. After Alexander's death, the diadochi (Alexander’s generals) carved up the empire and new royal coinages emerged. During the Hellenistic period, which lasted from Alexander’s death until the ascendancy of Rome, Greek coinage became so complex that, according to Carradice and Price (1988, p. 122), "no coherent account of it has ever been produced."
The names of Greek coinages, stater and drachma, are derived from commercial weights. The word stater means "that which balances the scales", and drachma originated from drax, the Greek word for a "handful of spits". As commercial weight standards varied across cities, coinages were struck to mixed standards. The Corinthian stater weighted 8.6 grams, the Aeginetic stater was 12.2 grams, the Milesian or Lydian stater was 14.1 grams, and the Euboeic stater was 17.2 grams. The drachma represented a smaller weight: 2.85 grams at Corinth, somewhat over four grams at Athens, and slightly over six grams at Aegina. Hence, the Corinthian stater was a three-drachmae piece, the Aeginetic stater was a didrachmon, with similar relationships existing elsewhere. Coinages included elaborate sets of multiples and fractions of the basic monetary unit. Multiples usually followed the dual system (didrachm, tetradrachm) and fractions were expressed as one sixth. The obol was one sixth of a drachma.
The legacy of Greek coins suggests that merchants were accustomed to use a heterogeneous coinage that was issued by a large number of mints. In this monetary environment, how did merchants quote prices and in what unit of account did they conduct commercial calculations? The hypothesis in this paper is that the unit of account was separate from the medium of exchange. Accordingly, merchants used the official mint weight of a leading coin of their hometown for accounting purposes. Thus, both commodity prices and prices of coins (exchange rates) were quoted in terms of a fixed quantity of metal, embodied by an index coin. Even specimens of the index coin had a market price that deviated from parity if they were badly worn. If for example the drachma of Athens served as index coin, then all other coins were valued in terms of the official silver weight of the drachma (somewhat more than four grams), while underweight drachma pieces traded at a discount. For this reason, the drachma did not enjoy a decisive advantage as medium of exchange, and it circulated side by side with other coins, even though its official silver weight served as unit of account.
The omission of value marks on Greek coins suggests that they circulated at market determined exchange rates. Value marks would have been a nuisance because they would have conflicted with actual exchange rates if coins were imperfect. In monetary history the absence of value marks is a reliable sign that coins traded at exchange rates that depended on their true metal weight and prevailing metal prices. Major coins lacked value marks from antiquity until the eighteenth century (with a notable exception during the late Roman Republic). On the other hand, the use of value marks indicates that coins were designed to circulate as tokens whose assigned values exceeded their true metal value. In Greek times, tokens were used only for emergency issues and minor denominations. Among the earliest tokens were those of Timotheus, an Athenian general, who paid his troops with bronze coins in the war against the Chalcidian League (364-359 BC). The coins, which were marked with one or two dots, were designed to pass for one or two obols. It remains an open question whether Timotheus’s promise to redeem the coins after the war and his prospects of victory were sufficiently credible to establish parity with the silver obol.8
In general, the quality of Greek coins was high, although there are examples of carefully filed coins, in particular from the early centuries of coinage. This confirms that coins circulated at market determined exchange rates. Free exchange rates between coins protected against deliberate abuse because it is not worthwhile to painstakingly clip and file coins if this reduces their value by the very amount to be gained through these practices. The abuse of coins is profitable only if maltreated coins can be passed on at official exchange rates, irrespective of their true metal weight. Since the valuation of coins required special skills, some cities, for example Athens in c. 375 BC, employed officials who tested coins and settled disputes among merchants. According to an inscription, the officials certified local coins (and possibly foreign coins that were at par with local ones), giving them legal tender status.9 Yet, the certifiers did not impose unrealistic exchange rates because this would have led to the widespread abuse of coins, for which, unlike in medieval Europe, there is no evidence in Greek times. For these reasons, in Athens the index coin was the drachma, which was accepted by tale (by counting individual coins) if it conformed to the official standard, and which, together with foreign coins, was traded at market determined rates if it was underweight.
Deliberate inflation through a reduction in the mint weight of the index coin was uncommon during antiquity. Usually, the mint weight of the index coin remained close to the corresponding commercial weight standard. Commercial weights that were inscribed with their coin equivalents show that the mint weight of the Athenian drachma was only about five percent less than the commercial drachma weight (Carradice and Price 1988, p. 92). Actually, this may have been the brassage (mint-charge). In the ancient world currency competition prevented deliberate inflation by a single city because the public could easily switch to more stable coinages, putting the mint of the inflating city out of business. The only substantial monetary debasement occurred in Ptolemaic Egypt, at the periphery of the Hellenistic world.10 The distant location of Egypt, which could be reached only by sea and through deserts, enabled the monetary authority to suppress currency substitution by controlling the influx of foreign coinage.
2. Roman Monetary Standards
During the first half of the third century BC, Rome gained control of peninsular Italy.11 Then, the struggle for supremacy in the western Mediterranean led to three wars with Carthage, the so-called Punic Wars (264-241, 219-202, and 149-146 BC). After civil wars in the first century BC, the Roman Republic gave way to the Imperial system, which survived until the fifth century AD. Rome operated several monetary systems during its long history: key dates include the currency reform during the Second Punic War and the introduction of imperial coinage. Accordingly, it is convenient to distinguish between three time periods in Roman monetary history: the early Republican monetary system before the Second Punic War, the Republican monetary system from the Second Punic War until the end of the Republic, and the Imperial monetary system.