The bank competition myth

Australian banks are upset. Their $30 billion per year gravy train of profits from the Australian people is finally being slowed down.

A levy on bank liabilities of 0.06% annually was announced as part of the 2017 Federal government budget, and is expected to raise about $1.5 billion per year, or 5% of bank profits.

To be clear, the banking system is a regulated cartel. Its primary function is to provide a public good in the form of the money supply of the country. As such, we would expect it to be uncompetitive, and use tight regulatory controls to ensure that the privileged position of private banks is not being abused.

In my book, Game of Mates, I explain that the result of this uncompetitiveness and lack of adequate regulation in Australia is that over half of the banks’ profits can be considered economic rents, which could be taken back with better regulation and shared with the public at large.

I want to use this blog post to explain in detail the underlying administrative mechanics of why any modern banking system is necessarily uncompetitive.

The first thing to know is that banks do two things. They make money by extending loans, which expands the money supply; a function that is an essential public service in a growing economy. Second, they settle obligations between parties both within their own bank, and between banks, which is another essential public service.

But letting private entities simply make money is risky. So our central banking system constrains the private banking system by making the banks settle payments between each other with a different currency held in accounts at the central bank. In Australia these are called Exchange Settlement Accounts. Every private bank in the system must have an account at the central bank so that they can do this second function of settling payments.

By controlling the second function of banks by making them use a currency controlled by the central bank, it indirectly controls the former function of money creation. No one bank can rapidly create new money by writing loans faster than the rest of the banks. If they do, when the borrower deposits the money created into an account at a different bank, like when they use the loan to buy a house from someone who banks with another bank, it will require the originating bank to settle this payment flowing from their bank to a different bank with their central bank money.

This process reduces their net asset position and increases their costs. They can’t continue to do this. What limits their rate of money creation through new loans is how fast other banks are creating money and transferring central bank money to them. Each individual bank is constrained in their money creation function by their settlement function.

Keynes wrote as such in his 1930 Treatise on Money:

…it is evident that there is no limit to the amount of bank money which the banks can safely create provided they move forward in step.

The words italicised are the clue to the behaviour of the system. Every movement forward by an individual bank weakens it, but every such movement by one of its neighbour banks strengthens it; so that if all move forward together, no one is weakened on balance.

The Australian bank data shows this process in action. Below are two graphs. On the left is the size of the loan book of Australian banks. There is a clear concentration here and a surprising regularity to the trends at all banks. To show these trends more clearly, on the right is the monthly growth of the loans made by the four major Australian banks. As you can see, there is no sustained deviation by any banks from the core growth trend. All banks are moving lock step, as they should.

The whole point of a central banking system is that the growth rate of loans for all banks in the system will quickly equalise. If you are a small bank, this means you can never grow abnormally fast in order to gain market share by competing for loans with the larger banks.

Any central banking system is therefore, by definition, unable to be competitive.

In Game of Mates, the solution proposed to stop the economic losses from the abnormal profits of the protected private banking cartel is to let the central bank itself offer basic low-risk lending and deposit functions directly to the public. Because it has the ability to create for itself its own central bank money, it is the only entity that can grow faster than the existing banks in the system.

Of course, the reality is that the solution would be a far greater hit to bank profits than the small levy proposed. In fact it would likely take back over $20 billion per year in profits from the private banks, which would be shared with the government through its profits on banking operations, and with its bank customers through lower costs. If the banks are upset about a levy of just $1.5 billion a year, they are going to really crack it when they hear this proposal!

*This proposal is actually widely called for by economists, and the idea can be mostly attributed to Nicholas Gruen. See here for example. 


  1. A bank of the Commonwealth acting as a public service, everything old is new again.
    Such an entity (govt public service) can interface with the public, and the Treasury can take back the ‘issue of money’.
    End the RBA.

  2. DreadnotMEMBER

    Thanks Cameron for a real explanation of how banks operate. Unlike the MSM that, in general, give the public the swill that a private bank is a simple intermediary between deposits and loans. What I don’t as yet fully get is the role of wholesale international funds in private banking. Why borrow internationally if banks create money?

    • Yes Dreadnot I would like an answer to this question also.

      Does the RBA or APRA tell the banks to finance some of their loans from borrowing or is it something to do with their capital ratios?

      • Trade balances are a major driver in external borrowings, also the rate of growth of their books can be ramped faster and hence more short term profit. The problem I would argue is what that external funding actually supports productive venture or speculation and inflation in the real economy (i.e not the CPI figure).

  3. The Australia Post outlets are already there for a branch network to beat the pants off any other bank.

    • Crocodile Chuck

      Yes, this is what Ahmed Fahour was supposed to implement @ Post when he came on board in 2009: a simple giropost offering for deposits, funded by buying AUS Treasuries.

      The Big Four [ABA] nobbled the idea-can’t have any ‘unfair competition’, esp. if its a public good!

  4. the cartel nature of the banks is probably the reason why the housing bubble has run so long/gone so high and also why there really is no credit cycle in this country.
    Imagine you are the CEO of a big bank. Your assets are disproportionately weighted to mortgages. And you know all the other banks are in the same position. You also know that the big 4 basically are the mortgage market. So there is no incentive to reduce your exposure to mortgages (the way Goldman did). Which leaves the only alternative of doubling down on mortgages knowing the other banks will follow. I think this is why credit and housing is so out of step with everything else and why the bubble still hasn’t burst.

  5. Actually, this is not entirely true. Given we have a reserve banking system that demands a certain amount of capital be held against loans, banks could opt to cut their dividend and retain this capital within the business. This would add to the Tier 1 capital base and allow them to grow loans faster, relative to the others who continue to pay dividends.

  6. There is also a limit to the amount of money banks can create as a whole as well as an individual limit.
    That is the amount of deposit “money” the public wants to hold as an asset given rates of return available elsewhere.
    Exceed that limit and the money returns to the banking system either as non-money term liabilities or as debt repayments.
    This is because demand for new loans moves in a different direction to the demand for bank money as an asset. Keynes also made this point when he described banks as performing 2 separate and ultimately unrelated functions (paraphrasing), providers of resources on the asset side and providers of money on liability side. When credit demand is strong people don’t usually want to hold their wealth in money. Which means overall money creation isn’t actually driven by lending; it’s driven by the portfolio decisions of the public to hold deposits v other assets and the banks’ willingness to buy assets from the public in exchange for deposits.
    Imagine a scenario where overall deposits grow strongly, overall lending shrinks materially. How is this possible. Very easily, deleveraging increases demand for money; loan repayments exceed new loans, people sell securities to the banks in exchange for deposits. The banks sell securities to the CB in exchange for their deposits at the CB. Money has grown because the bank and non bank public wanted more of it, lending has shrunk because people wanted fewer loans. This is exactly what happened in the US post QE.
    Apologies for long winded response; but the notion that overall lending drives deposit creation is a gross over simplification. Deposit demand (as asset) and banks willingness to buy assets (not lend) for deposits is the driving force behind overall deposit creation.

    • Gra ManMEMBER

      Sweeper, so what your saying is that the CB can create money by buying securities from banks. That money replaces money lost by deleveraging.
      QE = print and buy

      • No not money lost. If the commercial banks are able to and the CB willing to, broad money should grow during deleveraging. As both banks and households/business want more money. This didn’t happen during the Depression part partly due to the Feds response leading to the pro-cyclical debt-deflation dynamic Fisher described.
        Money lost in loan repayments is more than offset by new deposit creation via the banks buying safe assets from the public, lowering their yields encouraging a further move into money. In effect the commercial banks respond by rearranging their own portfolios away from new loans into safe securities and reserves held at the CB. So deposits and new lending move in opposite directions which seems counter-intuitive; but not if you conceptually split the banks in half and see them as performing two separate functions or meeting two unrelated demands.

      • Gra man IMJ you are not far wrong. The RBA prints into the overnight market to lower rates. That’s the mechanism. That’s where i differ a little from pfh. Private Banks don’t create the money. What we get is the RBA announcing an IR decrease and pouring money into the market to make it happen. You then get increased demand for money which the banks supply – that supply coming from some combination of RBA and offshore money.

        Note this is necessarily a simple assessment of a complex topic!

      • Flawse,
        It is a complex topic. However MMT does not shed any light on it at all.
        There are multiple constraints to deposit creation, at least 4:
        1. the individual banking one (profitability) mentioned in the article
        2. the desired proportion of reserves the banking system as a whole want to hold v deposits. this is where the CB becomes important. If the actual ratio fall below desired ratio, banks try to borrow in the inter-bank market or from the CB, bidding up the target rate, unless the CB provides more reserves. And contrary to MMT the CB doesn’t just provide any quantity of reserves at a fixed rate over any period.
        3. And importantly, the qty of deposits the public actually want to hold as an asset v other assets given other rates of return available.
        4. leakages into currency and foreign assets.

      • Following up from sweeper and flawse, this is definitely a topic where a whiteboard that shows flows would clarify .

    • Hi Sweeper. Can you help with Dreadnot’s question above? ‘What I don’t as yet fully get is the role of wholesale international funds in private banking. Why borrow internationally if banks create money?’
      Also , when you are talking about ‘money’ do we need to differentiate between M1 and M3? A term deposit is an interest bearing asset but are we differentiating between it as money and a government bond? Is it relevant that bank reserves are neither of these, being part of the money base?
      Lastly when you talk about the banks buying assets from the public, are these bonds that are owned by private individuals?
      I like to think that I understand this stuff better than most but still find there is a big gap that I hope you can fill.

      • Because there is a limit on the pool of saving and on the proportion people want to hold as money. If you hold money, you are giving up return.
        Banks can create deposits, but if a positive return is available either through debt repayments or term (non-money) financing, it returns straight back to the banking system. Meaning banks cannot profitably increase their lending this way.
        If the rate of return on other assets is lower overseas, it becomes profitable for the banking system to expand lending by borrowing offshore. And banks know this in advance of lending. Thus why the Treasury function is such an important part of banking – retaining deposits, take deposits from other banks (since the pool is capped), selling bonds etc. If it wasn’t banks would be pure risk managers on the asset side.

      • davie
        One SMALL issue not mentioned here is the Current Account Deficit. So there is a certain amount of money that does NOT return to banks deposits as money circulates. So this piece from Keynes (…it is evident that there is no limit to the amount of bank money which the banks can safely create provided they move forward in step.”), like a fair bit of his tripe, is factually incorrect. ( In the US case this perhaps applies as long as the US is the world reserve currency)

        More important than the CAD might be the ‘marginal propensity to import” resulting from any increased money supply. In Aus case our imports are something like 23% of GDP (from memory) and, despite there not seeming to be any work at all on this, one would guess our marginal propensity to import might be about 40%. So if a Bank creates ‘new money” about 40% of it ”disappears” before it comes back as deposits. (Summarily speaking) As long as the CAD exists Banks have to borrow foreign currency to plug that gap.
        Note this is a dynamic loop not a straight line thing. Theorists, as per the MB view, have it that Banks borrow money offshore and lend it into the local housing market. On the other hand they argue Banks can create loans which create deposits. When the later happens the money goes into the economy that results in 40% (?) of it being spent on imports for which we have to borrow foreign currency.
        In reality this is all happening at the same time.

      • Flawse,
        That passage from Keynes is literally right at the start of a 2 volume six hundred page Treatise where he later goes on to explain in detail how there is a real constraint on the qty of bank deposits the banking system as a whole can create. Yes in the first 10 pages or something he admits banks create deposits by buying assets (something he says isn’t controversial among academics). However he then he spends the remaining 500 something pages explaining why this doesn’t mean what “the most disinterested body of persons in the world, the Army of Heretics and Cranks” think it means. He explains in detail, that deposit creation by the banking system as a whole is not merely a residual of bank lending. Stressing that lending and deposit creation are two separate often contradictory services. Something which MMT’ers constantly miss. I’ve tried to provide examples below. It’s fair to say that that passage in isolation doesn’t come close to reflecting the point he was trying to make in the Treatise.

      • Hi Sweep
        I accept what you say.
        I guess i have a bit of an issue with a few of his statements . His statement that that he could see no reason for a price on ”çapital’ as there was no limiting factor on capital seemed to me to be a bit MMTish. Anyway best not complicate this discussion – I should have bitten my tongue!! 🙂

    • Sweeper
      Re the MMT debate – Perhaps we live in Oceana and it IS past 1984!!!
      Orwell’s nation-state of Oceana, the INGSOC party members were made to believe that two plus two equals five.

      • Yeh. Feels as though we are arguing against a divine maths – similar to the theory of the Trinity as 1 + 1 + 1 = 1 – where the parameters keep changing.
        ‘No I think it = 3. Your wrong because banks don’t lend reserves’

  7. Diogenes the CynicMEMBER

    I have always thought that Aus Post should become a bank but mainly for the transfer payments system backstop. To me it is a national security issue having your economy’s financial pipes owned and operated by private entities, if they have an IT issue then your payments system is down, if they fail by making poor decisions then your pipes clog up. The only issue with making AusPost into the full bank is the Govt would be tempted to privatise at some point for short term gains…

  8. fyi. that passage from Keynes is at the start of his Treatise on Money.
    Further on he expands on the dual and contradictory nature of banks (as I referred to above). ie. deposit creation is not a residual of lending.
    See Vol 2 Bk 7 of the Treatise on Money.
    “In so far as they are cash-deposits, he is acting both as a provider of money for his depositors, and also as a provider of resources for his borrowing customers. Thus the modern banker performs two distinct sets of services”
    He notes that the 2 contradictory roles are performed only due to historical accident – due to the creation of representative money replacing commodity money.
    “The dilemma of modern banking is satisfactorily to combine these two functions. As a purveyor of Representative Money, it is the duty of the banking system to preserve the prescribed objective standard of such money. As a purveyor of loans on terms and conditions of a particular type, it is the duty of the system to adjust, to the best of its ability, its supply of this type of lending to the demand for it at the equilibrium rate of interest, ie. at the natural rate…
    the complete attainment of one of its duties is sometimes incompatible with the complete attainment of the other, those in control of the banking system have to make up their minds which object is to prevail, or, if neither is to prevail, to achieve a just compromise between the two”.

    Then in a sentence which could describe MMT perfectly:
    “A partial selection of some amongst these truths and a blind eye to others of them have led to the opposed points of view which are characteristic between them of the vast bulk of non-academic monetary literature. On the one hand, there are the bankers, who at least maintain a certain sanity of demeanour by holding tight to rough rules-of-thumb which they have learn’t from experience. On the other hand, there is the most disinterest body of persons in the world, the Army of Heretics and Cranks whose numbers and enthusiasm are extraordinary”

    It is uncanny how well that describes MMT: “the most disinterest body of persons in the world, the Army of Heretics and Cranks whose numbers and enthusiasm are extraordinary”

    “It has been a principal object of this Treatise to give a clear answer to these perplexities. What is the true criterion of a creation of credit?.. We have found the answer to lie in the preservation of a balance between the rate of saving and the value of new investment”

  9. Note the Reserve Bank role in ‘setting’ interest rates. In lowering rates the RBA tips (prints) money into the overnight system which allows the Banks to supply the increased credit demanded at the lower rates. Again as this new money circulates we, in this economy as it is now structured, get increased imports which must be paid for. These are paid for either by foreign borrowing or sales of assets to foreigners.
    This idiocy is able to maintain itself as long as our Banks can borrow offshore at lower rates. If overseas rates were to rise substantially the two ends of the circle would not meet.

  10. Cameron, would Australia get a better outcome by removing any support for banking?

    I remember back pre-GFC, banks typically achieved ROE of 16% to 17%, reflecting returns necessary in a more risky environment. This was achieved in recognition of the risks taken by banks when lending money. More recently, with some support to reduce financial exposure, ROE has fallen to 12% to 14%, which I guess is a result of lower interest rates and tighter margins that deliver an acceptable return given the lower risks involved with social support provided.

    Surely, removal of that support would return banks to seeking higher ROE and thus delivering a more expensive service, which would be a negative for customers and shareholders? Is the implied backing of government a small price to pay (ie nothing, so long as bad debts are kept in check) to achieve better outcomes for savers and borrowers?

    Ultimately, banking relies on trust. If all depositors and lenders to banks lined up to take their money away, banks would collapse. They rely on trust and if that trust comes with the backing of a big brother, what is the problem? So long as there are prudential and regulatory controls, reducing the business risk has got to be a win-win I would have thought? Please tell me otherwise.