– August 24, 2018

Money is something everyone thinks they know something about. We think about it a lot, and most people think they know what it is. However economists will tell you that the question “What is money?” is actually one of the most difficult and tricky in economics while monetary policy is one of the most slippery and troublesome areas of economic policy, where confident expectations are regularly disappointed – on all ideological sides.

Because this is such a difficult topic, and because the effects of money and monetary policy are so difficult to capture in mathematical models, it is seriously undertaught and understudied on university courses.

For most students thinking about money means trying to understand the macroeconomic effects of interest rates on demand, while things such as the quantity and circulation of money, the demand for it and where it comes from are largely ignored.

This state of affairs has a number of sources (one for example is the obsessive focus on things that can be modelled, often with spurious claims to accuracy). A basic cause, however, is lack of historical perspective and understanding on the part of many (though not all) economists.

Most economists have a ‘Just So Story’ about the origins of money, first set out by Carl Menger. According to this money appears as a neutral medium of exchange that resolves the difficulties in barter due to mismatched wants: if you have a surplus of shoes and I have a surplus of corn but neither of us wants what the other has then trade becomes difficult. However if both of us are prepared to take a particular commodity that can be exchanged for anything (money) then everything works smoothly.

This is an elegant story but there’s one problem: the empirical evidence doesn’t support it. All of the evidence we have suggests that money arose as circulating debt claims, as a number of anthropologists have pointed out.

The Functions of Money

However that is only the start of the problem. Economists argue that money has four functions: medium of exchange, store of value, unit of account (the measure in which prices are calculated), promise of future payment. The story just recounted makes medium of exchange the original and in some sense the primary purpose of money, whereas the anthropologists argue that it was actually a promise of future payment that came first. Be that as it may the real problem is this: the assumption is that there is a single thing called money that has four functions. The problem is that the historical evidence contradicts this.

In many historical instances we have had distinct and different kinds of money for the different functions. Very often one of these will meet more than one function and there are always close relations between the various kinds of money but they are still distinct.

In Imperial China for example the medium of exchange money was ‘cash’, small copper coins of little or no intrinsic value that were minted in huge quantities (several billion a year at times).while the store of value money was taels, small silver ingots issued by private silversmiths with an assay stamp on them.

Cash was used as the unit of account in the form of ‘strings’ of cash. A ‘string’ was a thousand copper coins on a piece of string threaded through the square hole in the centre of the coins. Prices were expressed in multiples or fractions of strings. In addition, at various times there was paper or ‘flying’ money. These were circulating promissory notes, sometimes issued by wealthy individuals, often by the government. These were a circulating form of promises of future payment and were a way of monetizing either the credit of private individuals or of the state.

We can tell similar stories about the later Roman Empire where there was a gold store of value currency (the solidus) and a medium of exchange (the denarius) while prices were often calculated in sesterces (the sestertius had originally been a small denomination coin but was used this way after it was no longer minted).

This happened because the different functions required money with different qualities. With medium of exchange money the main concern was that it was easy to produce to a standard quality, easy to use, and widely accepted. The last took place mainly because of ‘network effects’ – the more people accept a medium of exchange money the more useful it is.

Store of Value

Once a kind of money has got over a critical level of acceptance it will be widely adopted for this purpose. Crucially, it does not matter for most people if the exchange money experiences mild inflation (the rate at which is exchanges for goods depreciates) because people seldom hold on to it for long. Inflation only becomes a problem if it is high and above all accelerating.

Store of value money however is different. Here even mild inflation undermines the whole purpose of this kind of money. It is held for long periods of time and used to purchase investment goods or to pay off large liabilities. In the first case a steady decline in the rate at which the money exchanges for goods is a serious matter.

For this reason store-of-value money has typically been specie, precious metals. These last physically and retain a constant or even improving rate of exchange for goods. In systems like the Roman or Chinese, where exchange money that has been accumulated can be converted into store of value money, the rate at which this is done depends on the inflation rate of the exchange money.

The other point is that the quantity of store of value money tends to grow slowly (because of the difficulties of mining precious metals) and tends to grow in line with the growth of productive activity as a whole. What is often difficult is to combine the two functions in one currency: if the exchange function is paramount, the currency will be subject to inflation while if the store of value function is paramount the currency will be ‘hard’ and will often not grow rapidly enough to meet demand for money, so deflating the economy and reducing economic activity.

Credit Circulation

In the modern world however it is the fourth kind of money, circulating promise of future payment, that has become predominant. Basically this is credit. Promise of payment money depends for its value on peoples expectations about the future and about the creditworthiness of issuers. If you believe that the growth in the future will be sufficiently high that enough wealth will be produced for the promise of payment to be cashed out, ultimately, in goods then people will accept it. This makes future payment money (credit) both enormously powerful and very fragile.

Credit is powerful because it works as a time machine. If a bank advances credit it creates money, based on the promise of future redemption. If the money created is used for investment this means that future income created by an investment has been brought into the present to actually make that investment. A real ‘by your bootstraps’ operation in fact. In the recent world a lot of investment is done this way, rather than by forgoing consumption (saving up exchange money) and converting it into store of value money or actual capital goods. Modern finance has enabled a much more rapid rate of growth and much higher levels of investment, via this time machine.

However the problem is that the future is uncertain, and people in groups can become unreasonably bullish and optimistic about the future. Even worse, governments can manipulate the way that the risk of the future not working out as expected is calculated: that is they can keep interest rates artificially low. In those cases too much future payment money will be created and when the expected growth is lower than anticipated or even non-existent a lot of it will become worthless.

Moreover, while creating future payment money for investment can be very beneficial, doing so for consumption is even more risky. This can also be very beneficial but the chances of having too much created are high. In that event you will have many consumers burdened by debt which they took on thinking their incomes would rise or their assets would increase in value and which they now cannot service if these things have not happened. That again makes the future payment money worth less than it was.

Types of Money

The conclusion from history is that today we should think very seriously about emulating the ancient Romans and Imperial Chinese and have different kinds of money. One would be a circulating medium that eased exchange and economic activity. Another would be a store of value money that retained its value and was somehow tied into the growth of productive activity as a whole (GDP growth perhaps).

There would be a way to easily convert accumulated exchange money into store of value money but the rate at which this happened would reflect the inflation if any of the exchange money. The would be a unit of account which could be either of the first two or some completely independent unit that was imply used for accounting purposes, to calculate relative prices.

We would still have future payment or  ‘flying’ money but the supply of this would not expand as much as it has recently. Moreover most of it would be interest bearing and the interest would reflect people’s collective expectations about the future. (These could still be wrong but would be much less likely to be wrong that the expert consensus or the views of politicians).

Right now, we have a disordered and dysfunctional money system. This poses very acute risks and dangers that because of the nature or money can have system-wide effects. We should look at how monetary systems have been reformed in previous times, and also rethink the way money is commonly taught and thought about.


Stephen Davies


Dr Steve Davies, a Senior Fellow at AIER,  is the Head of Education at the IEA. Previously he was program officer at the Institute for Humane Studies (IHS) at George Mason University in Virginia. He joined IHS from the UK where he was Senior Lecturer in the Department of History and Economic History at Manchester Metropolitan University. He has also been a Visiting Scholar at the Social Philosophy and Policy Center at Bowling Green State University, Ohio. A historian, he graduated from St Andrews University in Scotland in 1976 and gained his PhD from the same institution in 1984. He has authored several books, including Empiricism and History (Palgrave Macmillan, 2003) and was co-editor with Nigel Ashford of The Dictionary of Conservative and Libertarian Thought (Routledge, 1991).

Books by Stephen Davies

Get notified of new articles from Stephen Davies and AIER.