Saturday, 3 May 2014

Bank lending does not create money.

To understand why, it is essential to distinguish between two quite different services performed by commercial banks (combanks). One is to supply customers with day to day transaction money. The second is grant long term loans.
To effect the first, a combank will just credit £X to a customer’s account or tell them they can become overdrawn up to £X. Let’s assume the latter. Normally the combank requires collateral of course.
Now assuming the customer is ONLY AFTER transaction money, then they’ll be in credit about as often as they are overdrawn. That is, money has been created, but there is no long term loan involved.
To illustrate with a simple example, if the ONLY service performed by combanks was creating transaction money, then simple maths dictates that no customer would on average over the year be in debt. Reason is that for every customer going into debt for a short while, there must be another customer in credit. (I’m ignoring physical cash, BTW, to keep things simple.)

Genuine loans.
In contrast to creating transaction money, combanks connect lenders to borrowers (second of the above mentioned services performed).
Now individual combanks can simply credit the accounts of borrowers with money produced from thin air, but taking the economy as a whole, it’s a plain physical impossibility (never mind the economics) for one person to spend more than they earn unless someone else or some other group of people ABSTAIN FROM SPENDING i.e. save an equivalent amount of person hours of work, or person hours of blood sweat and tears, to put it another way.
A exception to the latter point is where an economy had spare capacity: i.e. where the above extra economic activity (say building the above house) did not exacerbate inflation. But I’ll assume the economy IS AT CAPACITY to keep things simple.

Shades of grey.
There is of course no sharp dividing line between money and non money. Second there is no sharp dividing line between current spending and capital / investment spending. E.g. is purchasing something designed to last one year a form of current spending or is it investment spending?
A third grey area is thus. There is no sharp dividing line between someone making a long term loan to a bank (e.g. putting money into a term account) and maintaining an stock of transaction money at a bank larger than they really need.
And a forth grey area is that combanks when granting a loan do not rely solely on savers who have deposited money for an extended period (e.g. in term accounts). That is, combanks indulge in a bit of maturity transformation: i.e. using money in current or checking accounts (transaction money) to fund long term loans.
But none of those grey areas detract from the basic point I’m making, which is:
Combanks create money when they create day to day transaction money, but that does not basically involve long term loans. While in contrast, long term loans do not involve money creation: reason being that those loans are basically funded from a stock of money which is not going to be spent in the near future, and which is thus more in the nature of a long term loan to the bank than a form of money.


  1. Thank you for this post. Forgive me for my ignorance on the subject but doesn't this oppose what the endogenous money people say? So combanks can create money through loans if the economy is not at full capacity?

    1. That’s right. I’m suggesting that if an economy is at capacity (and assuming the amount of day to day transaction money that people want remains constant) the any money “loaned into existence” by combanks ought to be classified as a loan rather than money. But of course there is no sharp dividing line between a loan to a bank (e.g. money in a term account) and genuine money (i.e. money in a checking account).

    2. Thanks for your response. One more quick question to highlight my ignorance. Can you give historical examples of economies actually operating at full capacity?

    3. By "full capacity" I mean something like "maximum level of economic activity compatible with acceptable inflation". Or if you like, "level of activity that corresponds to the Non Accelerating Inflation Rate of Unemployment". Of course that level of activity does not mean everyone is employed or that every plant, factory and machine is working at capacity - far from it. It's just the maximum feasible level of activity.

    4. Thanks again! So when economists calculate the 'output gap' I'm assuming they have the non-accelerating inflation definition in mind for potential output given the impossibility of full capacity taking place in the literal sense? Also, regarding the non-accelerating inflation definition, is it wrong to say that we really can't be too sure what inflation will be in a specific situation until we actually find ourselves in that specific situation? Thanks again!

    5. Correct. NAIRU is a valid and useful theoretical concept, but there is plenty of disagreement as to exactly what level of unemployment corresponds to NAIRU. A bit like the square root of minus one in maths. As I understand it, mathematicians cannot do without it, but no one can envisage it - at least speaking as a non-mathematician, I can't.


Post a comment.