Loanable funds? The money multiplier? Steve Keen and Richard Murphy dismantle the myths. Double-entry accounting shows what really happens when banks create money and governments run deficits. If economics taught reality instead of dogma, austerity would be unthinkable.
This is the summary of our discussion:
Steve Keen on Double Entry, Economics, and Ravel
“Economics without accounting is like physics without mass.” That, in effect, was the theme of my second podcast with Steve Keen. Our discussion revolved around three questions: why Steve came late to double entry, what that means for economics, and why it led him to create his Ravel software.
The conversation revealed much about the weaknesses of mainstream economics and the possibilities of a very different approach.
1. Why a late convert to double entry?
Steve began with biography. His early degree in arts and law avoided the accounting training that traditional economics courses required. He thought accounting was dull. But when he built his first model of Hyman Minsky's Financial Instability Hypothesis in 1992, the problem became clear. He could model debt, but he could not properly model money.
That nagging absence set him on a path. Teaching at the University of Western Sydney, he encountered students rejecting orthodox economics and found in the work of Augusto Graziani a decisive insight: money is not a commodity, nor just a bilateral promise. It is a triangular relation between buyer, seller and bank. That three-cornered structure means banks are not mere intermediaries but creators of money.
Steve absorbed this, spent time at the Levy Institute alongside Wynne Godley, and began to see the power of stock-flow consistent modelling. Godley's tables suggested a way of formalising money and debt. But there was still a problem: did banks really destroy money when debts were repaid? His instinct was no. The accountants said yes. In time, he realised he was wrong. Bank deposits are money; when debt is repaid, deposits fall, and so the money supply contracts. Double-entry was not an optional technicality. It was essential.
2. The consequence of double entry
I suggested in the discussion that the true gift of double entry is that it forces us to see every transaction as two-sided. Government spending, for example, is not money poured into a black hole. Someone always receives it. Someone always spends it again. There are always multiplier effects.
Steve agreed. Neoclassical economics prides itself on “general equilibrium.” But its models ignore the feedbacks that make an economy real. Double-entry, by contrast, forces integration: if an asset rises, a corresponding liability or equity must rise elsewhere. There are no black holes. The system balances.
This matters because orthodox models of money creation collapse when put to the double-entry test. The “money multiplier,” beloved of textbooks, only works if all lending is in cash—a fantasy in a world of electronic deposits. The “loanable funds” model assumes households lend their savings to firms and government, mediated by banks. In reality, banks create new money when they lend. The accounting shows it.
In short, double-entry is not just bookkeeping. It is the logic of a monetary economy.
3. Money as ledger and flow
I pressed Steve on another thought. Double-entry shows money as a record: every unit of money is just a ledger entry. But money is also a flow, dynamic and constantly changing. It behaves, in some sense, like a quantum particle: it alters when we observe or record it.
Steve's answer was that the key is dynamics. Orthodox economics still treats equilibrium as the natural state, even though the founders admitted it was a stopgap until better dynamic methods arrived. With today's tools, we can model economies as constantly evolving systems of flows. Double-entry is the discipline that ensures those models remain coherent.
4. Ravel: software for real economics
That takes us to Ravel, the software Steve has built. Its foundation is the Godley table: every row must sum to zero, ensuring that assets minus liabilities minus equity is always balanced. From these tables, the program generates systems of differential equations to show how stocks and flows change over time.
What makes this powerful is flexibility. Ravel lets you model the world as mainstream economists imagine it—loanable funds, money multipliers—and then correct it to reflect reality. If households do not in fact lend directly to firms, you can shift the asset to banks, and the system instantly rebalances. Errors are caught; the logic is enforced.
The result is a tool that can model government spending, bank lending, private debt, and financial instability with a realism that mainstream economics avoids. It shows, above all, that government deficits create money. Far from “crowding out” private investment, deficits expand the money supply and enable activity that otherwise would not exist.
5. Policy consequences
This conclusion is explosive. It means that austerity is a self-inflicted wound. It means that Rachel Reeves and others who think deficits are dangerous are damaging the social fabric they imagine they are protecting. It means that rather than slashing the NHS we should fund it; rather than pricing higher education we should restore free universities. It means that climate investment is not something we must “save for” but something we must finance through government spending, which itself creates the money we need.
In short: deficits are not bugs in the system. They are features. That is how fiat money works.
6. Why neoclassicals resist
Why then does mainstream economics resist this? Steve traced the history. Early neoclassicals like Marshall and Jevons admitted that equilibrium was a fiction, a crutch until better dynamic methods arrived. But when mathematicians proved equilibrium was unstable, economists doubled down. They preferred to assume away the difficulty—even to the point of assuming tastes could change—rather than face the need for a paradigm shift.
The result is what Steve called a “religious belief in equilibrium.” What began as shorthand became dogma. Instead of mathematics, it became “mathimagics”—myth, magic, and misplaced faith.
7. Towards a better economics
Our conversation ended where it began: on the need for realism. Accounting is not optional; it is the grammar of money. Without it, economics is blind. With it, we can see that government spending creates money, that deficits are necessary, that austerity is folly, and that dynamic models are the only honest way to understand a capitalist economy.
Steve's Ravel software is a tool to build a better economics and a better economy. It embodies the logic of double entry and the dynamics of flows. It shows what is happening in the real world rather than in the textbooks.
And it leaves us with a challenge. If economics is to serve people rather than myths, it must abandon its equilibrium religion and embrace accounting, dynamics, and reality. That is the task for all of us working for a future that deserves to be funded.
Comments
When commenting, please take note of this blog's comment policy, which is available here. Contravening this policy will result in comments being deleted before or after initial publication at the editor's sole discretion and without explanation being required or offered.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Thank you! Most interesting!
Might current “official economics” also serve the inluential-wealthy few, living in a world of ideologically sustained costly excess, to the cost of the many, most of whom create wealth through their labour and consumption?
1.government deficits create money
2.Rachel Reeves thinks deficits are dangerous
Reeves wants to “grow” the Uk economy, but that implies more money. But this can only be created by gov deficits – which Reeves thinks are dangerous.
The only thing that is dangerous is Reeves & her thinking based on: “mathimagics”—myth, magic, and misplaced faith” – i.e. fantasy.
Good post. How to transform into a story for ordinary people.
Thanks
Sometimes Mike I struggle to follow you (probably me ) but this is beautifully clear and concise.
In short a new equilibrium approach is required in which the emphasis is placed on the effects of two flows – the creation of money and its drainage – whether there is too much or too little in each – not simply treating creation and drainage as one flow!
I don’t feel comfortable with your use of the word “drainage”. It seems to me to encourage the inference that the money – the debt has gone somewhere else.
When effluent goes down a drain it doesn’t cease to exist, it goes somewhere else where one hopes it will be processed appropriately.
When I pay off my loan, the debt – the money ceases to exist.
There’s no afterlife for money.
I don’t mind you suggesting another word. My use is about money existing to do something useful like oil in a car engine then it’s drained off and can no longer do any useful work. What do you propose?
Meanwhile, out on the streets (from FCA Data).
The FCA has published the latest data on the retirement income market, covering the year April 2024 to March 2025.. This data
is collected through the following returns:
• REP015 – retirement income flow data, collected twice a year for each 6-month period from the period 1 April to 30
September 2018 onwards. This return provides data including the number of transferred plans, AUA (assets under
management) in relation to annuity and drawdown products, and number of plan holders taking withdrawals or UFPLS,
or purchasing annuities for the first time (by pot size and by age).
• REP016 – retirement income stock and withdrawals flow data, collected annually at the end of each financial year from
the period 1 April 2018 to 31 March 2019 onwards. The retirement income stock data covers the number of
uncrystallised plans, fully crystallised plans and partially crystallised plans with the withdrawals flow data capturing
information such as the number of plans with regular withdrawals/UFPLS payments (by age group and by pot size).
The key findings are outlined below:
• Total number of pension plans accessed for the first time in 2024/25 increased by 8.6% to 961,575 compared to 2023/24
(885,455).
• Sales of drawdown policies saw the biggest increase from 278,977 in 2023/24 to 349,992 in 2024/25 (25.5%).
• Sales of annuities increased by 7.8% from 82,061 in 2023/24 to 88,430 in 2024/25.
• The overall value of money being withdrawn from pension pots increased to £70,876m in 2024/25 from £52,152m in
2023/24. This is an increase of 35.9%.
• 30.6% of pension plans accessed for the first time in 2024/25 were accessed by plan holders who took regulated advice
(down from 30.9% in 2023/24).
• The number of DB to DC transfers continued to fall from 7,181 in 2023/24 to 6,418 in 2024/25.
Fascinating. Genuinely. Thank you.
Steve Keen was the first speaker on Sunday’s PJP day.
Corbyn asked for the slides for his talk to be on the PJP website, as they were difficult to see on the day.
Spreadsheet modelling is brilliant for the minority that are not turned off by sums. I would really love to see a physical model which demonstrated these flows, simplified as necessary. I can’t do it, but I have known 2 men (both dead) who could have. One, John Conway, built a computer that worked on liquid flows c. 1957. My late husband, Francis Evans, specialised in hands-on engineering models. There are probably some in the hands-on models community who could think of such a model, or eccentric engineers elsewhere.
They are a nice idea – but not very flexible, and also very expensive. So I am not sure who would make them.
Physical models do get ideas across in a visceral way that nothing else does. In the hands-on years, I lost count of the people who played with a model and said “wow, I never understood this before”. And part of the problem is getting to people who only understand money through household finance. It would be a splendid thing to have at next year’s party conferences, for example. I coukld try asking advice from someone still in the hands-on field, if you thought it was a good idea?
Why not?
But the problem is how do you show the destruction of money, rather than the flow of money?
I’m envisaging a collection of tubes and valves, with a quantity of water/liquid which can be supplied from a reservoir, or discarded via a tap at the lowest point. I’ll try to talk to Steve Pizzey about this early next week.
Good luck
Look up MONIAC, or the Phillips Machine. Bill Phillips, of Phillips curve fame, built a physical model of the economy back in 1949, complete with taps, valves, and fluids. Apparently, he salvaged bits of an old Lancaster bomber for his first machine!
https://en.wikipedia.org/wiki/Phillips_Machine
In reply to Schofield on September 24 at 1:11pm
I fear I completely misunderstood what you were saying. Please excuse my ignorance.
I assumed that by “drainage” you meant the opposite of “creation”; so then, why not “destruction”?
But you are talking in terms of flows so; is the creation of money by the setting up of a credit balance in the borrower’s account a flow of money from the bank to the borrower?
And is the reduction and eventual elimination of that credit balance by the transferring of one or more credit amounts to the accounts of one or more other account holders the flow that you refer to as “drainage”? Perhaps that could be better described as transference.
I think this might be where I realise I’m in a hole and should stop digging.
Linda Evans – I would appreciate seeing a physical model but I would also like to see a flowchart (picture) model using numbers. I find that way much easier to understand than words and I think that many others do too. I remember seeing someone replying on this blog saying that.
There is a flow chart in the Taxing Wealth Report 2024. Look at, I think, chapter 16.
[…] Steve Keen on rethinking money (YouTube video) […]