Debunking the “three functions of money”

{ This article has also been published on the New Money Hub web site }

In discussions about money, its three functions are often mentioned: “store of value, medium of exchange, and unit of account” — the order of these may vary. This is usually said in a peremptory tone, as being self-evident and not really worth discussing. These three functions of money constitute the tripod on which the accepted wisdom about money rests.

Let us start by noting that the usual approach is to try and define the nature of a thing, before describing its functions and features. It is therefore curious that money would be defined by its functions. The reason is that, since the beginning of modern economics, there has been great confusion about the nature of money — a confusion that has not been definitely resolved until today.

In 1913, Alfred Mitchell-Innes, building on the work of Henri Dunning Mcleod, wrote: “credit and credit alone is money”. With that terse statement, the essence of money was clearly described, the issue seemed ready to be settled. But Mitchell-Innes’s publications were not only forgotten for many decades, a century later the credit theory of money has still not been incorporated in many standard economics textbooks. But these same textbooks keep reprinting the “three functions” meme as a matter of course. Let’s have a look at these three functions from the point of view of the credit theory of money.

Money as store of value

We start with the observation that, generally speaking, debt is incurred with the intention of being repaid. We then note that a large majority of money in circulation is bank money, which is backed by loans. These loans have been taken out by individuals and corporations who naturally wish to repay these loans. And for those loans to be repaid, the money has to be spent.

What happens when the money is not spent? The following scenario provides a clue: if a company took out a loan, which resulted in an issue of money, and the holders of that money refused to spend it, that company would eventually go out of business. This may then cause the bank to default. The money would be lost.

This shows that the spending of money cannot be postponed indefinitely, otherwise the money may become worthless. The spending cannot be accelerated either, because the debtor must be ready to provide goods or services, and that may take time.

What this means is that in a healthy economy, at any point in time the money being spent by creditors has to be roughly equivalent to the productive capacity of those who wish to repay their loans. The dance of credits and debits is one in which both sides must cooperate to achieve optimal results. Money has to be spent, i.e. creditors and debtors must meet, at the appropriate time.

Bottom line: money, being a claim on future work, can indeed be seen as a store of value — but one that is neither absolute nor permanent. Its value actually depends on it being spent continuously at the appropriate rate.

Money as a medium of exchange

The use of the word “exchange” in this context is not really appropriate. One can exchange an apple for an orange, a green marble for a blue one, but money is not about exchange. Money is the principal mechanism by which debt circulates in the economy, it is about debt creation and debt cancellation.

Once issued, money serves primarily as a means of payment, and paying is about settling debts. When paying for goods and services with money, we may not know what proportion of the payment results in debt cancellation, or when that cancellation will actually take place. What we do know is that, as a result of a payment, the overall amount of credits and debits in the system will almost always be reduced. Of course, on any given day a very large number of debt creation and debt cancellation events occur in the banking system, so that the overall volume of credits and debits appears relatively stable.

There is no doubt that money is sometimes used as a medium of exchange. For example, “buying” a used bike from a neighbour does not change the amount of debits and credits in the money system — what happens is a change of creditor. But this is only a secondary function of money.

Bottom line: money is not primarily a medium of exchange, money is primarily a means of settling debts.

Money as unit of account

Here we have a confusion between the unit of measure, and the thing being measured. Money is not the unit of account, it is measured in the unit of account. It is just that we call the unit of account (Euro, dollar, yuan…) with the same name as the money-IOU itself (Euro, dollar, yuan…) That is, by the way, not a modern phenomenon: once could argue that the Roman denarius was also an abstract unit of account, and that the bank balances denominated in denarii were no less confused with the unit of account.

To see clearly what is happening, imagine an ancient country in which the unit of account was the silver shekel — a shekel being a unit of weight. Now imagine a coin with a metal value of one-tenth of a shekel, but on which the distinguishing mark of a bank or a temple was printed, and with the mention “one shekel”. In the marketplace that coin would simply be called a shekel, but the distinction between the money and the unit of account would be much more obvious.

Bottom line: money is not a unit of account, rather money is expressed in a unit of account.


We have shown that, of the “three functions of money”, one is imprecisely stated, another is misleading and the last one is simply incorrect.