When I first read about bookkeeping as a teenager, one of the things which confused me was the insistence that debits must equal credits for a given transaction. E.g. if a firm buys a delivery van for $20,000, it credits the "cash" account by $20,000 and debits the "vans" account by $20,000. But what if it had been offered a 10% discount? Its "vans" account would have an $18,000 debit. Are the vans somehow worth less because the firm didn't have to pay as much for them?
Now I understand that some of accounting, particularly valuation, is accounting convention, which is corrected over time (e.g. by depreciation, or sale of the van). For economic modelling, I've started using the idea of "raw net worth": not assigning monetary values to everything. So when a firm buys a van for $20,000, it loses a $20,000 asset and gains a van asset. It's not appropriate for accountancy, which must assign a fair value to a firm, but it's a much more direct model of what is actually happening, making it work very well for economic modelling.
I also split transactions into individual actions (each represented by a coloured arrow). So buying a van is 2 actions: (1) transfer of $20,000 from the firm to the seller, which decreases the firm's raw net worth by $20,000, and (2) transfer of the van from the seller to the firm, which increases the firm's raw net worth by the van.
(If you're in the mood for a slight complication, there are actually 2 more actions: (3) decrease in firm's equity by $20,000, and (4) increase in firm's equity by the van).
In other words, you’re not doing “simple” double-entry [DE] accounting. That’s OK. As long as your rules are coherent and achieve your objective, I won’t object.
Splitting a transaction in two (with each arrow representing one ‘half’ of a ‘trade’) makes it quadruple-entry [QE] accounting, I think. It does seem redundant to me, but your purpose seems to require it.
The “van” in your comment may have many different “values” - at different times and for different people - but it only has one “Exchange Value” at the point of sale, which is called it’s “sale price”, which is what the £20,000 (or discounted £18,000) of money does measure. Distinguish “price” and “value”; what you “think” a van is worth is irrelevant until you try to buy or sell it. It’s value TO YOU is entirely personal TO YOU. THAT ‘value’ doesn’t exist anywhere else but in your head.
The rule about equal CR and DR entries in different accounts is the mechanism that ensures the necessary Balance condition is maintained in the Sheet of accounting records. My Figure 5 (et seq.) tries to summarize how and why that works.
I'm beginning to see what you are trying to do. I once had a similar idea (~1980), viz., that there are 5 basic ways anyone's assets (your RNW) could change: 4 were ONE-way - finding (or creating), losing (or destroying), gift, theft, - the other was TWO-way - exchange.
As you say, new "values" (RNW) arise from finding/creating processes while destruction/deterioration is the one-way process which reduces global RNW. DE accounting certainly applies to trade/exchange processes.
Thanks for the link to your substack from the comment on mine. It's a very important topic, and I'm glad to see people discussing it.
I'm not an accountant, but I've spent a few years studying it now, particularly in relation to economics, and I have a couple of thoughts about your introduction to accounting. I hope it helps.
1. "Balance Sheet, Assets & Liabilities". Assets are not just what an entity owns, but also what it is *owed*. Most of a bank's assets are actually debts owed by another entity (e.g. borrowers, issuers of financial instruments, other banks).
2. If we subtract liabilities (including equity) from assets, we get net worth. The simplest way I've found to reason about debits and credits is: debits increase net worth; credits decrease net worth.
3. "Creditor and Debtor on the Same Account". I think it's best to think of accounts as not being shared between parties, but each party having their own account for the other. So, for example, when Bob has savings at Bank X, Bank X has a "Bob" account, and Bob has a "Bank X" account (whether he records it electronically, on paper, or even not at all). Bob's $100 of savings are a liability on the bank's "Bob" account, and an asset on Bob's "Bank X" account. This should help to avoid any confusion.
Your description of bank lending as the creation of two separate debts is exactly as I would expect to see: consistent with Perry Mehrling, Steve Keen, Richard Werner, etc.
I'll continue with a second comment, on a slightly different topic.
Thanks for your comment. I’m not an accountant either, so we may be able to help each other here. I can see your suggestions reflect your objective of getting a global appreciation of “economic issues” and can see the possible benefits of what you are saying, in that context.
Your point 1: I agree. No need to change anything there. You’re saying that I effectively own what is owed to me. I actually make that point in my articles; a bank customer already owns ANY credit-balance in his bank account, as creditor (even – or especially - the “magical” ones that banks call “loans”).
Your point 2: Neat! Debits reduce liabilities and also increase assets, so increase RNW; Credits reduce assets and also increase liabilities, so reduce RNW. Everyone needs some way of remembering how these two words work in two opposite directions in different accounts. I like you way, but will also keep my way.
Your point 3: I have to disagree; every account inevitably and automatically links two “people” – the Debtor and the Creditor. Your point seems to be that, because every account is being viewed by two different people, each one needs to have a separate account. But why complicate it? Why have two accounts – effectively saying the same thing, twice - when one serves both functions? By all means, do it, if it helps your global analysis, but my local objective of exposing bank fraud doesn’t need that complication.
Now, I have a double-barreled question for you: “Using your ‘arrow’ notation, can you, and how would you, represent *fraudulent accounting* by a bank?” The details of the fraudulent accounting are laid out in my “Deconstructed” articles (#2, #3 and #5).
Thanks for the reply, and I'm looking forward to discussing the subsequent articles!
I do need to disagree with what you wrote on point 1 though. There's a very important difference between being owed something and owning it. For every debt asset you have (a thing which is owed to you) someone else must have a matching liability; an owned asset does not have a matching liability elsewhere. Also, a debt is only a promise, and it's possible to make a promise which can't be kept. If I promised to give you Sydney Harbour Bridge next year, that wouldn't give you any ownership rights to it. It just means that I'd be almost certain to break my promise next year (and you might sue me).
On point 3, I'm not sure how much it matters, but while I agree that in principle there's no need for a model to duplicate information about debts (and in fact the arrow notation reflects this), an account which shows debts is inherently a representation from just one party's perspective. For example, if an account uses a debit to represent an event where the bank customer's balance decreases, that account is from the perspective of the bank. The customer's account with the bank (if they're keeping records) would show exactly the same event as a credit. I volunteered at a local credit union, and saw its account with the bank where it kept its reserves; the bank would have its own account for the credit union, which would be its mirror image.
I think it's also important to say that not every account describes a relationship between two parties. For example, a car manufacturer would have an account for the stock of unsold inventory, for which no other party is involved.
As to representing fraudulent accounting, the arrow notation doesn't do this, because it represents the underlying debts, rather than each party's view of the situation. To me, fraudulent accounting would be something like X debiting its "Y" account, thereby claiming that it is owed something by Y, when Y has not agreed to owe it to X, and shows nothing in its "X" account. When the claimed debt is due, X will simply have to correct its "Y" account.
When I first read about bookkeeping as a teenager, one of the things which confused me was the insistence that debits must equal credits for a given transaction. E.g. if a firm buys a delivery van for $20,000, it credits the "cash" account by $20,000 and debits the "vans" account by $20,000. But what if it had been offered a 10% discount? Its "vans" account would have an $18,000 debit. Are the vans somehow worth less because the firm didn't have to pay as much for them?
Now I understand that some of accounting, particularly valuation, is accounting convention, which is corrected over time (e.g. by depreciation, or sale of the van). For economic modelling, I've started using the idea of "raw net worth": not assigning monetary values to everything. So when a firm buys a van for $20,000, it loses a $20,000 asset and gains a van asset. It's not appropriate for accountancy, which must assign a fair value to a firm, but it's a much more direct model of what is actually happening, making it work very well for economic modelling.
I also split transactions into individual actions (each represented by a coloured arrow). So buying a van is 2 actions: (1) transfer of $20,000 from the firm to the seller, which decreases the firm's raw net worth by $20,000, and (2) transfer of the van from the seller to the firm, which increases the firm's raw net worth by the van.
(If you're in the mood for a slight complication, there are actually 2 more actions: (3) decrease in firm's equity by $20,000, and (4) increase in firm's equity by the van).
In other words, you’re not doing “simple” double-entry [DE] accounting. That’s OK. As long as your rules are coherent and achieve your objective, I won’t object.
Splitting a transaction in two (with each arrow representing one ‘half’ of a ‘trade’) makes it quadruple-entry [QE] accounting, I think. It does seem redundant to me, but your purpose seems to require it.
The “van” in your comment may have many different “values” - at different times and for different people - but it only has one “Exchange Value” at the point of sale, which is called it’s “sale price”, which is what the £20,000 (or discounted £18,000) of money does measure. Distinguish “price” and “value”; what you “think” a van is worth is irrelevant until you try to buy or sell it. It’s value TO YOU is entirely personal TO YOU. THAT ‘value’ doesn’t exist anywhere else but in your head.
The rule about equal CR and DR entries in different accounts is the mechanism that ensures the necessary Balance condition is maintained in the Sheet of accounting records. My Figure 5 (et seq.) tries to summarize how and why that works.
I'm beginning to see what you are trying to do. I once had a similar idea (~1980), viz., that there are 5 basic ways anyone's assets (your RNW) could change: 4 were ONE-way - finding (or creating), losing (or destroying), gift, theft, - the other was TWO-way - exchange.
As you say, new "values" (RNW) arise from finding/creating processes while destruction/deterioration is the one-way process which reduces global RNW. DE accounting certainly applies to trade/exchange processes.
Hi,
Thanks for the link to your substack from the comment on mine. It's a very important topic, and I'm glad to see people discussing it.
I'm not an accountant, but I've spent a few years studying it now, particularly in relation to economics, and I have a couple of thoughts about your introduction to accounting. I hope it helps.
1. "Balance Sheet, Assets & Liabilities". Assets are not just what an entity owns, but also what it is *owed*. Most of a bank's assets are actually debts owed by another entity (e.g. borrowers, issuers of financial instruments, other banks).
2. If we subtract liabilities (including equity) from assets, we get net worth. The simplest way I've found to reason about debits and credits is: debits increase net worth; credits decrease net worth.
3. "Creditor and Debtor on the Same Account". I think it's best to think of accounts as not being shared between parties, but each party having their own account for the other. So, for example, when Bob has savings at Bank X, Bank X has a "Bob" account, and Bob has a "Bank X" account (whether he records it electronically, on paper, or even not at all). Bob's $100 of savings are a liability on the bank's "Bob" account, and an asset on Bob's "Bank X" account. This should help to avoid any confusion.
Your description of bank lending as the creation of two separate debts is exactly as I would expect to see: consistent with Perry Mehrling, Steve Keen, Richard Werner, etc.
I'll continue with a second comment, on a slightly different topic.
I've just updated my article to include "& is owed", following your point 1. Thanks for that correction.
I'm glad you found that suggestion helpful!
Thanks for your comment. I’m not an accountant either, so we may be able to help each other here. I can see your suggestions reflect your objective of getting a global appreciation of “economic issues” and can see the possible benefits of what you are saying, in that context.
Your point 1: I agree. No need to change anything there. You’re saying that I effectively own what is owed to me. I actually make that point in my articles; a bank customer already owns ANY credit-balance in his bank account, as creditor (even – or especially - the “magical” ones that banks call “loans”).
Your point 2: Neat! Debits reduce liabilities and also increase assets, so increase RNW; Credits reduce assets and also increase liabilities, so reduce RNW. Everyone needs some way of remembering how these two words work in two opposite directions in different accounts. I like you way, but will also keep my way.
Your point 3: I have to disagree; every account inevitably and automatically links two “people” – the Debtor and the Creditor. Your point seems to be that, because every account is being viewed by two different people, each one needs to have a separate account. But why complicate it? Why have two accounts – effectively saying the same thing, twice - when one serves both functions? By all means, do it, if it helps your global analysis, but my local objective of exposing bank fraud doesn’t need that complication.
Now, I have a double-barreled question for you: “Using your ‘arrow’ notation, can you, and how would you, represent *fraudulent accounting* by a bank?” The details of the fraudulent accounting are laid out in my “Deconstructed” articles (#2, #3 and #5).
Thanks for the reply, and I'm looking forward to discussing the subsequent articles!
I do need to disagree with what you wrote on point 1 though. There's a very important difference between being owed something and owning it. For every debt asset you have (a thing which is owed to you) someone else must have a matching liability; an owned asset does not have a matching liability elsewhere. Also, a debt is only a promise, and it's possible to make a promise which can't be kept. If I promised to give you Sydney Harbour Bridge next year, that wouldn't give you any ownership rights to it. It just means that I'd be almost certain to break my promise next year (and you might sue me).
On point 3, I'm not sure how much it matters, but while I agree that in principle there's no need for a model to duplicate information about debts (and in fact the arrow notation reflects this), an account which shows debts is inherently a representation from just one party's perspective. For example, if an account uses a debit to represent an event where the bank customer's balance decreases, that account is from the perspective of the bank. The customer's account with the bank (if they're keeping records) would show exactly the same event as a credit. I volunteered at a local credit union, and saw its account with the bank where it kept its reserves; the bank would have its own account for the credit union, which would be its mirror image.
I think it's also important to say that not every account describes a relationship between two parties. For example, a car manufacturer would have an account for the stock of unsold inventory, for which no other party is involved.
As to representing fraudulent accounting, the arrow notation doesn't do this, because it represents the underlying debts, rather than each party's view of the situation. To me, fraudulent accounting would be something like X debiting its "Y" account, thereby claiming that it is owed something by Y, when Y has not agreed to owe it to X, and shows nothing in its "X" account. When the claimed debt is due, X will simply have to correct its "Y" account.