Money Supply Theory

This section looks at the most common explanation of inflation i.e. that it is related to the expansion of the money supply. It reviews the evidence for the theory and explains exactly how it is supposed to work.

If you ask most economists what causes inflation, they’ll probably mention that it is linked to the amount of money in circulation. As an example, they might ask you to imagine a simple world of bakers and brewers. There might be five bakers and five brewers each selling a loaf and a pint of beer to each other every day. Each costs £1 in gold coins. The total wealth of the world is £10. Now imagine what would happen if another £10 in gold coins is found and everyone gets a share of them.

Money supply example

The amount of money in the world has suddenly doubled to £20 but there are still only 5 loaves and 5 pints of beer being made each day. Everyone feels a bit richer but the only thing they can buy is bread and beer. As people think they have more money, they all try and outbid each other to get extra provisions. Very quickly the prices of all the bread and beer will increase until they cost around £2 and a new equilibrium is reached again. £1 in gold coins now only buys half a loaf or half a pint of beer.

That is a simple example of the theory behind the link between the money supply and inflation. In real life, it is a lot more complex and it can take a long time for prices to rise to the new equilibrium and hence the relationship is not always as visible as the theory suggests it should be. However, there is strong evidence that the theory holds over the medium-term as you’ll see below.

History of the theory

Sharp increases in prices have been observed in many periods of ancient history from ancient Babylonia to the Roman Empire. A particularly prolonged period of rising prices throughout the globe was seen in the 16th and early 17th Centuries. (See: “7-Inflationary Wave Theory.”)

At that time, money was defined in Europe in terms of precious metals. Silver and gold coins were issued as the primary medium of exchange. That era was the time of Christopher Columbus and the great discovery of the Americas. The key thing plundered from the New World and brought back to Europe was vast quantities of gold and silver. This made certain countries and financiers much richer as wealth was defined in terms of precious metals. The amount of money circulating in Europe therefore increased.

At the same time prices were rising in Europe and many scholars started to suggest reasons why this was happening. Amongst them was the famous astronomer Copernicus. He was one of the first people to propose the theory that the price increases seen were in fact related to the increases in the money supply.

Copernicus

Copernicus – one of the first theorists on inflation
Source: “Nikolaus Kopernik
Licensed under Public domain via Wikimedia Commons

The key formula

In its most simplified form, scholars such as Copernicus stated that Prices (P) vary in proportion to the supply of money (M) i.e. P ∝ M.
This simple discovery was to have major implications for centuries to come. Indeed, as we’ve seen in the previous chapter, it became part of the very definition of inflation for a long time.

Quantity theory of money

Many developments and variants of Copernicus’s theory have been put forward since. Probably the most notable was in 1848, when John Stewart Mill formally proposed the “quantity theory of money” and produced the “equation of exchange”. He showed that the simple formula is only applicable if both the size of the economy is stable (i.e. GDP is constant) and there has been no change in the number of times money is spent during that period by people i.e. people’s saving levels and spending levels haven’t gone up or down.

The full formula he proposed was thus: MV=PQ.

I will try and translate this jargon into words that might make some more sense. It is saying that at any period of time, the money supply (M) multiplied by the number of times people use that money (V) is equal to the average prices (P) multiplied by the total value of all the goods we produce in that time i.e. GDP (Q).

The difficult aspect of this is the number of times people use money, or the “velocity of money”, as it is known. This is to do with saving and spending levels. The basic idea behind this qualification is that if lots of people save money, fewer people will be buying goods. For a while at least, too many goods will be being made relative to buyers. Sellers therefore will be forced to drop prices as they compete for what business is left. The reverse situation also occurs when saving decreases and spending levels go up in an economy i.e. prices rise.

Evidence for the theory

Proponents of the theory (called Monetarists) can point to many more recent examples than the 16th and 17th Centuries where excessive money printing appears directly linked to rising levels of inflation. The most infamous of these was during the Weimar Republic in Germany in the early 1920s.

Indeed if you examine the long-term rise in inflation in the UK since 1900, it is strongly correlated to the growth in the money supply. The graph below shows an index of UK prices from 1900 and compares this with an index of the UK money supply. (In order to satisfy the equation of exchange, a deduction from the money supply has been made for the extra money required because of our expanding economy, i.e. the index is money supply less GDP*.)

The graph shows that over the last century inflation has fairly closely followed the money supply. However there have been quite long periods when one of the two was slightly higher. Indeed for the last couple of decades, consumer price inflation has been lagging behind increases in the money supply in the UK and this situation must be resolved at some point.

Money supply vs inflation

Sources: Price index – ONS longitudinal series. Money supply – The Bank of England’s preferred measure (M3/M4/M4x). GDP – measuringworth.com.

Some issues with the theory

What the above chart also illustrates is that changes in the money supply are not always directly reflected in the prices of goods and services. Indeed the two can go for many years or even decades in opposite directions. This is partly because of changes in people’s desire to save/spend (i.e. the velocity of money) and partly because the newly created money does not always flow directly into goods and services – see below.

No one tells the population that the money supply has increased. The very first response to an increase in money supply is often for people to save, and so effectively they keep the amount of money in circulation the same. Therefore prices do not immediately change either. As that money spreads further around the economy, more and more gets into circulation. Prices then rise. Furthermore, money supply and velocity of money are often correlated. As money supply increases, velocity of money goes up (and vice versa in a recession) and so the price changes can be bigger than the money supply calculations alone might predict.

To further complicate matters, this process is very much dependent on where the increases in money supply have gone (i.e. into retail prices or into asset prices) and more generally on consumer sentiment and the state of the business climate and cycle. It is therefore not surprising that there is often little short-term synchronicity between the money supply and inflation. This is clearly illustrated in the data from the UK in recent years.

UK money supply

Sources: Bank of England, 12-month change in money supply M4 and ONS annual RPI inflation.

As the above chart shows, there has been little correlation** between increases in the money supply in the UK and retail inflation in the twenty years between 1992 and 2012.

Many economists now acknowledge that changes in the money supply are not predictive of retail inflation over short-term periods or in low-inflation environments. In a recent paper examining the real world correlation between money supply and inflation of all countries between 1969 and 1999, the following conclusion was reached:

This strong link between inflation and money growth is almost wholly due to the presence of high-inflation or hyperinflation countries in the sample. The relation between inflation and money growth for low-inflation countries (on average less than 10 per cent per year over 30 years) is weak, if not absent.

Money supply and asset price inflation

But what these simplistic analyses of monetary theory miss out is the complexity of inflation. To comprehend this, you need to understand exactly how the money supply is expanded. Money is normally created in an economy when private banks make loans to individuals or companies. (See box below: “How money is created.”) Typically this money is first used for speculation or to purchase assets e.g. houses, companies/shares, bonds, commodities (or more complex derivatives of them). It is not normally created to spend directly in the economy on goods and services, although typically it ends up there.
Therefore when the money supply is expanded, as happened markedly during the 80s and the following decades in the UK, it creates inflation in the assets that the money is first used for. This so-called “asset price inflation” is clearly illustrated in the chart below with data from the UK.

Asset price inflatiom

Sources: Price index – ONS longitudinal series. Money supply – Bank of England (M3/M4/M4x). GDP – measuringworth.com. FTSE – Finfacts and SwallowPark. Net Public Debt – ukpublicspending.co.uk . House prices – ONS.

The solid bars on the chart show an index of the money supply (M4/M4x) less GDP growth, i.e. the net increase in the money supply over and above that required by the UK economy during this time. The lines show where that money then flowed to. These depict the index prices of key asset classes such as shares (i.e. FTSE All Share Index), bonds (i.e. Net Public Debt) and house prices. It shows that a series of bubbles were created over this period in these assets. The prices of all of them have kept up with or exceeded the overall increase in the money supply.

The dotted line on the chart shows an index of goods and services prices (RPI) and how the expanded money supply since 1980 has yet to have its full impact in this area. The gap between the two and the implications of this are discussed more in later chapters.

Two separate money supplies? Two separate inflations?

What is key to understand about the quantity theory of money concerning inflation and the money supply is that it affects two separate economies: wealth (i.e. asset prices) and consumer spending (i.e. goods and service prices). Money flows between these two economies. For example when someone sells a house and spends the proceeds in the real economy on goods/services like holidays or health care, they are transferring money from one economy to the other. Similarly when they save their salary in a pension, money is transferred in the opposite direction.

It has been argued that the two sectors obey the quantity theory separately (and over the medium-term in totality too). However economists rarely highlight this issue. This is probably because it is difficult to demonstrate as countries do not normally publish data separately for the money supply in the two areas.

APPENDIX: How money is created

Although many people talk about the money supply in the context of Monetarist theories, comparatively few really understand how money is actually created and how the money supply is increased.

Contrary to popular conception, very little of the money that exists in the UK has been created by the government and very little is in the form of coins and notes. 97 per cent of our money has been literally created from nothing by commercial banks when people or companies have sought loans. Our money is not usually created by the Bank of England, as many erroneously believe (except recently during Quantitative Easing). Mervyn King, the ex-governor of the Bank of England, made this very clear in an article he wrote back in 1994: “In the United Kingdom…money is created by the banking system.”

Moreover in 2014, the Bank of England published a paper clearly outlining exactly how private banks create the money in our economy and debunking the commonplace misconception that people deposit money first and then banks lend it out. (They also outlined how the “money multiplier” theory was also incorrect too.)

What really determines the money supply is the willingness of commercial banks to lend money. This is in part a reflection of their ability to lend within the constraints of their assets and the need to have a certain amount of reserve capital covering those loans. This was why they did not lend as much after the 2008 recession, as their assets have been reduced by bad loans and at the same time regulation demanded that they hold higher reserves.

For more information, visit Positive Money.

* The logic for deducting GDP from the money supply index is a direct application of the theory of exchange which states: change in money supply + change in velocity = change in prices + change in quantity of goods produced i.e. GDP. The graph ignores the effect of velocity of money, as this cannot be measured independently of the other variables. See: Laidler, D. (1985), “The Demand for Money: Theories, Evidence, and Problems”, 3rd edition, Harper and Row. The consumer pricing data was the ONS longitudinal series, which largely mirrors RPI and its predecessor the UK Cost of Living Index. The money supply index used a broad measure of money supply and the one that is the Bank of England’s preferred measure. This was originally M3, then became M4 and is now M4x.
**  For the statistically minded, the relationship over the last 20 years is slightly negative, r2 = -0.09.