I tried to trace the eurocoins issued by Finland in the state financials. All I could find is an entry in the annexes indicating the estimated amounts needed in the next couple of years to cover returns, accompanied the total liability amount; this appears not to be recorded in the balance sheet. The Bank of Finland (part of the eurosystem) apparently reports (in its balance sheet) coins as a net of coins in circulation (a liability) and right of return to the state (an asset).
Interesting! It seems to me that there are few people, if any, who truly understand coins and we shouldn't discount the possibility of outright errors in their accounting treatment.
I have to correct myself when it comes to the Fed. Below I wrote how coins should be accounted for by the Fed, but that is not the case. All outstanding coins -- not just those held by the Fed -- are recorded as an asset of the Fed (and a liability of the Treasury - recorded only by the Fed and not the Treasury?!) and they ARE part of Currency in circulation (on the credit/RHS of the Fed balance sheet) once they are delivered to depository institutions. It's all nicely explained here: https://www.federalreserve.gov/monetarypolicy/bst_table1popup.htm
On the debit side:
" Treasury currency outstanding: Coin and paper currency (excluding Federal Reserve notes) held by the public, financial institutions, Reserve Banks, and the Treasury are liabilities of the U.S. Treasury. This item consists primarily of coin, but includes about a small amount of U.S. notes"
On the credit side:
"Currency in circulation: Currency in circulation includes paper currency and coin held both by the public and in the vaults of depository institutions. The total includes Treasury estimates of coins outstanding and Treasury paper currency outstanding."
So the Fed's accounting treatment of coins must look like this:
Purchase of $1000 worth of coins from the Mint: Debit 1000 to Treasury currency outstanding, credit 1000 to TGA.
Sale/delivery of coins to a deposit institution (DI): Debit 1000 to DI's Reserve account, credit 1000 to Currency in circulation.
Do all agree?
I find it weird that the Fed in this way keeps track of some Treasury liabilities outstanding, instead of the Treasury doing it itself, like in the case of Treasury notes and bonds. For one thing: in what way can coins held by the public be an asset to the Fed? (edit: See my reply to Brian below for my attempt to understand why the Fed does this.)
No. Coins are not a government "liability." They are pure assets, like gold or silver bullion, or your house, except originally, gold and silver had nothing to do with money, but coins were there first. And yes, in theory, they work as a tax credit, but let's not get this backwards, a credit for *whose* liability. Not a government liability. The term "liability" comes from the liability to redeem paper dollars, or other paper notes such as bonds, in coin, if so asked. So in theory, "tax credit works," as it does for Federal Reserve paper Notes, but in law, even that is not formally written, it's just people like us scrambling around to find out what this outdated word "liability" means that people forgot to drop from their discussions around currency. Everybody, get up to speed, now!
Um, I don't own any coins that have any "pure asset" value -- their only purpose, fundamentally, is to pay any taxes, fees or fines levied in Canada, though they are also accepted at local stores and banks.
Thanks for the follow-up, Brian! We can continue here our earlier discussion.
The Fed reports coins it (reserve banks) holds -- not coins in circulation -- as an asset. Some Chicago Fed researchers mention in footnote 4 to a 2018 paper (https://www.chicagofed.org/publications/chicago-fed-letter/2018/395) that coins are liabilities of the Treasury, not the Fed. But I haven't found coins on any Treasury/government balance sheet yet. Logically coins should be part of the public debt -- otherwise depositing the "trillion dollar coin" at the Fed to get the Fed to credit the Treasury's account (TGA) would make no sense at all. It must be a token of public debt, even if it is interest-free and without any predefined due date.
Any examples from other countries of coins as tokens of public debt? Historical examples? I'm too tired now, after Christmas shopping, to look for them. I'm sure they exist, though.
This from the GAO kind of hints at coins being a way of financing the deficit, just like issuing Treasury debt is, but stops short:
"Because the Mint, a part of Treasury, is included in the budget, the recognition of the earnings from coins—the difference between production costs and face value, called seigniorage—is shown in the federal budget as a reduction in needed borrowing for the government, after the deficit or surplus for the year is calculated." https://www.gao.gov/assets/gao-04-283.pdf
If it was just a nice income, seigniorage, then why would it not be affecting the deficit or surplus? Now they calculate the deficit, and instead of borrowing the whole amount through Treasury debt issuance, they first deduct coins issued that year (because those are public debt tokens, I argue -- or at least they should be).
- One issue with coins is that the issuance is a fairly small dollar amount in the context of government finance, and they stay out in circulation for a long time. They might get buried in other items.
- If there is no expectation that they return, they could legitimately be dropped from the balance sheet of the issuer.
- The GAO quote implies that they don’t include seigniorage in the entries going into the deficit as they define that term, they are a fudge factor that explains part of the wedge between the deficit and net borrowing. Seignorage may not be the only fudge factor involved. In my primers, I don’t sweat those tiny details, but the government accounting rules are more complicated than what shows up in economic model accounting.
I agree that the comparatively small amount might have an effect here. But $53 billion is not that small, is it? I think this is more about confusion, perhaps due to some historical reasons. Perhaps the Fed is doing what it does (see my comment above to Antti-Juhani) because it is mandated/supposed to track also coins in its Currency in circulation account? That would make sense to me. If it includes a Treasury liability on the credit side of its balance sheet, then it must include it also on the debit side of its B/S, in order to net it out (as it is not a Fed liability, and everything on the credit side is considered a Fed liability).
In the case of public debt, I don't agree with you on dropping / not recording it even if there is no "expectation of return". There is always going to be a "float" (in insurance terminology) of public debt, also of zero-interest debt (which all Treasury debt held by the CB in practice is?). Some of it could be in the form of a "billion dollar coin" on the CB balance sheet and it wouldn't make things different. It is debt and it can be repaid by the public/Treasury if it so chooses (just like the UK repaid the Consols).
Coins are not debt, not liabilities. Coins are the way you think about silver or gold bullion, except they are nothing to do with silver or gold, the metals--look historically (the Innes book), there is no connection between the quantity of metals contained in any coin and their value. But they are not debt, they have value in the same way your house has value.
If the metal value is a lot lower than the face value, then they are tokens. And tokens might not be directly debt (as in IOU), but they are part of recordkeeping of debts and credits. There is a reason why the Fed has all the coins in circulation registered as credits, and why it says that all coins outstanding are Treasury liabilities. Oliver Davey and I are trying to figure out how all this recordkeeping works (check "The Money View vs. The Accounting View"). The debts are always debts of real resources, but that is not easy to see if one is used to consider money a 'thing', as you obviously are.
The ratio of metal to its face value doesn't matter, as ninety-percent silver coins were treated the same as today, for their face value, not for their silver content. So just figure out how the ninety percent silver coins were treated and treat the modern ones the same. All that record keeping doesn't matter, either.
The Fed officially calls coins a Treasury liability (see my reply to Antti-Juhani above). They are recorded on the debit side of its balance sheet as such, just like Treasury bonds the Fed holds are. And just like Fed notes, they are credits on the account Currency in circulation. Tax credit is not a wholly mistaken way of understanding coins, and Innes, too, got a lot right.
We all have formed our own way of making sense of money. It's a mighty, madness-inducing task. That's why it's hard to convince the other party of the superiority of your way of understanding it?
edit: Tom edited his comment after my reply, partly to address my reply. An interesting way to have a discussion.
I think banknote issuance covers coin issuance as coins are just banknotes broken down into smaller denominations. The banknote has to exist first. I'm sure the Canadian Treasury issues coins and replacements through their subsidiary agent - the Canadian mint. Also Mint would have some accounting process for seigniorage.
The issue is whether the Bank of Canada is involved in coin distribution. The private banks could buy them directly from the Mint, and so there is no connection to the BoC’s balance sheet.
As noted, I did a quick scan of the Canadian Mint financials, and they had a face value liability, but it may be just silver coins with a C$ face value. If there is no right to return non-precious coins to the Mint, they could just strike them off the balance sheet, and the seigniorage profits just flow into the income statement.
Even with $1 and $2 coins, I don’t think that’s a whole lot of profits.
Mr. Romanchuck, I submit there is no difference between a non-precious coin and a precious coin. As for a "precious" coin, that would be final payment--one wouldn't return it to the Treasury, for redemption, though I get your discussion is about torn bills, and worn or damaged coins, whether the mint has promised to replace them or not, but casting "expected expenses" as "liabilities" is really stretching it a bit.
From what I saw of the financials of the people who issue Canadian silver coins - The (Royal) Canadian Mint - there is a difference between them and other Canadian coins. There is a face value liability on their balance sheet, which the footnotes indicate are associated with silver coins.
And in practice, there is certainly a difference. You can walk into a bank and ask for quarters, loonies, nickels, or whatever, but they won’t give you the silver coins for their face value. You need to go to wherever those coins are distributed (Post Offices, maybe other dealers). And you certainly never see them used in commerce.
Not to bomb you with comments, but I had meant the non-precious coins, as to right of return, comparing them to the historic pre-1965 coins as they were once used *in 1965*, so I missed the trail on that one. I think Miss Ryan also meant seigniorage on non-precious, real, coins, or should have...
My local office of the national tax agency of Japan will take real non-precious coins, well, any physical cash (it's prohibitively expensive not to pay in cash, actually), back as revenue, except it turns out I was told I owed, "zero."
As for never seeing precious coins in commerce, I have a 1965 40% silver half dollar, and certainly it had come at some point into our hands via commerce, so...
1) Comparing a fiat currency to a non-fiat currency (since pretty much everybody except Canada in 1965 were non-fiat) is questionable. If you have a gold peg, all your monetary liabilities can be converted into gold (directly or indirectly, as in the Bretton Woods, where only foreign central banks had convertibility rights). There is no reason why accounting conventions must be the same.
2) Modern effectively worthless coins tend to stay in circulation for a long time. How governments account for them legitimately has a great deal of leeway. (You can write down the liability on the basis that there is a legitimate belief they will not be redeemed.) The amounts involved are trivial, so there is no particular need to care. (As for the Trillion Dollar Coin, that is just an accounting gimmick that counters a stupid accounting gimmick in the United States.) Returning coins to the government to pay taxes (not actually allowed in Canada) is just the taxpayer redeeming a government liability to cancel out the government's tax receivable asset.
3) Sure, you can buy precious coins as a collectible, and buying and selling collectibles is a form of commerce, My intended meaning is that nobody expects to see such coins in use when paying for groceries (or any other non-collectible store) unless somebody has done something stupid.
I checked the Australian distribution procedures. Maybe similar to BoC.
"The Royal Australian Mint (RAM) issues coins to banks on behalf of the Treasury, and the number of coins released into circulation is based on bank demand:
How coins are distributed
The RAM issues coins directly to banks, who then place them in circulation.
How coins are removed from circulation
Banks remove damaged coins from circulation and send them back to the RAM to be recycled. "
I tried to trace the eurocoins issued by Finland in the state financials. All I could find is an entry in the annexes indicating the estimated amounts needed in the next couple of years to cover returns, accompanied the total liability amount; this appears not to be recorded in the balance sheet. The Bank of Finland (part of the eurosystem) apparently reports (in its balance sheet) coins as a net of coins in circulation (a liability) and right of return to the state (an asset).
Interesting! It seems to me that there are few people, if any, who truly understand coins and we shouldn't discount the possibility of outright errors in their accounting treatment.
I have to correct myself when it comes to the Fed. Below I wrote how coins should be accounted for by the Fed, but that is not the case. All outstanding coins -- not just those held by the Fed -- are recorded as an asset of the Fed (and a liability of the Treasury - recorded only by the Fed and not the Treasury?!) and they ARE part of Currency in circulation (on the credit/RHS of the Fed balance sheet) once they are delivered to depository institutions. It's all nicely explained here: https://www.federalreserve.gov/monetarypolicy/bst_table1popup.htm
On the debit side:
" Treasury currency outstanding: Coin and paper currency (excluding Federal Reserve notes) held by the public, financial institutions, Reserve Banks, and the Treasury are liabilities of the U.S. Treasury. This item consists primarily of coin, but includes about a small amount of U.S. notes"
On the credit side:
"Currency in circulation: Currency in circulation includes paper currency and coin held both by the public and in the vaults of depository institutions. The total includes Treasury estimates of coins outstanding and Treasury paper currency outstanding."
So the Fed's accounting treatment of coins must look like this:
Purchase of $1000 worth of coins from the Mint: Debit 1000 to Treasury currency outstanding, credit 1000 to TGA.
Sale/delivery of coins to a deposit institution (DI): Debit 1000 to DI's Reserve account, credit 1000 to Currency in circulation.
Do all agree?
I find it weird that the Fed in this way keeps track of some Treasury liabilities outstanding, instead of the Treasury doing it itself, like in the case of Treasury notes and bonds. For one thing: in what way can coins held by the public be an asset to the Fed? (edit: See my reply to Brian below for my attempt to understand why the Fed does this.)
But coins are still a government "liability" (tax credit, if you will) are they not? Maybe we can think of them as zero coupon perpetuals?
No. Coins are not a government "liability." They are pure assets, like gold or silver bullion, or your house, except originally, gold and silver had nothing to do with money, but coins were there first. And yes, in theory, they work as a tax credit, but let's not get this backwards, a credit for *whose* liability. Not a government liability. The term "liability" comes from the liability to redeem paper dollars, or other paper notes such as bonds, in coin, if so asked. So in theory, "tax credit works," as it does for Federal Reserve paper Notes, but in law, even that is not formally written, it's just people like us scrambling around to find out what this outdated word "liability" means that people forgot to drop from their discussions around currency. Everybody, get up to speed, now!
Um, I don't own any coins that have any "pure asset" value -- their only purpose, fundamentally, is to pay any taxes, fees or fines levied in Canada, though they are also accepted at local stores and banks.
I'm not talking about gold dubloons here, LOL ...
Thanks for the follow-up, Brian! We can continue here our earlier discussion.
The Fed reports coins it (reserve banks) holds -- not coins in circulation -- as an asset. Some Chicago Fed researchers mention in footnote 4 to a 2018 paper (https://www.chicagofed.org/publications/chicago-fed-letter/2018/395) that coins are liabilities of the Treasury, not the Fed. But I haven't found coins on any Treasury/government balance sheet yet. Logically coins should be part of the public debt -- otherwise depositing the "trillion dollar coin" at the Fed to get the Fed to credit the Treasury's account (TGA) would make no sense at all. It must be a token of public debt, even if it is interest-free and without any predefined due date.
Any examples from other countries of coins as tokens of public debt? Historical examples? I'm too tired now, after Christmas shopping, to look for them. I'm sure they exist, though.
This from the GAO kind of hints at coins being a way of financing the deficit, just like issuing Treasury debt is, but stops short:
"Because the Mint, a part of Treasury, is included in the budget, the recognition of the earnings from coins—the difference between production costs and face value, called seigniorage—is shown in the federal budget as a reduction in needed borrowing for the government, after the deficit or surplus for the year is calculated." https://www.gao.gov/assets/gao-04-283.pdf
If it was just a nice income, seigniorage, then why would it not be affecting the deficit or surplus? Now they calculate the deficit, and instead of borrowing the whole amount through Treasury debt issuance, they first deduct coins issued that year (because those are public debt tokens, I argue -- or at least they should be).
- One issue with coins is that the issuance is a fairly small dollar amount in the context of government finance, and they stay out in circulation for a long time. They might get buried in other items.
- If there is no expectation that they return, they could legitimately be dropped from the balance sheet of the issuer.
- The GAO quote implies that they don’t include seigniorage in the entries going into the deficit as they define that term, they are a fudge factor that explains part of the wedge between the deficit and net borrowing. Seignorage may not be the only fudge factor involved. In my primers, I don’t sweat those tiny details, but the government accounting rules are more complicated than what shows up in economic model accounting.
I agree that the comparatively small amount might have an effect here. But $53 billion is not that small, is it? I think this is more about confusion, perhaps due to some historical reasons. Perhaps the Fed is doing what it does (see my comment above to Antti-Juhani) because it is mandated/supposed to track also coins in its Currency in circulation account? That would make sense to me. If it includes a Treasury liability on the credit side of its balance sheet, then it must include it also on the debit side of its B/S, in order to net it out (as it is not a Fed liability, and everything on the credit side is considered a Fed liability).
In the case of public debt, I don't agree with you on dropping / not recording it even if there is no "expectation of return". There is always going to be a "float" (in insurance terminology) of public debt, also of zero-interest debt (which all Treasury debt held by the CB in practice is?). Some of it could be in the form of a "billion dollar coin" on the CB balance sheet and it wouldn't make things different. It is debt and it can be repaid by the public/Treasury if it so chooses (just like the UK repaid the Consols).
Coins are not debt, not liabilities. Coins are the way you think about silver or gold bullion, except they are nothing to do with silver or gold, the metals--look historically (the Innes book), there is no connection between the quantity of metals contained in any coin and their value. But they are not debt, they have value in the same way your house has value.
If the metal value is a lot lower than the face value, then they are tokens. And tokens might not be directly debt (as in IOU), but they are part of recordkeeping of debts and credits. There is a reason why the Fed has all the coins in circulation registered as credits, and why it says that all coins outstanding are Treasury liabilities. Oliver Davey and I are trying to figure out how all this recordkeeping works (check "The Money View vs. The Accounting View"). The debts are always debts of real resources, but that is not easy to see if one is used to consider money a 'thing', as you obviously are.
The ratio of metal to its face value doesn't matter, as ninety-percent silver coins were treated the same as today, for their face value, not for their silver content. So just figure out how the ninety percent silver coins were treated and treat the modern ones the same. All that record keeping doesn't matter, either.
The Fed officially calls coins a Treasury liability (see my reply to Antti-Juhani above). They are recorded on the debit side of its balance sheet as such, just like Treasury bonds the Fed holds are. And just like Fed notes, they are credits on the account Currency in circulation. Tax credit is not a wholly mistaken way of understanding coins, and Innes, too, got a lot right.
We all have formed our own way of making sense of money. It's a mighty, madness-inducing task. That's why it's hard to convince the other party of the superiority of your way of understanding it?
edit: Tom edited his comment after my reply, partly to address my reply. An interesting way to have a discussion.
I think banknote issuance covers coin issuance as coins are just banknotes broken down into smaller denominations. The banknote has to exist first. I'm sure the Canadian Treasury issues coins and replacements through their subsidiary agent - the Canadian mint. Also Mint would have some accounting process for seigniorage.
The issue is whether the Bank of Canada is involved in coin distribution. The private banks could buy them directly from the Mint, and so there is no connection to the BoC’s balance sheet.
As noted, I did a quick scan of the Canadian Mint financials, and they had a face value liability, but it may be just silver coins with a C$ face value. If there is no right to return non-precious coins to the Mint, they could just strike them off the balance sheet, and the seigniorage profits just flow into the income statement.
Even with $1 and $2 coins, I don’t think that’s a whole lot of profits.
Mr. Romanchuck, I submit there is no difference between a non-precious coin and a precious coin. As for a "precious" coin, that would be final payment--one wouldn't return it to the Treasury, for redemption, though I get your discussion is about torn bills, and worn or damaged coins, whether the mint has promised to replace them or not, but casting "expected expenses" as "liabilities" is really stretching it a bit.
From what I saw of the financials of the people who issue Canadian silver coins - The (Royal) Canadian Mint - there is a difference between them and other Canadian coins. There is a face value liability on their balance sheet, which the footnotes indicate are associated with silver coins.
And in practice, there is certainly a difference. You can walk into a bank and ask for quarters, loonies, nickels, or whatever, but they won’t give you the silver coins for their face value. You need to go to wherever those coins are distributed (Post Offices, maybe other dealers). And you certainly never see them used in commerce.
Not to bomb you with comments, but I had meant the non-precious coins, as to right of return, comparing them to the historic pre-1965 coins as they were once used *in 1965*, so I missed the trail on that one. I think Miss Ryan also meant seigniorage on non-precious, real, coins, or should have...
My local office of the national tax agency of Japan will take real non-precious coins, well, any physical cash (it's prohibitively expensive not to pay in cash, actually), back as revenue, except it turns out I was told I owed, "zero."
As for never seeing precious coins in commerce, I have a 1965 40% silver half dollar, and certainly it had come at some point into our hands via commerce, so...
1) Comparing a fiat currency to a non-fiat currency (since pretty much everybody except Canada in 1965 were non-fiat) is questionable. If you have a gold peg, all your monetary liabilities can be converted into gold (directly or indirectly, as in the Bretton Woods, where only foreign central banks had convertibility rights). There is no reason why accounting conventions must be the same.
2) Modern effectively worthless coins tend to stay in circulation for a long time. How governments account for them legitimately has a great deal of leeway. (You can write down the liability on the basis that there is a legitimate belief they will not be redeemed.) The amounts involved are trivial, so there is no particular need to care. (As for the Trillion Dollar Coin, that is just an accounting gimmick that counters a stupid accounting gimmick in the United States.) Returning coins to the government to pay taxes (not actually allowed in Canada) is just the taxpayer redeeming a government liability to cancel out the government's tax receivable asset.
3) Sure, you can buy precious coins as a collectible, and buying and selling collectibles is a form of commerce, My intended meaning is that nobody expects to see such coins in use when paying for groceries (or any other non-collectible store) unless somebody has done something stupid.
I checked the Australian distribution procedures. Maybe similar to BoC.
"The Royal Australian Mint (RAM) issues coins to banks on behalf of the Treasury, and the number of coins released into circulation is based on bank demand:
How coins are distributed
The RAM issues coins directly to banks, who then place them in circulation.
How coins are removed from circulation
Banks remove damaged coins from circulation and send them back to the RAM to be recycled. "