Coinage began in Lydia, in the interior of West Anatolia, in the form of electrum in the second half of the 7th century B.C., around 630. Electrum is an alloy of gold and silver that varies in its natural state between 65% and 85% gold, and neither weighing nor the touchstone can tell the percentage in between with any precision. Therefore, the metal would seem to be a particularly poor choice for coinage, and it was widely considered so until recent decades. But more recently, the logic of the Lydian choice has emerged. Interestingly, it is the very uncertainty of the value of electrum that is the key to the benefit of its coinage. As a corollary, the subsequent coinage in silver in Greece, beginning about 80 years later, requires a separate explanation. I shall develop these themes below.
When Lydia decided to coin, money was already widely present in the form of gold and silver in its trading region for centuries. Had the Lydians known how to separate the raw electrum between gold and silver, they might well have done so. But they did not know how; their ability to divide up electrum between silver and gold came later. Field research in the late 1960s uncovered remains of a refinery for separating unadulterated electrum between the metals on the site of Sardis, capital of Lydia. The refinery dates to about the time when King Croesus of Lydia decided to substitute fresh gold coins for the electrum ones. There is some speculation that the Lydians could have separated the gold and silver earlier had they wanted to. But if so, they manifestly found it more economical to coin the electrum.
The Lydians must also have possessed large deposits of the alloy to be tempted to monetise it despite its unfortunate features. For many years, the general professional opinion was that electrum flowed directly down from the Pactolus River right beside the city of Sardis. But in the early 2010s, the startling discovery was made through mineralogical research that the flow of precious metal from the Pactolus River must have been pure gold (Cahill et al. 2014). It is therefore considered likely now that the Lydians obtained their electrum instead from the northwest region of their empire, in today’s Turkey.
The Lydian motives
Why did the Lydians coin the electrum? After all, they could simply have circulated the unadulterated metal, cut into proper parts, as money. The reason emerged around the year 2000, when the exciting discovery was made that the Lydians could control the gold content of their electrum coins closely. The royal coins of the mint at Sardis that were studied with advanced non-destructive methods contain 55% gold, plus or minus 2%, which could only be so voluntarily. Moreover, the Lydians always attained their desired mix of gold and silver by adding silver rather than reducing gold, which means mixing rather than separating. Mixing is a much simpler process and was long well-known.1
The advantage of coining electrum becomes apparent. The coins provided more security about the contents than the unadulterated pieces of the metal did. They offered an insurance premium. Of course, this premium depended on trust in the issuers. But judging from the success of electrum coins in the whole monetary region surrounding Lydia – including Mesopotamia to the east, the Black Sea traders to the north, Syria to the south, where the Phoenicians were active, and mainland Greece – the proper conditions were met. Conformably, the evidence shows only moderate counterfeiting of the coins, though issuers were many. Besides the Lydian king, the Greek colonies in Ionia on the eastern coast of the Mediterranean adjacent to Lydia took a large part in production. These colonies may also have been the first to coin, though we shall probably never know.
Following the Lydian experiment, coinage did not take off anywhere for 80 years and when it did it was in mainland Greece. Only with Alexander of Macedonia’s conquests in the late 4th century B.C. did coinage spread widely in the entire region to the north and south of Greece and to the east deep into central Asia. Mainland Greece started coining at about the time of Croesus’ substitution of gold coins for electrum ones. Greece also decided to coin strictly in silver. The basic reason may be the discovery of large silver deposits in Laurion, to the southeast of Athens, around the time.
But why did the Greeks decide to coin the silver instead of circulating it in the familiar forms as ingots or bars? In the case of silver, there was no insurance premium needed of the sort that electrum required.2 The scales would have done reasonably well, especially with the help of experience and the touchstone. It is often supposed that the coinage yielded large profits despite the added production costs. But this is highly debatable. If the conjecture was true, how come the clever Phoenicians, Mesopotamians, and Egyptians failed to follow suit for over a century and a half? Their metallurgical skills were as well-honed as those of the Greeks. Coinage would have been child’s play for them.
All sorts of speculation are possible. An interesting and debated theory, by Schaps (2001: 93-103, 2004: 16-17, 108-110), is that the Greeks had only acquired advanced monetary habits more recently than others in the region, and therefore found greater allure in the innovation of coins during a period of demographic and economic expansion. Other hypotheses are possible. But there is one overriding consideration: the Greeks almost surely subsidised coinage. This must be part of the answer.
The evidence of public subsidy
In the footsteps of Lydia and Ionia, early Greek coinage had an unusually developed system of denominations. The silver denominations went down from a tetradrachma at the top (4 drachmas) to one obol (1/6 of a drachma) to 1/8 of an obol (1/192 of a tetradrachma) in about nine steps. By all counts, this is exceptional in the entire pre-Renaissance world and not seen in Europe for many centuries following the fall of the Roman Empire. Medieval Europe had virtually only a single, light coin for five centuries beginning in the 7th century, composed of silver: variously the penny, denier, denaro, or pfennig (Spufford 1988: 27). Any range of denominations to speak of only began in 1201, with the appearance of the grosso in Venice, another silver coin. To quote Grierson (1975: 27): “it was only between the beginning of the 13th and the mid-15th century... [that] Latin christendom came to have at its disposal a wide range of denominations, reverting thus in some measure to the monetary pattern of antiquity” [my italics]. This mid- and late-medieval range of denominations also rested on a variety of metals and alloys, never pure silver alone (as in Greece).
The case of China deserves separate notice. During the middle ages, China possessed a system of coin denominations that it had inherited from antiquity; but the system consisted strictly of lower denominations in the form of copper/bronze. The Chinese system never provided individual coins of sufficient value to buy a day’s consumption, whereas the top Greek denomination would easily buy over a month’s consumption. True, the Chinese typically possessed complementary silver money for purchase of big-ticket items, not always, but never in the form of coins. The first wide circulation of silver coins in the country came in the 1820s, consisting of Mexican coins.
The issue is obvious. To produce a low-value coin costs more as a percentage of market value than to produce a high-value coin of the same material with the same methods and differing only in size. Furthermore, the relevant technology in coining “was little changed from Greek and Roman times” in late-Medieval Europe (Sargeant and Velde 2002: 50). Thus, the costs of coining (brassage) in ancient Greece were about five or six times larger per unit weight of silver on the lower denominations than the higher ones (my estimate from Melitz 2016, based on Sargeant and Velde 2002: 51). Therefore, even if the highest denomination Greek coin was highly profitable (despite the consequent encouragement of substitutes and invitation to others in the metals industry to offer lower prices), all denominations of notably lower weight must have been at a loss. The conclusion that the system of denominations represented a government subsidy is almost unavoidable.
Conclusion about motives
The reasons for such subsidy are also easy to find. One would be the social aim of providing a collective service to the community in facilitating trade. But the state might also foresee an economic gain for itself from promoting monetisation. As the largest spender and recipient of money in the community in the form of precious metals in numerous small lots, the government had much to gain from the spread of coinage in order to economise its own transaction costs in its budgetary affairs. A proper analogy would be the incentive of contemporary governments to encourage popular use of computers, at public expense, in order to induce online declarations of taxable income. In my effort to model the beginnings of silver coinage (Melitz 2016), I relied strictly on the government’s enlightened self-interest. But that is simply a choice.
Cahill, N, J Hari, B Onay and E Dokumaci (2014), “Depletion gilding of Lydian electrum coins and the sources of Lydian gold”, Proceedings of the International Congress, Israel Museum, Jerusalem, 25-26 June.
Grierson, P (1975), Numismatics, Oxford University Press.
Melitz, J (2016), “A model of the beginnings of coinage in antiquity”, European Review of Economic History 21: 83-103
Sargeant, T and F Velde (2002), The big problem of small change, Princeton University Press.
Schaps, D (2001), “The conceptual prehistory of money and its impact on the Greek Economy,” in M Balmuth (ed.), Hacksilber to coinage: New insights into the monetary history of the Near East and Greece, American Numismatic Society, pp. 93-103.
Schaps, D (2004), The invention of coinage and the monetization of ancient Greece, Ann Arbor: University of Michigan Press.
Spufford, P (1988), Money and its use in Medieval Europe, Cambridge University Press.
 All bibliographical support is in Melitz (2016).
 Of course, the same question can be posed for Croesus regarding gold, but the latter’s choice of coining is easier to understand since he was replacing electrum coins.