Wednesday, November 23, 2016

Double-Entry and "Quadruple-Entry" Bookkeeping

(This post is not part of the series "New Monetary System From Scratch". This post is my attempt to establish a common understanding of bank accounting among all the discussion participants in Nick Rowe's blog comments here. I present here my own understanding and ask people to let me know if they agree with it or not.)

I have created a somewhat busy spreadsheet to illustrate some basics of bank accounting as I understand them. I suggest you have a look at it before you read the rest of the post. Going forward, I assume, for clarity's sake, that all transactions are recorded in the general ledger. That might not be the case in reality (see explanation here).

My spreadsheet example is adapted to an example Jamie, one of the participants to the discussion, has given in his laudable, easy-to-follow slide set here.

"Quadruple-entry bookkeeping", as I understand it, refers to two different real-life ledgers, or "books". Jamie gave an example in one of his comments:

Forget about creating money for a moment. Imagine the simplest possible economy where only one economic transaction takes place anywhere in the entire economy during an accounting period.

Imagine that the transaction is that my business sells you a bicycle for €100. Let’s look at this from both our perspectives.

Jamie’s perspective: I lose a bicycle and gain €100.
Antii’s perspective: You gain a bicycle and lose €100.

Macro-economic perspective: No gain or loss of either bicycles or money as the changes in Jamie’s position and Antii’s position cancel out. The macro-economy doesn’t care who has the bicycle and who has the money.
Note that there are FOUR position changes in the macro-economy that we must account for even in this simplest possible economy with only one transaction (two changes in bicycle inventories, two changes in money inventories).

If both Jamie and I kept records of our transactions (in a double-entry bookkeeping format), there would clearly be two different ledgers: Jamie's ledger and my ledger.

In reality, Jamie would probably make four entries on accounts in his ledger (see example here). Two of the entries would be related to his income: credit (CR) "Sales" and debit (DR) "Cost of goods sold"). The other two would be related to his assets: CR "Inventory" and DR "Cash"; "cash" includes checking account, if we are in the electronic age. (Notice that these are not the "credit-debit pairs"; those are "CR Sales, DR Cash" and "CR Inventory, DR Cost of goods sold"; see the example I linked to above.)

To simplify things, we can assume that Jamie sells the bicycle "at cost", so that his inventory is diminished by €100. It seems the purpose of Jamie's example is to highlight the asset positions, so we can overlook the income side (zero effect if goods are sold "at cost"). So, two entries: CR 100 Inventory, DR 100 Cash.

In the example, I, Antti, am not a business. That most likely means that I keep records only of my assets – not of my income. Two entries: CR 100 Cash, DR 100 Inventory (of assets/of bicycles).

Jamie's two entries and my two entries together are the four entries needed in "macroeconomic accounting", following the principle of "quadruple-entry bookkeeping", as suggested by Jamie. Looking at the entries, we could conclude that one person's credit is another person's debit, and vice versa.

What seems to remain unclear in the discussion is what happens in bank accounting. What kind of entries does a bank make in its ledger in different situations?

Here are two scenarios:

1. Let's take Jamie's example above and assume that we are talking about checking accounts, not physical cash. As far as I know, the bank would make only two entries in its ledger in this case: CR 100 on "Jamie's checking account" and DR 100 on "Antti's checking account". (Would taking a macro-perspective lead to sextuple-entry bookkeeping in this case?)

2. Nick's dad agrees on a $1,000 loan with his bank (see, again, my spreadsheet example). Outside of the bank ledger, a loan contract and possibly other documents are created and filed. In the bank ledger, two entries are made: DR 1000 on "Nick's dad's loan account" and CR 1000 on "Nick's dad's checking account".

In Nick's blog's comments, Jamie says:
Loans are assets for a bank. Assets are usually recorded in asset registers. I can’t say if banks have a specific name of their register of loans but I am pretty confident that this is correct.

I might interpret him wrong, but this sounds like he's suggesting that there are two different entries made in the bank's ledger to record (1) an asset of the bank ("green loan"), and (2) Nick's dad's loan/debt/liability ("red loan")? This is obviously not the case, as it is clear that the debit balance on Nick's dad's loan account is at the same time (considered as) an asset of the bank and a liability of Nick's dad (two perspectives).

The framework Jamie presents in his slide set doesn't seem compatible with the framework Nick has established by differentiating between green money and red money. In Nick's world, there are no loan accounts. There are only checking accounts that can be "overdrawn", and to Nick a positive (credit) balance is "green money" while a negative (debit) balance is "red money" – both primarily seen from the account-holder's (customer's) perspective. Nick also makes clear that he doesn't consider – in clear contrast to conventional accounting language – "green money" a liability of the bank (Jamie's "red money") nor "red money" an asset of the bank (Jamie's "green loan"). (I think the subject of overdraft and what I call a traditional loan – two alternative accounting techniques – deserve a "deep-dive" blog post. Let me know if you agree.)

I understand that we all use different implicit models when we try to make sense of the world. That's fine. But we all need to be able to tell the model from phenomena (reality). I think Nick does well in this regard, because it wouldn't cross anyone's mind that Nick thinks that the bank really uses red and green bits of paper, moving those between customers' boxes. But when I look at Jamie's slide 8, I cannot tell if Step 1 is supposed to be something that takes place in reality or not. As far as I know, it doesn't. And even if it did, I can't see any purpose in it. This is because the bank doesn't need to "create" or "manufacture" money (or medium of exchange) or loans. This has been understood since the overdraft was invented in Scotland in the 18th century. It was invented when a banker, with the help of a customer, realized that an account doesn't have to have "money" in/on it and still a payment can be made "from" it.

Quite a mess we've got ourselves in?


  1. Antti: You put a lot of work (and thought) into this spreadsheet and post. Thanks.

    You begin with what I think is a serious mis-assumption: "1. Let's take Jamie's example above and assume that we are talking about checking accounts, not physical cash."

    Why might this be serious? Let's build a bank from nothing. Assume ten people get together, decide to begin a bank, and each contribute $100 cash each. The bank begins with $1000 cash.

    Next assume USA rules. Banks are required to deposit reserves of 10% of deposits. Deposits are $1000 so $100 will be sequestered (or sent to the Fed).

    After the reserve deduction, $900 is available for loans. Remember, this $900 is in cash. It is available for transfer to other banks if needed.

    Now assume that this newly formed bank makes a $9000 loan. Wait! It cant't! It only has $900 available for loan! Actually, it can make a $9000 loan so long as the loan deposit remains on the accounts of the bank. (The USA reserve rule is that 10% of deposits remain in reserves. 10% of $9000+$1000 is $1000.)

    The bank's problem comes when the borrower wants to move $9000 to other banks in a single event (such as the borrower wants to start a new bank). There is only $900 cash available. The solution to this problem is for the new bank to sell the loan and transfer the remaining cash to the bank-of-borrowers-choice.

    I think this risk is the reason that Jamie pairs green money with red loan AND red money with green loan.

    Jamie (in his slides) may be presenting both the bank balance sheet and bank ledger. Is that what you are thinking?

  2. Roger said: "Jamie (in his slides) may be presenting both the bank balance sheet and bank ledger."

    I don't think so. That would be a serious mistake from Jamie's side; combining a report with records.

    I would call my assumption a mis-assumption, because the whole point was to look at the accounting. And I don't understand why you take up the "money multiplier" story which I think has been quite widely discredited already? The Fed creates more reserves when they are needed, so the amount of reserves forms no constraint. But perhaps I misunderstood you?

  3. Antti: “Jamie’s … laudable, easy-to-follow slide set”

    As I said earlier, if a model is easy to follow, it allows debate to focus on disagreements on content – which is what is important. If the model is not easy to follow, we never get on to the content.

    Also, someone said something about “architecture” in comments on one of your posts. I think that is a good word. “Strategic” is another good word. We are creating a macro-economic architecture or a strategic view of banking or whatever.

    The “architecture” and “strategic” words tell us to focus on what is REALLY important and ignore everything else.

    Antti: "Quadruple-entry bookkeeping", as I understand it, refers to two different real-life ledgers, or "books"

    That’s not how I think about it. Quadruple entry means that LOGICALLY there are four entries irrespective of by whom / where / when / how / how many times they are recorded.

    Antti: “To simplify things, we can assume that Jamie sells the bicycle "at cost", so that his inventory is diminished by €100”

    Most mainstream economists don’t understand inventory, or they understand it but ignore it in their thinking. I often suspect that they don’t quite understand profit either. Mainstream economists think at the level of Keynes identity:

    I + C = C + S

    so part of the role of an architectural model IMHO is to explain to them how that identity “works” in the real world i.e. success would involve Nick & Co understanding unambiguously how concepts like inventory and profit fit with Keynes’ identity.

    JKH has stated on many occasions that Keynes’ identity is true at all times. I agree. That means that it must be true for every individual transaction in the economy, so we can ask how a single transaction of a given type would impact the identity e.g. buy the parts for a bicycle, manufacture a bicycle, sell a bicycle to a household at cost, sell it at a profit, take out a loan at the bank.

    I doubt that many economists could provide an answer to that challenge, so that suggests a level of granularity that our models would have to meet to help resolve that problem. That is the level of granularity of my model. That is a deliberate decision on my part.

  4. Antti: “What seems to remain unclear in the discussion is what happens in bank accounting. What kind of entries does a bank make in its ledger in different situations?”

    I don’t think that this matters. It’s only the underlying logic that matters. I suspect that is one of the main points where we disagree.

    Antti: “The framework Jamie presents in his slide set doesn't seem compatible with the framework Nick has established by differentiating between green money and red money”

    Yes, that is correct. I put forward my model because I thought that Nick’s models were wrong and he seemed to need multiple models to describe a single real world. My uses of green and red money are analogous to, but different from, Nick’s but I wanted to retain the concepts of green and red because, like Nick, I think that they are useful and easy to understand for non-experts.

    I do think that Nick’s general approach to modelling is one of his strengths in comparison with other mainstream economists. Nick tries to devise simple models complete with explicit assumptions that are easy for non-specialists to understand (if not always, as in this case, to agree with). He also tries to get to the underlying logic. Devising a model that is easy to understand is far more difficult than being a contrarian!! Anyone can be a contrarian. Black is white! Up is down! In is out!

    Antti: “I think the subject of overdraft and what I call a traditional loan – two alternative accounting techniques – deserve a "deep-dive" blog post”

    I think that is too much detail for me when my main objective is to help Nick & Co to understand how their accounting identity fits with actual accounting, and then to explain the same thing to non-economists. It may still be appropriate for your objective but I am not totally clear on what your objective is.

    Antti: “But we all need to be able to tell the model from phenomena (reality)”

    I agree but an accounting system is not reality. It’s just another model.

  5. Antti: “And even if it did, I can't see any purpose in it. This is because the bank doesn't need to "create" or "manufacture" money (or medium of exchange) or loans”

    In Keynes identity, the “I” term is mostly about businesses purchasing raw materials and manufacturing their products to create inventory e.g. a car manufacturer manufactures cars. For a car manufacturer, inventory is created by “real-world” manufacturing. For other businesses, such as a hotel, the inventory is “manufactured” in a computer system e.g. a room is made available for a specific night. I use the term manufacturing for real-world manufacturing and “manufacturing” (in quotes) for the creation of inventory merely in a computer system. In that sense, I model banks as “manufacturing” inventory (loans and money) in their systems. I am trying to create a model that is as consistent as possible across sectors. Even so, banks are still different from other businesses as they create the medium of exchange.

    Also, one of the core concepts of accounting is conservation. In physical manufacturing, conservation is imposed by the laws of physics. In a hotel, conservation is imposed by a combination of the underlying assets and time e.g. if a hotel has 10 rooms, it can only make available 10 rooms each night. In finance, conservation is imposed by forcing financial assets and liabilities to be created in pairs which add to zero (like matter and anti-matter in physics).

    That’s the end of my initial comments. I am happy to clarify any of the comments I have made but I don’t really want to discuss the validity of my modelling conclusions. I believe that my conclusions are appropriate for my models. You can accept or reject what I have done, or use any parts of it (the style and/or the content) as a way of developing your own models. However, I don’t think that it is realistic to agree a common model across the various people who have been contributing to Nick’s blog and your blog, particularly when there is no agreed objective for the models, no common method for modelling and no mechanism for resolving disputes.

    Are you studying for a qualification? Or are you just studying for your own personal reasons (like me)? Are you developing models for any specific purpose / objective like a thesis?

  6. Jamie: Thanks for the long (not too long for my taste), thoughtful comments!

    Instead of providing a long answer touching separately on everything you said, I'll try to provide some kind of (long) conclusion (in two parts).

    I think your attempt at quadruple-entry is laudable, but as you say, our (Nick, you, me, etc) goals might not be common. Or even if we shared the main, overarching goal (which we probably do; we want to understand how the economy or a human society in general works?), Nick and I might not agree with you if you think that the discussion around red/green money calls for a quadruple-entry approach. I can of course only talk for myself.

    You seemed to hint, from the start, that if someone doesn't think in terms of quadruple-entry, then that someone is thinking like a household, not like a macroeconomist?

    What I've been trying to say, from the start, is that one can present a bank's double-entry bookkeeping simply as double-entry -- without explicitly mentioning or presenting how things look from the non-bank customer's perspective -- and still be talking about the economy as a whole.

    I guess we could say that this kind of view recognizes the bank as an intermediary between non-banks, and not an independent actor like Nick or Nick's dad (a household). If the bank is seen like this, we can see that in the bank ledger there are mainly records of liabilities and assets of non-banks, of these "independent actors". From this it follows that the banking system as a whole can be seen as some kind of "macro record-keeper".

    You might think that the banking system isn't really a "macro record-keeper", because it doesn't keep records of real assets of non-banks. And you would be right. But when I discuss the banking system, then I discuss the financial positions of non-banks. I do keep in mind that there are transactions in real goods which usually create these financial positions.

    Enough about quadruple-entry. What I'm going to present in my main series of posts (New Monetary System From Scratch) is much more in line with your thinking, even if it isn't based on quadruple-entry. I take it as a given that Andy and Betty view things from their personal vantage point, and that we could show how their (often imaginary) personal accounting looks like, but I focus on the banking system as the record-keeping device of their financial positions.

  7. To your questions in the end of your last comment:

    I had been worried about the world economy since 2008. I followed the financial markets, and macro commentary, very closely. My intuition said that the massive rise in debt levels around the world is bad, that in some way it is "borrowing growth from the future", and cannot go on indefinitely, and that it will eventually cause massive problems and a world-wide depression down the road (after the can cannot be kicked anymore -- no one knows when exactly). If you're familiar with the discussions that have taken place since 2004, one could say that I shared most of the fears and worries of Bill White (ex-BIS, now OECD). I respect him highly, but at the same time I saw that there are intelligent people who don't seem to agree with him. I wanted to find out why, or more accurately, I wanted to prove "Bill White's case". That was the best way for me to remain sane, because it all looked to me, and many others, like a slow-motion train wreck which we should do something about.

    OK. I needed to understand how debt really works at macro level. It seemed this was where people didn't agree. Disagreement of intelligent experts is usually a sign of a hole in theory. And I found out that in order to understand how debt works, one needed to understand how money works. And reading about that, I was very fast staring at the gaping hole in the economic theory.

    I didn't give up. I had learned some interesting stuff about how debt works at macro level by trying to reason about it, also independently, for many years before I even touched monetary theory. And it seemed that the understanding on debt I had built partly by independent thinking, provided me with a new perspective on the monetary system. That meant that what I read about monetary theory, I took with a grain of salt, all the time comparing it with the model in my head. I did adjust my model, too, but only when I had to. I was also being the best critic of the theory I was building, and more than once I got this horrible feeling when I thought I might have succeeded in killing the theory I had learned to like -- and live with. (I quit my day job in 2014 to pursue my theory full-time. This was right after I had realized that what I have come up with looks like a new theory which has the potential to change the way we see money.)

    So I would say my reasons are first and foremost personal. It had never crossed my mind that I'd want to become a professional, academic economist, or similar. I just wanted to understand how the world works. Overall, I put very little weight on people's credentials.

    Now that I need an income, I'm considering doing a PhD, which might be the easiest way to get an income while doing what I love (in Norway all PhD students get much more than a "living wage"). As I didn't major in Economics, but Accounting, it might be more complicated to get to a PhD program.

    How does this compare with your personal reasons?

    1. I don't have anything to add to what I said in my last post to you on Nick's blog.

  8. You asked about paradigm shifts. A few points.

    There are several of aspects of our discussion that might qualify as a paradigm shift, but I will leave academics to discuss what precisely is, or is not, a paradigm shift.

    A bank used to be a building where we took pieces of paper and coins. Now my bank is a combination of:

    Apps on my laptop and iPad
    Plastic cards I use to buy things in shops and over the internet
    Android Pay on my mobile phone
    An occasional telephone call.

    This change will continue. For example, I can now get higher interest rates on my savings by peer-to-peer lending, which cuts out the bank completely.

    This is similar to changes in stockbroking that have already occurred. When I was younger, I bought shares by phoning a broker who placed the trade and charged me a lot for that service. Now I use an app and pay very little for a trade. The broker has disappeared. This process is called disintermediation.

    Other major changes are taking place in economics. Twenty years ago, we would have had no access to Nick or JKH and would have had to rely on textbooks. Now we can discuss complex issues with them and bypass the textbooks, and we can do it at no cost. Also, we can think through issues and challenge the experts. The experts are not used to this challenge. It’s not clear to me where this will end.

    In the UK at least, economics students are also challenging the experts.

    Within the economics profession, there are also changes. Nick’s post was about New Keynesian economists and how they have ignored money. Nick wants to challenge that. Also, New Keynesians are very dismissive about accounting. As I have said before, that seems like a chemist who is studying explosions being dismissive of atoms and their interactions. New Keynesians talk about “micro-foundations” but maybe accounting and money are the micro-foundations of economics. Putting accounting and money back at the heart of economics would be a further paradigm shift but the existing generation of New Keynesian economists will resist such a change.

    Another possibility of paradigm shift is that traditional economists might be displaced by theories from physicists or others as they get bored of the natural sciences and seek new challenges.

  9. hello!! Very interesting discussion glad that I came across such informative post. Keep up the good work friend. Glad to be part of your net community.
    contabilidade bh

  10. You’ve got some interesting points in this article. I would have never considered any of these if I didn’t come across this. Thanks!.
    contabilidade bh