Most probably, early coins were only mildly profitable at best, and quite possibly were even subsidised by the state as part of a political strategy of encouraging their spread. The reason for this political strategy would be the considerable economies of transaction costs that the state could gain from the spread of coinage in its own revenue-seeking and spending activities. A proper analogy would be the interest that contemporary governments have to encourage popular reliance on computers, at public expense, in order to induce online declarations of taxes.
Research traces the beginning of coinage with increasing accuracy to around 630 BC in the Greek city-states in Ionia or in Lydia, or both, in the contemporary part of West Turkey east of the Aegean Sea. The earliest coins were made of electrum, a mix of gold and silver. The common view, even among knowledgeable scholars, is that early coinage was highly profitable, at least for the Lydian kings Alyattes (610-560 BC) and Croesus (560-547 BC). It is “usually understood [that] the electrum coins were highly overvalued”, say the archaeologists Cahill and Kroll (2004, p. 613, with minor rephrasing), by which they clearly mean highly profitable. In an influential book, Le Rider (2002, pp. 96-100) estimates a profit rate of about 15 to 20%. However, I argue that this position is very dubious (see Melitz 2015).
Consider first the historical context in which coinage began. The innovation occurred in societies that belonged to a vast trading network where the use of the precious metals as money went back at least over 1,000 years to the East, where the Assyrian empire stood, and at least many centuries to the South, in the Levant, where the Phoenicians had been active as international traders since Homeric times (that is, centuries before Homer). The producers of traded goods, wholesalers and traders possessed accurate scales for weighing gold and silver, were expert at detecting fineness, and could use the touchstone for help. Their benefit from coins would depend entirely on their ability to dispense with weighing and assessing and simply count, based on trust, but the associated savings could hardly make much difference in a large transaction, while the early electrum coins, consisting very roughly half and half of gold and silver, were only useful for big-ticket items. According to the estimates, one of the largest coins (a stater) might buy an ox, and the very smallest coins (tiny) would be far too valuable to serve as small change.1
In accordance with this line of reasoning, coinage took off very slowly outside of Ionia and Lydia, and it is arguable that its take-off was particularly slow where monetary habits with the precious metals were most deeply engrained. It took about 80 years before a few Greek city-states on the mainland and offshore started to coin, beginning in 550 BC. They did so in silver. Coinage subsequently spread widely in the Greek city-states over the next half-century or so, but almost nowhere else. After King Croesus fell to the Persian King Cyrus in 547 BC, the Persians imitated the coinage they found in conquered Lydia and encouraged its spread. The Persian coins were also in gold and silver following Croesus’ example (the latter had introduced separate coins in the two metals shortly before, in 550 BC). But the Persian coinage had most success in the western part of their (Achaemenid) empire and made little headway in the more sophisticated east. Quite significantly too, the Phoenicians who traded far and wide started to coin only in the middle of the 5th century and their Carthaginian outposts somewhat later at the end of the 5th century. Egypt, hardly a commercial backwater, did not begin to coin until the late 4th century BC (Lorber 2012). It was clearly the conquests of Alexander the Great in the last third of the 4th century BC and the subsequent political expansion of Rome in the next four centuries that led to the wide spread of coinage in the ancient world outside of China and India (where coinage had begun independently but much later than in Ionia and Lydia). North and central Italy and most of Europe also saw coinage arrive only under Roman influence. Rome itself started to coin late, around 300 BC, and coinage really only took off there with its military advances in the third century, that is, about two and a half centuries after coinage had covered most of Greece. This early history of slow progress is difficult to reconcile with the idea that the early coins were a source of exceptional profits.
As a further consideration, the area where early coinage took place was one of high metallurgical skills. Had the activity been highly profitable, we would have expected competition to have eroded the profits. Indeed there were over 300 different early issues of electrum coins in Ionia and Lydia, many of them private. Should not these competing issues have precluded extraordinarily high royal profits? And why did Croesus ultimately abandon electrum in favour of bimetallism? Had electrum coins become unprofitable?2 World monetary history also tells us that very profitable coinage attracts counterfeiting. Counterfeiting was an enormous problem prior to the introduction of highly sophisticated and mechanised methods of producing coins both in China since the Han dynasty in the 2nd century BC and Europe since medieval times. Yet we know of no corresponding problem in the Lydian and Greek poleis. To be sure, counterfeits crop up, but the problem is contained. Velde (2014) estimates “the counterfeits found in the numismatic trade” of Lydian electrum coins as “1 to 2% of the total number of coins”. Of the early Greek coins, Wallace (2001, p. 131) says: “It is a fact that the alloys of Greek coins were very carefully calibrated and scarcely ever adulterated”.
One last consideration is particularly damaging to the thesis of highly profitable early coinage. The states that produced the earliest coins displayed peculiar behaviour in the historical instances. They provided a range of denominations of coins that took centuries to emerge elsewhere (especially if we ignore the Roman republic and empire), and they did so despite the higher costs of producing the lower than the higher denominations. Lydian electrum coins and those of the nearby Greek colonies in Ionia had nine different convenient denominations going from one stater (the relevant unit of account) down to 1/192. The early Greek silver coins likewise had denominations going down from a tetradrachma at the top (four drachmas) to one obol (a sixth of a drachma) to one eighth of an obol (a hundred and ninetieth of a tetradrachma) in about nine steps. Even as wide a range of values as one to 24, to say nothing of one to 48 or 96, with the top coin worth quite a lot indeed, well over a week’s wages for an independent worker, as was the case is a remarkable feature of the early Lydian and Greek examples. Other early coinage systems only achieved the same result with at least two metals.3
Coinage was political
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. Sargeant and Velde (2002: 51) provide a table showing brassage costs (costs of production) as a percentage of value for silver and partly gold, drawn from various sources and covering nine examples for different parts of late medieval Europe. If we focus on denominations differing by around 20 to 30 to one, the costs of brassage are about five to nine times higher on the lower denomination than the higher one in the relevant readings (six of them). Thus, a brassage cost of 1% on a ‘stater’ (any unit) means 5% to 9% on a coin of one twentieth to one thirtieth of a stater. Furthermore, the technology “was little changed from Greek and Roman times” (Sargeant and Velde: 50).
It would seem quite apparent from all this evidence that the willingness of the Lydian government and the Greek city-states to absorb the cost of producing an extremely wide array of denominations of coins in a single precious metal must have reflected a political strategy of promoting coinage. Such a policy would also be easy to explain. As repeatedly stressed in the literature, the government itself had much to gain from the spread of coinage in managing its budgetary affairs because of its numerous payments and receipts of bullion in small individual lots (think of public salaries, the pay of mercenaries and soldiers and individual tax receipts).4
Cahill, N and J Kroll (2005), “New archaic coin finds at Sardis”, American Journal of Archaeology 109: 589-617.
Crawford, M (1970), “Money and exchange in the Roman world”, Journal of Roman Studies 60: 40-48.
Gitin, S and A Golani (2001), “The Tel Miqne-Ekron silver hoards”, in M Balmuth (ed.), Hacksilber to coinage: New insights into the monetary history of the Near East and Greece, American Numismatic Society: 27-48.
Grierson, P (1975), Numismatics, Oxford University Press.
Howgego, C (1990), “Why did ancient states strike coins?”, Numismatist Chronicle 150: 1-25.
Kraay, C (1964), “Hoards, small change and the origin of coinage”, Journal of Hellenic Studies 84: 76-91.
Kraay, C (1976), “Archaic and classical Greek coins”, Berkeley: University of Berkeley Press.
Kroll, J (2008), “The monetary use of weighed bullion in Archaic Greece,” in W Harris (ed.), “The monetary systems of the Greeks and Romans”, Oxford University Press, Chapter 1.
Kroll, J (2012), “The monetary background of early coinage”, in W Metcalf (ed.), The Oxford Handbook of Greek and Roman Coinage, Oxford University Press: 33-42.
Le Rider, G (2001), La naissance de la monnaie : Pratiques monétaires de l’Orient ancien, Paris: Presses Universitaires de France.
Lo Cascio, E (1981), “State and coinage in the Late Republic and Early Empire”, Journal of Roman Studies 71: 76-86.
Lorber, C (2012), “The coinage of the Ptolemies”, in W Metcalf (ed.), The Oxford Handbook of Greek and Roman Coinage, Oxford University Press: 211-234.
Melitz, J (2015), “A formal analysis of the beginnings of coinage in antiquity”, CEPR Discussion Paper no. 10795.
Sargeant, T and F Velde (2002), The big problem of small change, Princeton University Press.
Scheidel, W (2008), “The divergent evolution of coinage in Eastern and Western Eurasia”, in W Harris (ed.), The monetary systems of the Greeks and Romans, Oxford University Press: 267-286.
Scheidel, W (2009), “The monetary systems of the Han and Roman Empires”, in W Scheidel (ed.), Rome and China: Comparative perspectives on ancient world empires, Oxford University Press: 137-207.
Spufford, P (1988), Money and its use in Medieval Europe, Cambridge University Press.
Van Alfen, P (2015), “Archaic small change and the logic of political survival”, American Numismatic Society.
Van de Mierop (2014), “Silver as a financial tool in ancient Egypt and Mesopotamia”, in P Bernholz and R Vaubel (eds.), Explaining monetary and financial innovation, Springer: 17-29.
Velde, F (2014), “A quantitative approach to the beginnings of coinage”, Federal Reserve Bank of Chicago, available online.
Von Glahn, R (1996), “Fountain of fortune: Money and monetary policy in China 1000-1700”, Berkeley: University of California Press.
Wallace, R (2001), “Remarks on the value and standards of early electrum coins”, in M Balmuth (ed.), Hacksilber to coinage: New insights into the monetary history of the Near East and Greece, American Numismatic Society: 127-134.
1 On the general historical context in which coins began, see Gitin and Golani (2001), Kroll (2008, 2012), and Van de Mierop (2014), among others. Concerning the purchasing power of the Lydian coins, see Velde (2014) and van Alfen (2015). Bullion also continued to dominate coinage in medieval Europe in large commercial transactions for many centuries after the spread of coinage (see Spufford 1988). China never even coined silver during its so-called “silver century”, 1550-l650 (see von Glahn 1996). It should be noted too that electrum in its natural state was an uncertain mixture of gold and silver. Even though this uncertainty matters in explaining why early coinage took place in this metal, it is not much relevant here since the only issue is whether the early coins were really a source of large and sustained windfalls over decades (see Melitz 2015).
2 Interestingly, the argument of the archaeologists and numismatists that the royal Lydian coinage was highly profitable hinges entirely on inferences about the events at the time when Croesus switched to bimetallism around 547 BC. However, these inferences suffer from a failure to consider that we have no way of knowing at what exchange rate the earlier electrum staters were traded for new gold staters at the mint. I discuss the matter in more detail in Melitz 2015, note 11.
3 In the case of the Roman Republic, the two metals were bronze and silver. In medieval Europe, they were silver and alloys of silver and baser metals, and they also came after many centuries, namely, only in the thirteenth (see Grierson 1975: 27, and Spufford 1988). In the Chinese case, a wide range of market values of denominations (so to speak) was only achieved by incorporating other materials besides copper/bronze into the monetary system like hemp and silk cloth or, as often true since the late tenth century AD, paper money.
4 Kraay 1964: 89-90, 1976: 322-323, Crawford 1970, Lo Cascio 1981, Howgego 1990, Scheidel 2008, 2009.