The Technological Role of Fiat Money

This article argues that fiat money’s only technological role in an economy is to act as societal memory: money allows people to credibly record some aspects of their transactions and make that record accessible to other people. This recordkeeping role is demonstrated in the three standard paradigms of fiat money: the overlapping generations, turnpike, and search models. In these models, if a new economy is created by removing the money and replacing it only with a historical record of all transactions, known to everyone in the economy, then the original monetary allocation is still achievable as an equilibrium. This article is a less technical presentation of the ideas in the author’s study, “Money is Memory,” which is forthcoming in the Journal of Economic Theory. The article appears in the Minneapolis Fed’s Quarterly Review with the permission of Academic Press. The views expressed herein are those of the author and not necessarily those of the Federal Reserve Bank of Minneapolis or the Federal Reserve System. Imagine that John and Paul meet. John has apples and wants bananas. Paul wants apples but doesn’t have bananas. In monetary economies, this problem is solved by Paul giving John money in exchange for apples. John then uses the money to buy bananas from someone else—let’s say George. If John doesn’t give the apples to Paul, then John doesn’t get the money and can’t buy the bananas from George. This scenario demonstrates that adding fiat money to an economy expands the set of allocations available in the economy. In this sense, money is a technological innovation, just as much as the train or the steam engine is. But, of course, unlike those devices, money itself is intrinsically useless. So, what technological role does money actually play? In terms of the reallocation of intrinsically valuable resources, we can think of the scenario above as a situation in which John is considering making Paul a gift of apples. If John makes the gift, George will give him bananas in the future; if John doesn’t make the gift, George won’t give him the bananas. The money that John receives from Paul and offers to George is merely a way to let George know that John has fulfilled his societal obligations and given Paul his gift of apples. Thus, monetary economies can be viewed as merely large interlocking networks of gifts. If agents in the economy know the history of all gifts, then any allocation of resources achievable in the economy with money is achievable without it. George, for example, can react to different histories of John’s gift-giving in the same way that he reacts to John having different amounts of money. This analysis leads to a simple conclusion: If the function performed by money can be superseded by a perfect historical record of transactions, then money’s only technological role must be to provide that record. In other words, money is a form of record keeping, or societal memory. I make this argument formally by considering the three standard paradigms of fiat money: the overlapping generations, turnpike, and search models. In each of these models, I create a new economy by removing the money and replacing it only with a historical record of all transactions, a history known to all agents. I do not provide agents with any additional ability to enforce contracts. I show that in each of the models the original monetary allocation is still achievable as an equilibrium in the new environment with perfect societal memory. I conclude that the only technological role of money is to provide a (possibly limited) form of societal memory of transactions. My argument demonstrates the vacuity of the three standard explanations of the role of fiat money in an economy: money acts as a store of value, a medium of exchange, and a unit of account. From a technological point of view, we can see that none of these functions really require money. Money does not represent a new way for society to accumulate wealth. Money does not reduce the costs of transferring resources from one person to another. There is no immediate technological reason that money should be a better numeraire than other goods. The traditional explanations for the presence of money in an economy are more descriptive of its functions than explanatory. The true explanation for money’s presence is that money is a record-keeping device. I am not the first to notice that money has a recordkeeping role; anyone who makes an argument along these lines has to acknowledge an enormous debt to the work of Robert Townsend (1987, 1989, 1990). Townsend has studied several environments in which the optimal arrangements feature monetary tokens that help agents remember trading histories. Other researchers [among them, Joseph Ostroy (1973), Robert Lucas (1980), and Rao Aiyagari and Neil Wallace (1991)] have noted, as I emphasize here, that fiat money helps keep track of past transactions. In contrast to those researchers, though, I emphasize the generality and the singularity of money’s record-keeping role. My point is a twisted version of Milton Friedman’s famous dictum: Money is always and everywhere a mnemonic phe-