1. Our Money System - most textbooks get it wrong

Some readers may be wondering why there is even a chapter on money in this book at all. Surely there is nothing to it. Surely money just exists and circulates in the economy and occasionally the government will print more of it... the end. Well, if you're one of those people then you're in for a big surprise. Other readers may have come across explanations describing how bank deposits are lent and re-lent in a sequence and the total quantity of money (the "money supply") is then capped according to a formula known as the "money multiplier". Those readers will be in for a surprise too because that explanation is badly wrong. The workings of the monetary system are plagued with misinformation with disastrous effects on the economy.

It may seem perfectly natural to start a book on economics with a chapter about money. Surprisingly though, most textbooks appear to treat the way money works as a detail on the periphery of economics. Indeed many people that have taken degrees in economics will scarcely have learned anything about the nature of money at all, and to make matters worse, when students do get taught about the monetary system they are regularly given false or misleading information. This was highlighted by the Bank of England in 2014 when they published a long overdue paper that was quite damning of undergraduate teaching. The paper contains sentences like:

The reality of how money is created today differs from the description found in some economics textbooks.
the money multiplier theory [...] it is not an accurate description of how money is created in reality.
the relationship between monetary policy and money differs from the description in many introductory textbooks.
operates in the reverse way to that described in some economics textbooks.

This paper is not the only indication that the textbooks have things wrong. Indeed as far back as 1980, Nobel prizewinning economist Milton Friedman said of our monetary system:

"It's a process that, even today, few bankers understand."

The full details of the monetary system would comfortably take up an entire book on its own, so the description given in this chapter is necessarily a simplification, but it is generally compatible with the Bank of England's 2014 paper even if it does use less technical language.

Everlasting tokens and spendable IOUs

It is useful to consider for a moment how most people think money works… It is natural to assume that the government prints it (either on paper or electronically) and then it circulates in the economy forever more. We could label this kind of money "everlasting tokens". Some people might point out that everlasting tokens can't really exist simply because notes and coins wear out but we will assume for simplicity that notes and coins are indestructible. The reason the money can be thought of as everlasting is simply because there is no legal constraint on how long a note or coin can be used for. If this was indeed the complete story of how our money worked then we could conclude that the total amount of money that existed would either remain constant or, if the government created some more, it would increase. And we could certainly be sure that the total would never decrease, it would be a one way process of expansion.

In the real world everlasting tokens do indeed exist in the form of notes and coins, but in recent decades they have come to constitute only a small fraction (less than 5%) of the money that circulates in the economy. The vast majority of the money we spend is a second type with the strange characteristic that its total amount can fluctuate both up and down in a way that is not under the direct control of the government. This type of money can best be described as "spendable IOUs." The concept of a spendable IOU may sound strange, and in order to explain it, we should first consider the behavior of an ordinary IOU, the kind people could create themselves:

Say that Mick wished to borrow £10 from his friend Jim. He can write “IOU £10” on a piece of paper, and hand that to Jim in return for a £10 note. It is important to appreciate that the IOU did not exist previously. Mick did not need to obtain the IOU from somebody else. It was just created at that instant with pen and paper. The IOU that Mick created now has value. It acts as a record or proof of the loan and could be referred to if Mick tried to claim the loan never took place. But despite its value, Jim would find it almost impossible to purchase anything with it.

When Mick later repays the loan, Jim should no longer keep hold of the IOU because Mick will no longer owe Jim any money so Jim should return the IOU to Mick. The IOU has now done its job. It has no further purpose and Mick should now dispose of it. Indeed Mick would be positively keen to destroy the IOU because it could cost him money if it was lost or stolen and got into unscrupulous hands.

As you can see, the life cycle of an IOU has three stages

Life cycle of a conventional IOU

It gets created out of nothing. It did not exist previously, it did not need to be obtained from somebody else.
It has value but is usually not spendable.
When the loan is repaid, the IOU has no further purpose and expires out of existence.

Now let us go back a step and consider the fact that the IOU was not spendable in a bit more detail. The reason the shopkeeper would not accept the IOU as payment for goods was because he didn’t know anything about Mick’s creditworthiness, in all likelihood he won’t even know who Mick was. If the shopkeeper by amazing coincidence happened to know Mick, and he thought he was a reliable person, he may have considered taking the IOU in payment but in general this would be an extremely rare event.

So the spendability of an IOU is dependent upon whether shopkeepers have confidence that they could easily get the everlasting tokens from the person that issued the IOU. Now consider the spendability if the source of the IOU was not an individual but a large well known organisation with a reputation to maintain. The likelihood of the IOU being accepted as payment is now higher than in the case of most individuals but still pretty low. If a shopkeeper regularly accepted these kinds of IOUs then they would be left with the onerous task of contacting all these various organisations asking them to honour their IOUs. Even if they all promptly paid up it would still be a bureaucratic headache and certainly not worth doing for small payments. So now imagine that the organisation was actually a bank... now the likelihood of acceptance will become higher still. Shopkeepers are likely to assume that banks will honour their IOUs, what's more, shopkeepers will generally have accounts with banks and banks can handle the bureaucracy of processing IOUs from other banks. Now this type of IOU could certainly be spendable. Indeed, in the real world, virtually all businesses accept these as payment. This type of IOU is so universally accepted that the vast majority of people do not even realise that what they are accepting are IOUs at all. People are so comfortable and used to accepting IOUs from banks as money that they rarely see any need to actually seek the everlasting tokens, the notes and coins, from the issuer. They simply hold on to the IOUs in their bank accounts and use them for themselves to buy other things. People just assume that bank IOUs are simply an electronic representation of some homogeneous thing called "money". Little do they realise that money comes in such radically different forms.

Let us pause for a moment to recap:

IOUs from individuals have value and so could in principle be accepted as payment for goods and services but it's exceptionally unlikely.
IOUs from banks can, and are, easily be accepted as payment for goods and services.

Why would a bank write an IOU?

So now we have established that an IOU from a bank could, at least in theory, function as a form of money but this begs the question - why on earth would a bank ever write out an IOU? Surely banks have loads of money! Surely banks can do everything they want to do with good old fashioned everlasting tokens. They could take everlasting tokens from depositors and use that to lend out to borrowers, as simple as that. The reason banks will write out IOUs is because it allows them to employ a trick to make more "loans" than the sum total of their current stock of everlasting tokens. You will notice that the word loan was in quotes and that is because when a bank employs this trick it is not really making a loan at all. The trick banks use instead is more accurately described as an IOU swapping arrangement.

If you wanted to "borrow" say, £1000 from a bank, the bank will ask you a series of questions to assess your creditworthiness, and then get you to sign some sort of loan agreement. This loan agreement is essentially an IOU you create with your signature and are handing to the bank. It is your promise to pay the bank some money (plus interest) in the future. What the bank will give you in return is is usually not everlasting tokens. What they give you, will be a cheque book or debit card, which represents an IOU from the bank.

Your non-spendable IOU is exchanged for a spendable IOU from the bank. And just like any other IOU, the bank’s IOU did not exist previously. It did not have to be found from somewhere else. It was just created on the spot with ink and paper or a few clicks on a keyboard. It did not come from the deposits of other customers. And also just like any other IOU, it will expire back out of existence when the loan is repaid.

If we consider money as anything that is widely accepted in exchange for goods and services, then these spendable IOUs are most definitely money. So we can now say that most money is created when loans are made by banks and disappears back out of existence when those loans are repaid. At any one moment, thousands of people in the economy will be creating money by taking out new loans, whilst thousands of others will be destroying money by paying back loans. This idea is so shocking to many people that it is worth repeating but this time in the exact words of the the Bank of England in 2014:

"Just as taking out a new loan creates money, the repayment of bank loans destroys money."

We can visualise this creation/destruction dynamic as follows: imagine a bathtub with water simultaneously flowing in from the tap and out through the plughole. The flow into the bath from the tap corresponds to new loans being made, while the water flowing out through the plug hole corresponds to existing loans being repaid. The amount of water in the bath (plus a very small quantity of everlasting tokens) corresponds to the total amount of money in the economy, the 'money supply'. So the money supply depends critically on the relative rates of flow into, and out of, the system. The consequences of this unstable dynamic on the economy are absolutely enormous (more on that later) and yet few economists and even fewer politicians are even aware of it. Most university-level economics students graduate with little clue that the money supply has this behavior..

If you only remember one thing from this chapter, it should be this diagram and the sentence "Loans create money, loan repayments destroy money." This is far and away the most important idea for understanding our current economy.

The spendable IOUs from banks are often referred to in the textbooks as "credit." This word is misleading because people instinctively feel that only borrowers have anything to do with credit. Little do they realise that even if you have no borrowings at all, when you buy your weekly shop at the supermarket with a debit card, what you hand over at the checkout is in fact credit. Credit is created by borrowing, but thereafter will circulate in the economy between people that may have no borrowings at all. Using the word credit for this type of money is also a misleading because the word credit can mean many things other than money. Another phrase often used to describe this type of money is "demand deposits", so called because the bank customer that has this type of money in their account can demand to have their money in the form of notes and coins (everlasting tokens) at any time. This phrase is certainly better than the word credit because it is at least unambiguous. There are a number of other phrases that you may come across to describe this type of money including "sight deposits", "bank money" and "checkbook money". The general public will generally be unaware of the true nature of spendable IOUs and will think of them simply as the money in their bank account.

Don't banks pay interest themselves?

When people hear this explanation of our monetary system, they are sometimes skeptical. This is because they thought that when a bank lends say £1000, they must be borrowing £1000 from and paying interest to, somebody else. Therefore the idea of banks creating money out of nothing must surely be false. But there is in fact no contradiction between the idea of banks paying interest and this explanation, both statements are true. To see why, we need to look at a bank "loan" again, and this time follow up what happens to the money:

For the sake of simplicity, let us imagine for a moment that there is only a single bank in town. Let us also imagine that someone wants to borrow £1000 to buy a second hand car.

Step one is to go to a bank, and get £1000 of freshly created spendable IOUs.

Step two is making the purchase, i.e. the buyer swaps his spendable IOUs for the car….

Now the car seller has the cheque or debit card payment.

Step three is the car seller storing the £1000 of spendable IOUs in his own bank account. The car seller is now lending his money to the bank and will be expecting his bank to pay interest on that money.

We now end up with at stage four, where the bank is earning interest from the car buyer and is simultaneously paying interest to the car seller. The bank is now making its money on the difference between the two rates of interest. In the case that there is more than one bank, things become more complicated, but the gist will remain unchanged.

So it is true that banks make their money on the difference between what it earns in interest making "loans" and what it pays in interest to depositors - and it is also true that banks create money out of nothing. There is no contradiction between the two ideas.

Making payments between banks

Imagine that Anna who banks at bank A wants to pay a cheque for £100 to Bill who banks with bank B. The starting situation is shown in figure?

In order for this to happen, bank A needs to modify its records so as to reduce the amount of spendable IOUs in Anna’s account and bank B correspondingly needs to increase the amount of spendable IOUs in Bills account. The new situation is shown in figure?

As far as Bill and Anna are concerned the transaction is now complete, Anna has transferred her spendable IOUs to Bill. But things are not settled as far as the banks are concerned. The problem is that bank B is now lumbered with an obligation to pay £100 of everlasting tokens to Bill should he demand it. Simultaneously bank A is relieved of its obligation to Anna. Left in this state Bank B is out of pocket. Bank B will only agree to this arrangement if bank A transfers £100 of everlasting tokens to it as settlement. This will now complete the transfer to the satisfaction of all Anna, Bill and the two banks.

It is useful at this point to consider exactly how bank A can send everlasting tokens to bank B. So far the only type of everlasting tokens we have considered are notes and coins. So one might assume that banks are sending lorry loads of notes and coins to each other to settle the transfers of spendable IOUs between their customers. This would be a very clumsy way of doing things and the banking sector have worked out an alternative. What they have done is created an entirely separate type of money known as "reserves" specifically for this purpose. Banks do not directly have reserves themselves but instead have reserve accounts at the central bank, rather like how individuals have accounts at high street banks. So for bank A to send £100 of reserves to bank B it has to effectively ask the central bank to make the transfer.

Bank reserves are a much misunderstood form of money. There are innumerable articles which state or imply that a bank can directly lend out its reserves to customers. This is simply not true. It is impossible for members of the public to own bank reserves because they cannot have accounts with the central bank. Only banks can have accounts with the central bank. Even financial institutions like pension funds are not allowed accounts at the central bank. This mistaken idea about reserves being lent out to the public is so widespread that Paul Sheard, Chief Global economist at Standard and Poor's rating agency was moved to write a paper entitled "Repeat After Me: Banks Cannot And Do Not 'Lend Out' reserves". This is not some popular article for the public but instead one aimed at professionals in the finance sector. The article was needed because of the number of economists that are simply unaware of this.

How can banks get by with so few reserves

You may remember that earlier we said that the vast majority of the money being in the form of spendable IOUs and only a fraction (less than 5%) being in the form of everlasting tokens. But this raises a potential paradox. If a transfer of £100 of spendable IOUs between bank customers requires a transfer of £100 of reserves, then surely we need equal amounts of both types of money in the system or banks may find themselves short of reserves to enable those transfers.

The resolution of this apparent paradox is all to do with the sheer size of banks and the fact that money will usually flow in both directions between them. If banks A and B are both large, then during the ebb and flow of money transfers between thousands or millions of individuals in any one day, it is likely that the amount of money being transferred from A’s customers to B’s will be matched by a rather similar flow of money from B’s customers to A’s. What the banks do is to simply keep a running total of the transactions and then pay the net amount of everlasting tokens that is required to balance the transactions at the end of the day. This net amount will in general be only a small fraction of the total amount of spendable IOUs managed by either bank. The fact that the total reserves in the system are only a fraction of the total money supply is the reason the monetary system is known as "fractional reserve banking"

So banks do need to have some, but not much, reserves in order to operate. Some reserves to allow transfers of money between its customers and customers of other banks, as well as some notes and coins to satisfy its customer's (generally small) demands for cash.

Bank runs

So now we know that banks do not need a very large amount of reserves in order to pay other banks. But there is a situation where the flow could become decidedly one-way: this is known as a 'bank run'. If bank A is perceived to be at risk of going bankrupt then this can quickly become a self fulfilling prophecy as customers rush to transfer the money in their accounts (the spendable IOUs) to other banks. Now bank A will start having to transfer much larger sums of reserves to the other banks without much flow in the opposite direction, and because they have so few of them, they will soon run out altogether. Customers that have been slow to exit will be unable to transfer their spendable IOUs to other banks, because the other banks will not accept them without the corresponding reserves. Customers will then be left with IOUs from bank A which cease to be spendable. Bank A will now be deemed bust and its remaining customers stand to lose their money.

Notice that this situation is not necessarily due to any fault of the bank. Even an unfounded rumor of a bank run can trigger a real bank run! This precarious situation is a built-in feature of fractional reserve banking. In order to reduce the likelihood of bank runs, governments often offer to guarantee some or all of the spendable IOUs in people’s bank accounts. Though this of course means that the taxpayer has to cover the cost of any bank failures.

A capping mechanism that doesn’t apply

It is commonly stated in textbooks and popular explanations that there is a mechanism that puts a ceiling on the money supply. The so called the 'money multiplier' model which is based on something called the reserve ratio.

This reserve ratio is the ratio of the amount of everlasting tokens a bank has (notes and coins plus reserves in their account at the central bank) as a fraction of the amount of spendable IOUs the bank has on it's books. The explanations state that governments set a legal limit on the size of this ratio. Say that, for example, this limit was one tenth. This means that the amount of spendable IOUs a bank could create would be ten times as much as the amount of everlasting tokens they possessed. At first glance this regulation would appear to put an upper limit on the amount of spendable IOUs that a bank can create. However, as was pointed out by William R White, former deputy governor of the bank of Canada

no industrial country conducts policy in this way under normal circumstances.

In practice central banks will always allow banks to obtain new everlasting tokens (in exchange for other assets) whenever they request them. So in practice, the money multiplier concept is rather like stating that a child was limited in his spending to whatever was in their piggy bank, but omitting to mention that the parents will add whatever extra money the child desires to the piggy bank when asked. So the a reserve ratio could put a ceiling on the money supply if central banks refused to create more everlasting tokens, but in practice it does not. And just for good measure, the reserve ratio limits don't apply at all in many countries, including the UK. So the multiplier model is doubly wrong. As professor Charles Goodhart, former member of the Monetary Policy Committee of the Bank of England once said of the model:

"it should be discarded immediately"

Other mechanisms to influence the money supply

Central bankers are almost continuously inventing new mechanisms to influence both the size of the money supply and the likelihood of banks going bust in an economic downturn. For example the Basel regulations (versions I, II and III), sometimes paying interest on reserves to encourage certain behaviours, other times not paying interest on reserves to encourage other behaviours, lending quotas and many many others. If we were to describe the current set of regulations it would probably be out of date in a few short years. There is a natural tendency for the rules to get relaxed during periods where the economy appears to be running well and tightened in the aftermath of financial crises. The frequency of changes is a clue that the system is fundamentally unstable.

Is interest repayable?

Some people, when presented with information about how our monetary system works, become concerned about where the supply of money for interest payments could possibly come from. They say things like: “If the total size of the money supply was fixed then there is no possible source of money for paying interest.” This may be followed by: “So this proves that the money supply is forced to increase forever, otherwise borrowers could never pay the money back.” They suspect that a proportion of borrowers must, by simple mathematics, be unable to repay the interest. This view is quite widespread and has been stated in numerous popular articles and videos. This view is however, mistaken.

You can get a clearer idea of what is going on by breaking down the problem in several steps. Let us start with a simplified model economy. In this model the only money that exists is the everlasting kind, notes and coins and the total money supply is fixed. We also start off considering what could happen if there was no interest payable on loans. This is illustrated the diagram below. The diagram shows "the economy" on the right hand side consisting of households and industry. One can imagine the money paid as wages flowing from industry to households, being matched by an equal and opposite flow of money from households to industry as people buy goods. We are using the term "wages" here quite loosely to include any form of earning. For example business owners will get much of their income directly from the profits of their companies in the form of dividends. The diagram also shows the flow of loans from banks to the economy and an equal and opposite flow of loan repayments. In this scenario it is clear that it is possible to reach an equilibrium state where we have equal and opposite flows. For example we could have £10billion of wages being paid each day and £10billion of goods being purchased and £1billion of loans being made each day matched by £1billion of repayments (of earlier loans) each day. In this equilibrium there is no need for any injection of new money, the existing money supply can continue to flow indefinitely.

Now let us consider what happens if we make the more realistic assumption that there is interest to pay on loans. This is illustrated in figure below. The figure is actually incomplete and this appears to be the model that is in the minds of those that believe that interest is unrepayable without a continuous growth in the money supply. In this diagram we have a problem. If the rate of creation of loans is matched by the rate of repayment then how can we have an additional flow of interest payments? The flows in this diagram can not reach equilibrium.

The resolution of this apparent paradox is the realisation that the banks do not accumulate the interest payments in some huge vault so that the bankers can gleefully count their money at the end of each day. Instead that money is the sources of the bank's income. The bank's need to pay their staff, rent their buildings, buy their computers and pay dividends to their owners who will want to buy their yachts and Ferraris. All these outflows of money are heading right back into the economy. So the full picture is better described by the figure below. Now an equilibrium can be reached with the interest payments being an equal and opposite flow to bank spending. No paradoxes to resolve, no new money required.

The fact that there is no in-built mathematical paradox to avoid when paying back interest does not necessarily guarantee that all loans will be paid back and the system is always in perfect balance, far from it. If someone borrows a large sum on the basis that they expect healthy future income to repay the interest there is always scope for things to go wrong. They may lose their job, or they may become ill, perhaps their business plan was flawed and the product they are making proves unpopular. Any of these problems may lead to a situation where the loan repayments are larger than the borrower can ever reasonably pay back. This is possible even if the loan was interest free. This diagram is intended to illustrate more that the system can balance and be in equilibrium not that it always will be. There are many problems caused by fractional reserve banking, some of them severe as we shall see later but inherently unpayable interest is not one of them.

Banks buying things creates money

Very few people know that banks create money when they make loans and only a fraction of them know that the repayment of loans destroys money... but now for some even less well known information - when banks buy things it creates money. This idea is mentioned incredibly rarely outside of the most technical of banking documents. Few professional economists know this fact. One place it is mentioned is in the book Where Does Money Come From in which it says

..."the bank creates new money when it buys assets, goods or services on its own account, or pays its staff salaries or bonuses."

Another mention is in the Mcleay et al paper Money Creation in the Modern Economy where it states:

"Deposit creation or destruction will also occur any time the banking sector (including the central bank) buys or sells existing assets from or to consumers, or, more often, from companies or the government."

To see why this is so let us consider the process of bank A purchasing a new laptop for one of its employees for £1000 from a local computer store. For simplicity let us imagine that the computer store business itself has an account at bank A and just before the purchase has taken place, the store has £50,000 (i.e. spendable IOUs) in its account. How can the bank pay the store £1000? Theoretically it cold send a bank employee round with £1000 in cash but as you might imagine this us unlikely. We already know that the bank can not give reserves to the store so the only option left is to credit the store's bank account with £1,000 of spendable IOUs. These are IOUs from the bank. The bank can not get those IOUs from anywhere else, that would make no sense at all. Instead the bank simply creates them afresh in the process of crediting the store's account making the store's account go from £50,000 to £51,000.

At this point some readers may may be wondering - if banks can buy things with freshly created spendable IOUs then what is to stop them going crazy and buying everything under the sun. We need to point out the fact that even though banks can create spendable IOUs with a few taps on a keyboard, these IOUs are a liability not an asset. They are a bank's promise to pay. So let's say that the computer store owners wished to spend their newly gotten IOUs on some new display cabinets and the display cabinet company had an account with a different bank B then bank A will have to hand over £1000 of reserves to bank B in the process. So the ability to buy stuff with your own spendable IOUs is not a free lunch.

As you might imagine, and as the McLeay quote suggests, when banks sell things, money is destroyed.

Quantitative Easing

Shortly after the global financial crisis of 2007/8 the central banks in many western countries started a process called quantitative easing, or QE for short. The phrase quantitative easing was all over the media at the time and was almost always (inadequately) described as "money printing", often followed up with the (vague and misleading) words "to boost the economy". So now you have just learned about the money system it seems the perfect point to tell you about the true meaning and true effects of quantitative easing but before we can do that we need to consider what circumstances would lead central bankers to begin QE in the first place.

The build up to QE

QE is most likely to be implemented after the bursting of a housing market bubble. We will be going into the reasons why the housing market even has bubbles and why they burst in later chapters but for now suffice it to say that a housing bubble corresponds to a dramatic rise in real estate prices followed by an equally dramatic fall. One of the biggest bubbles of recent times was the one in Japan in the early 90s. See figure below.

In the run-up to the bursting point, huge amounts of money is being lent out (created) for the purposes of buying ever more unreasonably overpriced houses. This corresponds to a strong stream of water flowing into the money supply bathtub discussed earlier and there will be a strong flow of money going down the plug hole due to the correspondingly large mortgage repayments. Then when the bubble bursts, there is a collapse in the enthusiasm for new mortgages. So the flow into the bath from the tap suddenly slows to a trickle and here's the key part - the flow down the plug hold does not suddenly fall. Mortgages often last for decades and so the rate of flow down the plug hole remains strong. It should be obvious that in the absence of any special intervention, the bath will begin to empty. The money supply will begin to fall.

A falling money supply is a painful economic environment. People have less money to spend and so naturally sales of almost every kind of good or service will slow, causing all sorts of economic problems and people start asking "why does there seem to be less money around?" A common belief is that banks and or rich people were somehow hoarding it. Little do they realise that the real reason is that money is literally disappearing. One might imagine that it might be possible for producers to simply lower their prices and so perhaps the economy can simply adapt to the falling supply but in practice this is very hard to coordinate. Imagine large companies up and down the country asking their workers to accept pay cuts... tricky. Also consider all the thousands or millions of supply contracts between companies where one is agreeing to supply some item or raw material to another at a set price - they would all have to be torn up and renegotiated. The problems are so great that a falling money supply is generally a form of economic vandalism. Companies will go bust, people will be made unemployed - a general disaster. This is exactly what happened in the great depression in the 1930's after the stock market bubble burst, during which the money supply fell by around a third.

Controlling the money supply under normal circumstances

The mechanism of choice for central banks to influence the money supply has long been to adjust interest rates. The interest rate acts to encourage or discourage potential borrowers from requesting new loans. So in a situation where it looks like the money supply will rise, central banks will likely raise rates to discourage borrowing, slowing the rate of flow of water into the bath. Conversely, in a situation where it looks like the money supply will fall, central banks will likely lower rates to encourage borrowing increasing the rate of flow of water into the bath.

Lowering rates to encourage borrowing works just fine when the reduction in borrowing enthusiasm is small, but if the reduction is more dramatic it should be noted that zero is the lowest interest rate that can be charged, and if that still fails to encourage sufficient borrowing then something more radical will need to be done...

Counteracting a fall in the money supply - helicopter money

Imagine you are in charge of the central bank and you are aware of the falling money supply problem. How do you counteract this fall?... Bearing in mind that the fall in the money supply is general, i.e. everyone will feel the effects of this fall then one could reasonably argue that the central bank could create the money and give it to everyone. The concept of a central bank giving money to everyone is known as "helicopter money" because it is as if the money was literally thrown out of helicopters on to the population. At this point some readers may be wondering to themselves - if the general population are not allowed to hold central bank reserves, how is even possible for central banks to give anybody any money. One possible answer is to literally print the notes and coins and give them to people but the more practical and likely answer is that the central bank gives people electronic money in their own banks accounts indirectly. Say for example a central bank wanted to give an individual (let's call him Tom) £1000 for whatever reason and say that he had an account with a high street commercial bank, let's call it Acme bank. What the central bank would do is instruct Acme to add £1000 to Tom's bank account. This corresponds to Acme creating £1000 of fresh IOUs. Of course Acme will not want to do this without compensation, so the central bank will at the same instant, also credit Acme's own account at the central bank with £1000 of new reserves. The new IOUs created by Acme correspond to an increase in the money supply of £1000. The £1000 of reserves in Acme's account at the central bank do not count as an increase to the money supply because reserves are not spendable by the general population.

Note that the spendable IOUs created through this mechanism will not expire. Because they were not created through any loan arrangement. There is no point in the future where anyone will go to their bank and say "I'm now giving back that £1000 I owe you". Helicopter money appears a perfectly reasonable way to expand the money supply and has been taken seriously and discussed at the highest levels, including by heads of central banks but it has not actually been put into practice in this way to any significant degree.

Central banks choose quantitative easing instead

An alternative to giving money away to counter a falling money supply is for a central bank to create new money to simply buy stuff but now you need to consider what kinds of things they could buy. They could not reasonably buy physical goods like cars and washing machines because then they would be lumbered with a colossal pile of things in a gigantic warehouse. This is a non starter. What they may do instead is to buy government bonds.

Jargon buster: "Bonds", "principal", "coupon" and "maturity"

Bonds are a particular mechanism for borrowing money that is often employed by large corporations or national governments. Say that they wanted to borrow £10 million for 5 years. They know full well that whoever might consider lending them the money would require some compensation in the form of interest. What they could do is print up a piece of paper with something like the following words "We promise to pay the bearer of this document a sum of £10 million in five years time. Additionally we will pay the bearer 3% at the end of every year" and then offer that document for sale. In this case the £10million is called the "principal", the term of 5 years is called the "maturity" and the 3% is called the "coupon". Obviously the size of the coupon has to be large enough to tempt investors. People that purchase bonds like this may hold on to them

The purchasing of large quantities of government bonds by a central bank has come to be known as quantitative easing or QE. It is commonly described as "printing money" and indeed it is true that QE creates new money but the printing money description is misleading because it implies that the money enters the economy and will remain there forever more. If one needed to explain QE in one sentence it would be better to describe it as "the temporary creation of money with built-in compulsory un‑printing at a future time". The following description will show why this is so:

Say that the central bank wants to purchase £1million worth of bonds from a pension fund called Investo Inc. Investo will have an account with a private bank, say Acme Bank. In order for the central bank to purchase the bonds it will want Acme to create £1million of spendable IOUs and assign them to Investo. The only way that Acme will agree to this is if the central bank simultaneously gives Acme £1million of reserves.

So there you have the money creation part of QE - a process the appears rather like "money printing" as repeatedly put forth in the popular press. What is missing however is any consideration of what happens to the government bonds over time. The issuers of the bonds (the government) are still obliged to pay both the coupon and principal on those bonds to the bond owners owners, i.e. the central bank. When the bonds mature and the principal is repaid, the money used to pay the principal is destroyed, shrinking the money supply in the process.

If the central bank wishes to maintain the original increase in the money supply, then every time some of the bonds they are holding mature it will have to repeat the QE process and purchase more bonds to compensate. This means that, in the absence of any new bubble, it is very problematic for a central bank to stop QE.

What about the coupon payments on the bonds? What happens to the coupon payments is more obscure, indeed during the research for this book, many senior economists were asked and none were confident of the answer. Months later a news story broke that the UK chancellor, George Osborne, had "found" a new source of money that could be used by the government. The story revealed that indeed even the central bank was not quite sure what to do with the coupon payments! The money had thus far, been deposited in a dormant account the Bank of England where it took no part in the economy. It was as if it was simply being buried forever, equivalent to destroying it. Somehow the government and the central bank between them had decided that the coupon payments could contribute to the profits of the central bank and therefore be used by the government (central banks profits are almost entirely handed over to governments). This idea was thought so hard to explain to the general public that even the, normally serious, BBC Newsnight television program resorted to describing what had happened as the chancellor George Osborne finding this new source of money "down the back of the sofa!".

This decision to allow the government to spend the coupon payments was significant because it effectively means that any government borrowing via bonds that are held by the central bank are interest free.

Extra detail: Is QE in the US different?

The description of QE given here so far is how it works in the UK. The Bank of England have said explicitly that they purchase bonds or shares from non-banks (utilising a private bank as a middleman) and demand deposits are created in the process. In the US however, the Federal Reserve (the US central bank) have stated that they are buying bonds and shares directly from banks. The Fed can buy things from banks simply by adding reserves to the bank's account with them. Remember that reserves are not spendable by the general public and so are not counted as part of the money supply. So it appears like American QE does not create new money. However, the quantities of bonds and shares that the Fed has purchased are far higher than the amount that banks would ordinarily hold. So what has happened is that as each round of QE has taken place, the banks have made use of their newly added reserves in order to buy more bonds and shares and as already discussed, when banks make purchases it creates new money. In effect then the banks end up acting as middlemen in the US just as they are in the UK and so American style QE creates new money just as it does in the UK.

For a fully worked out example, see here.

Is there an alternative to fractional reserve banking?

Many readers will have been surprised by the odd nature of our monetary system and may be wondering if money inherently needs to work this way. The answer is no. A monetary system known as full reserve banking simply means exclusively employing everlasting tokens and not allowing banks to create spendable IOUs. This means that the total money supply is automatically fixed unless the government decide to print more. A consequence of this system is that when you go to a bank and ask to borrow say £1000, the bank will have to actually check that it has £1000 of everlasting tokens, available from depositors, to lend out. This stunningly simple idea is how most people would assume that banks work already! We will put forward arguments as to why full reserve banking would be preferable to the system we currently have but the arguments will require more concepts from upcoming chapters, so this will have to wait until later.

Money FAQ

Q: Where's the gold?

A: Gold used to be an integral part of the monetary system up until 1971. But since then the use of gold has been abandoned. Banks no longer have any commitment to hand over any prescribed amount of gold in return for money as was previously the case.

Q: Can you define money?

A: Defining money is rather like defining a paperweight. Almost anything can be a paperweight, but some things are better than others. A dead fish can be a paperweight, but perhaps not a very good one! Similarly with money. Almost anything can be money, but some things are better than others. Given this fact, instead of a hard and fast definition, we should instead say the following: The degree to which something can be considered money is the degree to which it is widely accepted in exchange for goods and services. Just to test out this definition, lets consider two things that sometimes cause disagreements, namely demand deposits and gold. Demand deposits (spendable IOUs) are most definitely money in as much as they are very widely accepted in exchange for goods and services. Gold however is rather poor quality money. Just go down to your local supermarket and see if you can pay for your weekly food shopping with a small piece of gold. If you got very lucky you may succeed but your chances are slim.

The most common definition of money found in textbooks is actually a combination of three factors, 1. a medium of exchange, 2. a unit of account and 3. a store of value. So you may be wondering why 2 and 3 have not been considered here. The answer is that 2 is simply a side effect of 1, i.e. if something is regularly used as a means of purchasing things then it is a virtually certainty that is will naturally become a unit of account. Point 3, the store of value, is not even absolutely true. For most currencies in most of history their value decreases over time. It is truer to say that the value of a currency should not fall too fast. But again this is already expected if the currency is regularly used to buy things. In summary, if 1 is true, 2 and 3 will naturally follow and there is little sense in making 2 and 3 part of the definition.

Q: Do all loans create money?

A: No. Only bank loans create money. Some institutions (for example peer-to-peer lenders) can lend pre-existing money and those loans do not create new money. These institutions may be barred from calling themselves a “bank” by regulators, even though what they do may in fact correspond to a dictionary definition of a bank.

Q: Are notes and coins also IOUs? After all, on a UK ten pound note for example, it says “I promise to pay the bearer on demand, the sum of ten pounds”

A: The reason it says these words is a hangover from earlier times when a note was indeed an IOU from the central bank for gold or precious metal coins. These days a banknote is not an IOU for anything other than another identical banknote! So it is not really an IOU at all by any reasonable definition. Indeed dollars and euro banknotes carry no such promise despite their monetary systems working in essentially the same way. For a more academic discussion of this issue see Central Bank Money: Liability, Asset, or Equity of the Nation? One of its conclusions is the following "CBM [central bank money] cannot be characterised as a liability of the central bank because the central bank, in fiat money systems, is under no legal obligation to do anything other than recognise that it has indeed issued the CBM, so that the holder of CBM cannot ask for repayment of CBM in anything other than CBM."

Frustratingly the more widely held view is that notes and coins are a liability of the central bank which makes very little sense.

Q: Where is the discussion of double entry bookkeeping with assets and liabilities?

A: It is extremely common to discuss the monetary system in that way, indeed there are people that think that our modern monetary system is defined by double entry bookkeeping. However, double entry bookkeeping is merely an error checking mechanism for transactions that could already be recorded in other ways. Double entry bookkeeping is a convenient way of recording IOU swapping arrangements of the type seen in bank "loans". If there is any confusion or doubt over any issue, you should first consider and understand the underlying transactions involving IOUs and everlasting tokens before any attempt to work out how to record those things with double entry bookkeeping.

Q: Is bitcoin money?

A: Bitcoin is poor quality money because it is not widely accepted in exchange for goods and services but it would be wrong to say that it it is simply not money at all because there are most definitely some things you can buy with them including illegal things sold on the dark web. Criminals love Bitcoin.

Q: What about Modern Monetary Theory (MMT) is that correct?

A: Modern Monetary Theory is set of ideas about the economy but with particular emphasis on the workings of our monetary system. It is by no means mainstream but in recent years seems to have gained a small army of devout followers. In my opinion MMT contains an odd mix of true and false assertions. For more detail click here.

Did you like contents of this chapter? If not click here.