the things being traded. But more worryingly, it was clear that the
fei
were not a medium of exchange in the sense of a commodity that could be exchanged for any other—since most of the time, they were not exchanged at all. Indeed, in the case of the infamous shipwrecked
fei
, no one had ever even seen the coin in question, let alone passed it around as a medium of exchange. No, there could be no doubt: the inhabitants of Yap were curiously indifferent to the fate of the
fei
themselves. The essence of their monetary system was not stone coins used as a medium of exchange, but something else.
Closer consideration of Adam Smith’s story of commodities chosen to serve as media of exchange suggests that the inhabitants of Yap were on to something. Smith claimed that at different times and in different places, numerous commodities had been chosen to serve as the money: dried cod in Newfoundland; tobacco in Virginia; sugar in the West Indies; and even nails in Scotland. Yet suspicions about the validity of some of these examples were already being raised within a generation or two of the publication of Smith’s
Wealth of Nations
. The American financier Thomas Smith, for example, argued in his
Essay on Currency and Banking
in 1832 that whilst Smith thought that these stories were evidence of commodity media of exchange, they were in fact nothing of the sort. 19 In every case, these were examples of trade that was accounted for in pounds, shillings, and pence, just as it was in modern England. Sellers would accumulate credit on their books, and buyers debts, all denominated in monetary units. The fact that any net balances that remained between them might then be discharged by payment of some commodity or other to the value of the debt did not mean that that commodity was “money.” To focus on the commodity payment rather than the system of credit and clearing behind it was to get things completely the wrong way round. And to take the view that it was the commodity itself that was money, as Smith did, might therefore start out seeming logical, but would end in nonsense. Alfred Mitchell Innes, the author of two neglected masterworks on the nature of money, summed up theproblem with Smith’s report of cod-money in Newfoundland bluntly but accurately:
A moment’s reflection shows that a staple commodity could not be used as money, because
ex hypothesi
the medium of exchange is equally receivable by all members of the community. Thus if the fishers paid for their supplies in cod, the traders would equally have to pay for their cod in cod, an obvious absurdity. 20
If the
fei
of Yap were not a medium of exchange, then what were they? And more to the point, what, in fact, was Yap’s money if it wasn’t the
fei
? The answer to both questions is remarkably simple. Yap’s money was not the
fei
, but the underlying system of credit accounts and clearing of which they helped to keep track. The
fei
were just tokens by which these accounts were kept. As in Newfoundland, the inhabitants of Yap would accumulate credits and debts in the course of their trading in fish, coconut, pigs, and sea cucumber. These would be offset against one another to settle payments. Any outstanding balances carried forward at the end of a single exchange, or a day, or a week, might, if the counterparties so wished, be settled by the exchange of currency—a
fei
—to the appropriate value; this being a tangible and visible record of the outstanding credit that the seller enjoyed with the rest of Yap. Coins and currency, in other words, are useful tokens to record the underlying system of credit accounts and to implement the underlying process of clearing. They may even be necessary in an economy larger than that of Yap, where coins could drop to the bottom of the sea and yet no one would think to question the wealth of their owner. But currency is not itself money. Money is the system of credit accounts and their clearing that currency represents.
If all