Modern Monetary Theory (MMT): How Fiat Money Works

[imagep]

Think of a bookie - he's got a lot of people that owe him money, and once in a while he owes other people money. But the whole thing can be done in his book, with nothing changing hands, as long as the winners are content to keep their winnings "on the books." Which is what we do when we keep money in banks. If the bookie was so inclined, he could change his clients' balances to reflect, say, the sale of a car from one client to another.

Yup, thats EXACTLY the system that I am describing.

No, it really is out of thin air. A loan is just a matching set of brand new liabilities (borrower's deposit) and brand new assets (promissory note).

But nobody's savings account is debited to get "funds" for the loan. Your accounts are not touched when I borrow money. This was Richard Werner's point.

Sometimes I think it's just a language difference, because you say the right things about reserves, but what you said at the top of this post isn't correct. Anyway, we'll figure it out.

It's a metaphore. It's the same money circulating over and over again. If I have a shovel, and lend it to my neighbor, we both have a claim to a shovel, but there is still only one shovel. And if he lent it to someone else, and they lent it again, there could eventually be a million claims to that shovel, but still only be one shovel in existence, and it could only dig one hole at a time.

If I deposit some money, then I'm not using that money while it is in my bank account, so it's available for someone else to borrow. But if I demand my money back (cash or issue a check or use my bankcard, or whatever), then the bank has to return my money to me effectively by using someone elses deposit to give to me. Since banks pool all their electronic money into just one account at the fed, and since dollars are dollars and all spend just the same, money is fungible, and it doesn't particularly matter the source of giving me my money back, I don't care, I just want my money back. And the person getting the loan could also care less about the source of the money that they borrow, ever dollar is just the same to them. But at any one time, while a bunch of parties may have a claim to the same dollars (just like the shovel), they can't all use it at the same time, except for transactional time of course. So to the extent that not all bank transactions clear instantly, yes, banks create money, but that is temporary in nature, as is all bank created money.

When banks lend money, the M2 and higher metrics increase, but only because those metrics do not subtract out for the liabilities that are created in lending transactions. If we were to take the M2 and subtract out liabilities, we would end up with a figure very similar to the M1.

 

[James972]

But at any one time, while a bunch of parties may have a claim to the same dollars (just like the shovel),1.

actually we say that that when a bank takes in a dollar the marginal reserve requirement equals 10 percent of a bank's demand and checking deposits( in this case $1.00.) So if the bank has one shovel, to use your analogy, it can make 9 more for distribution. But the Fed controls how many dollars or shovels the bank can create over time. Do you understand.

 

[imagep]

I've seen a couple of studies posted that supposedly "prove" that banks just directly add digits to accounts, I've read them, and they are obviously faulty.

One of them listed about 5 points that they presented as "proving" that. None of them were valid.

I don't remember all of them offhand, but one of them was the fact that no employee ever had to check to see how much money the bank had in reserve before creating a loan. But why would they? Banks aren't constrained by reserves, worst case is that the bank didn't have enough excess reserves to cover the loan, and it borrowed that money overnight. The person in charge of approving loans knows this, he does this stuff every day. Also, since in both of these studies, the banks were aware of the study and the proceedure had already been agreed to. The bankers knew in advance that they were going to make this loan, and they knew that the check was going to be deposited directly back into an account at the same bank, thus with no money leaving the bank, there was no need to check to see if the bank had sufficient funds.

Another point made was that the banks reserve account with other banks (apparently the way reserve accounts are handled in other countries) didn't change. But my observation was that it didn't change a single penny. Surely, if the bank made any other transactions, it's reserve account(s) would have changed at least by a smidgen, so I can only conclude that the accounting of reserve accounts isn't in real time. But even if it was in real time, why would it have changed when all the bank did was to issue a check for a specific amount and then take that check back as a deposit - there was no net gain or loss to the banks holdings, thus there would have been no need for any reserve accounts to change, if the bank had $x.yz in reserve before this transaction, after the transaction it woul have $x.yz - transaction amount + transaction amount, would of course would still be $x.yz

 

[JohnfrmClevelan]

No, you can (almost) always get your money because not everyone demands all of their money out of the bank the same day. Besides, the bank has an option of issuing you a certified check, and the bank is allowed to clear checks overnight, so all the bank would need to do is to borrow the money overnight on the interbank lending system or direct from the fed.

During the great depression, bank runs were much more common than they are today, but the fact that we had bank runs (and still occasionally do), is proof that banks actually lend from deposits. You don't seriously think that the bank somehow magically stores your electronic or checking deposits in their physical vault do you? Most banking is done just by marking up or down accounts. If you withdraw your money, the bank then marks down your account, and issues you a check (creating the liability of that check). When you deposit that check at another bank, that bank then marks up your account and the fed marks down the original banks account.

In the event of a countrywide bank run, the most depositors could theoretically withdraw in cash, total, would be (MB). And that's after all reserves were converted to bills and drained from banks. The rest only exists as M1/M2, bank-created credits. After that, the govt. would have to buy up bonds and/or deficit spend in order to add to MB and rebuild the banks' reserve balances.

What is in bank vaults and drawers and ATMs is vault cash, which is a portion of MB and counts as reserves. Banks don't loan out of that, though.

 

[imagep]

actually we say that that when a bank takes in a dollar the marginal reserve requirement equals 10 percent of a bank's demand and checking deposits( in this case $1.00.) So if the bank has one shovel, to use your analogy, it can make 9 more for distribution. But the Fed controls how many dollars or shovels the bank can create over time. Do you understand.

I do understand, effectively you are correct, but I'm just pointing out that it doesn't work that direct. If I deposit a dollar in a checking account, the bank can lend 90% of that dollar out (keeping 10% as a required reserve), and the deposit from that loan can then be lent out again minus the 10% reserve amount, on and on until 100% of the dollar is being held as a required reserve, allowing the banking system to ultimately lend a total of $9 from just a $1 deposit.

But it's not like you deposit a dollar, and they then directly lend out $9. There is a process that is gone through to come up with that $9, and it's actually the same dollar being lent out and redeposited over and over again. When people say that a dollar deposit can result in $9 of lending, that's correct, but it's a simplification of a more complicated process than it sounds like.

Now lets say that I borrowed a dollar, demanded that I receive the dollar in physical money, and then used that buck to roll up a joint and I smoked that joint and discarded the roach. The bank has lost the opportunity to loan that dollar out over and over again. Fortunately for the banking system, most of the proceeds from loans are used to purchase a specific item, and thus the money isn't smoked away.

 

[James972]

In the event of a countrywide bank run, the most depositors could theoretically withdraw in cash, total, would be (MB).

yes because the Fed would be there to prevent it!!

 

[imagep]

In the event of a countrywide bank run, the most depositors could theoretically withdraw in cash, total, would be (MB).

Actually, they could, but that wouldn't happen in our modern banking system. We have rules in place that prevent that which didn't exist during the great depression. The first thing that would happen is that when banks run low on money, they are not required to pay out in cash, they are allowed to issue bank checks, and there is nothing you can do about it, either take the check or go away empty handed. The second thing that can happen is that the banks can declare a banking holiday, and shut down until the crazyness is over.

And that's after all reserves were converted to bills and drained from banks. The rest only exists as M1/M2, bank-created credits. After that, the govt. would have to buy up bonds and/or deficit spend in order to add to MB and rebuild the banks' reserve balances.

Most likely, most people wouldn't be able to instantly withdraw all of their money in cash, and they shouldn't expect to be able to do that. So they would get a cashiers check from one bank, and deposit it at another bank, which would then have a limit on how much cash they could withdraw.

What is in bank vaults and drawers and ATMs is vault cash, which is a portion of MB and counts as reserves. Banks don't loan out of that, though.

Sure, but they could lend out cash if they wanted to. Obviously, no bank keeps enough cash to handle a cash bank run, but since cash bank runs are very rare these days, and are stopped before the bank is out of cash, it's not really an issue.

 

[imagep]

actually we say that that when a bank takes in a dollar the marginal reserve requirement equals 10 percent of a bank's demand and checking deposits( in this case $1.00.) So if the bank has one shovel, to use your analogy, it can make 9 more for distribution. But the Fed controls how many dollars or shovels the bank can create over time. Do you understand.

But only one hole could be dug at a time. As soon as someone spent their deposit, that deposit no longer exists, and someone elses deposit has to replace it (or the bank has to borrow to make up the difference).

So yea, while in theory in aggregate we may all have a claim to gazillions of dollars, since there are not gazillions of dollars in existence (after claims have been settled), no new money is actually created - the only thing that is created is more claims to the same money. As long as we don't all demand our money out of banks at the same time, there is no issue. And even if we did all demand cash out of banks simultainiously, there is no law that says banks have to give us cash, they can give us checks (I suppose that could be considered as the bank creating new money).

 

[JohnfrmClevelan]

Money is fungible. There is no distinquishable difference between the original investor money (which is required to even open a bank) and money from deposits or any other source of money. So how can you prove that original investor money wasn't used in a particular loan?

Because investor money would show up as equity on the bank's balance sheet. If the bank had $1 million in equity and I took out a loan for $1000, the bank's equity would drop by $1000. And that is exactly what the Werner paper showed does not happen. Nobody's account goes down to make my loan happen.

What we mean by "money is fungible" is that it makes no difference if we hold a government-created paper dollar or a dollar in a bank account, they are interchangeable for spending purposes. But the paper dollar is MB, while the bank dollar is most likely M1.

Sure. So? And if banks literally horde all the deposits (what you are claiming), it's not even really a liability because those exact same deposits are in their vault or are offsetting in their accounting system - much like an individuals safe deposit box.

But M1 is greater than MB. A bank cannot cover everybody's deposit with a hard asset like reserves. Most of a bank's assets are promissory notes.

The reason banks consider deposits liabilities is because all of the money from depositors, and any other source, is co-mingled together, either in vault cash or in electronic moneys stored at the federal reserve. The word "reserve" is abiguous, there are clearly two types of reserves, one type is the required reserve, which is clearly defined by the fed as a percentage of non-time deposits (not a percentage of outstanding loans made by the bank, this is an undisputible fact). The other type of reserve is the excess reserve, which is any money that the bank has which is above the required reserves.

That doesn't mean you have two types of reserves, it just means that banks can (and do) hold excess reserves. Required reserves are, simplified, equal to 10% of the bank's total liabilities.

The reason banks consider your account balance a liability is because it represents dollars that the bank owes you. If you withdraw $100, you take $100 of the bank's reserves with you.

When you make a deposit, the bank gets both assets (reserves) and liabilities (your balance) from the transaction. But in reality, the assets and liabilities are separate - reserves are reserves, and your balance is your balance. If they were not separated, then banks couldn't create money by expanding their balance sheet and creating new dollars. That is when we would have 100% reserve banking.

If what you are claiming was true, then the required reserve rate on deposits would be 100% (that's what you are saying), yet we know that to be NOT true.

No, I never said anything like that. What did I say to make you think that?

Note that is says "SPECIFIED DEPOSIT LIABILITIES". You have claimed that the reserve ratio is based upon the amount of loans made,

No, the reserve ratio is based upon the bank's total liabilities - our bank balances. If I said anything else, then I misspoke.

...but obviously, a loan is an ASSET to the bank, not a liability. The "specified deposit liabilities" are obviously DEPOSITS, this should be obvious, because it literally says "deposits", plus it says "liabilities" and we all know that a loan that a bank makes to someone else is not a liabilty to the bank that lends the money, it is an ASSET, there is no ambiguity to this issue.

The promissory note is the bank's asset; the deposit (which probably left the bank soon after the loan) is the bank's liability.

 

[JohnfrmClevelan]

(cont.)

Of course, but that was the old west, today monetary policy is determined by the federal reserve. You are talking pre-gold standard banking, our banking is post gold standard. I find it ironic that MMT points out that economic text books are frequently incorrect because they were never updated to reflect the reality of our current system, and here you go referring to an eve more ancient system than the gold standard.

But banking is still done that way, that is the whole point. (As is was during the gold standard, too.) Reserve accounts are basically settlement accounts, and we have set laws in place to specify how much a bank needs in its settlement account. That's it. That's also the point when we point out that lots of other countries don't even have reserve requirements - in that case, banks are fine as long as they have enough reserves for settlement at the end of the day.

Most of the money we use is bank money, created the same way the Old West Bank created it - bank I.O.U.s, borne of loans to people and businesses. The only thing that is different with a central bank is that the central bank handles settlement. Instead of banks crediting each other with settlement accounts, our central bank creates MB, and that is what fills out settlement accounts now.

 

[James972]

But only one hole could be dug at a time. .

no, 9 holes can be dug at one time since the bank can lend 9 people shovels after one is deposited.

 

[James972]

no new money is actually created

actually tons and tons is created over time to keep up with an always growing GDP!!!

 

[JohnfrmClevelan]

It's a metaphore. It's the same money circulating over and over again. If I have a shovel, and lend it to my neighbor, we both have a claim to a shovel, but there is still only one shovel. And if he lent it to someone else, and they lent it again, there could eventually be a million claims to that shovel, but still only be one shovel in existence, and it could only dig one hole at a time.

If I deposit some money, then I'm not using that money while it is in my bank account, so it's available for someone else to borrow. But if I demand my money back (cash or issue a check or use my bankcard, or whatever), then the bank has to return my money to me effectively by using someone elses deposit to give to me. Since banks pool all their electronic money into just one account at the fed, and since dollars are dollars and all spend just the same, money is fungible, and it doesn't particularly matter the source of giving me my money back, I don't care, I just want my money back. And the person getting the loan could also care less about the source of the money that they borrow, ever dollar is just the same to them. But at any one time, while a bunch of parties may have a claim to the same dollars (just like the shovel), they can't all use it at the same time, except for transactional time of course. So to the extent that not all bank transactions clear instantly, yes, banks create money, but that is temporary in nature, as is all bank created money.

OK - so you are talking about reserves as "money."

 

[austrianecon]

Both of those articles support my position on this.

and mine..

 

[JohnfrmClevelan]

and mine..

So this has just been a language problem all along? We all agree with the papers in all respects?

 

[csbrown28]

But that's not exactly how it works.

Any bank can lend the full $20k regardless of their reserve position, because they are allowed to cover the loan overnight by obtaining the funds needed overnight.

Agreed, of course...

In the event that the loan money is deposited in the same bank, then they only need to borrow an addititional $2,000 to cover the loan.

Agreed, though in my hypothetical, it was deposited in another bank.

If the money is deposited in another bank, then they borrow the full $20k assuming that the bank doesn't already have enough excess reserves to cover the loan. If the bank already had ample excess reserves, then it wouldn't have a need to borrow any money at all.

In the hypothetical I create the bank had $10k in reserves, but setting that aside to avoid confusion, let's say the bank had no reserves, I think I'm going to disagree with you on this point. Why would the bank have to borrow $20k? to cover a $20k loan. You seem to be forgetting that the $20k loan would be the liability and the promissory note (the loan) would be an asset worth $20K (plus interest). If it borrowed $20k AND held the $20k value in the promissory note the banks leger wouldn't balance.

This is exactly why when you take a loan for a car, the bank never, ever owns the car. They own the contract you sign that you used to acquire the money buy the car and it's the contract that is the asset that balances the loan which is the liability. Within the contract is the right to take possession of the car should you fail to meet the terms. In the same way, the bank doesn't have to borrow $20k to cover a 20K loan, just the 10%. The Promissory Note is the asset that balances the loan. However the bank still needs to acquire the reserves, which it can do at the end of the day.

Thus, a bank would only need to borrow 10% of $20k to meet is reserve requirement of $2,000, even if the money was deposited in another bank. The promissory note would end up on the asset side of the ledger, but since the loan is balanced against the promise, it equals zero. The interest is unrealized profits.

However, in this way, customer funds are lent, only not to other customers but to other banks within the FRS. It's here that I suspect a key legal distinction might be made, but I'm not certain yet.....

The proceeds of loans create new deposits which in turn become new reserves, a portion of which can be lent again. That's how our banking system effectively creates new money. They don't directly and literally add digits to accounts, although it may "feel" like they do. Thats where a lot of MMTers get it wrong.

When you say "proceeds" I'm not sure what you mean, unless you just mean the money created by borrowing (the loan). I'd agree that a loan made to a customer becomes a deposit, a loan to a bank, which in turn is held as reserves reserves within the FRS, but here you claim that a bank loans customer deposits, it doesn't simply mark up accounts, but if a bank doesn't have the reserves until the end of the day when it acquires them from somewhere else, then they are adding digits to accounts.

Given my claim that the bank does not loan customer funds, but simply creates them (the liability) and the loan (the asset) balances out the transaction. Then all that is needed is the reserves to meet the reserve requirement.

That's my understanding anyway. I realize that much of this is fungible and I suspect there could be specific language used by banks in the Fed that makes important distinctions that are hard if not impossible to determine otherwise.

 

[austrianecon]

So this has just been a language problem all along? We all agree with the papers in all respects?

No, the papers don't actually support your position. You said banks don't loan deposits. That's factually not true. We agree when they don't lend out reserves. But deposit reserves are limited to just 10% of total deposits.

 

[austrianecon]

In the hypothetical I create the bank had $10k in reserves, but setting that aside to avoid confusion, let's say the bank had no reserves, I think I'm going to disagree with you on this point. Why would the bank have to borrow $20k? to cover a $20k loan. You seem to be forgetting that the $20k loan would be the liability and the promissory note (the loan) would be an asset worth $20K (plus interest). If it borrowed $20k AND held the $20k value in the promissory note the banks leger wouldn't balance.

The Bank will borrow $20k because they could loan out the other $18k at higher rate. Simply point, it's about leveraging the right way to make the maximum return.

 

[JohnfrmClevelan]

No, the papers don't actually support your position. You said banks don't loan deposits. That's factually not true. We agree when they don't lend out reserves. But deposit reserves are limited to just 10% of total deposits.

Are you calling the reserves that are transferred between reserve accounts "deposits"?

 

[austrianecon]

Are you calling the reserves that are transferred between reserve accounts "deposits"?

Wow.. no, what I am saying is that when a Bank takes your deposit.. 10% goes into reserve account. The rest is loanable.

 

[JohnfrmClevelan]

Wow.. no, what I am saying is that when a Bank takes your deposit.. 10% goes into reserve account. The rest is loanable.

OK, then. Glad we cleared that up. You're wrong.

When you deposit a $1000 check, they mark up your account, an added liability for the bank. At the Fed, $1000 in reserves are transferred from drawer's bank's reserve account to depositor's bank's reserve account, the matching asset to the check. There is nothing tangible to loan out to anybody. But since only $100 of those reserves are necessary to cover the $1000 deposit, the bank has $900 in excess reserves, which allows it to create another $9000 in loans. Hopefully, we agree up to this point.

The new loan is created by the bank expanding its balance sheet, because there is nothing tangible to loan out. The bank can create a loan for $9000 by marking borrower's account up by $9000 and executing a promissory note for $9000 (plus interest), and the necessary reserves are already in place. M1 increases by $9000, and MB stays the same.

Those papers don't back you up at all. They both confirm the credit creation version of banking.

 

[austrianecon]

OK, then. Glad we cleared that up. You're wrong.

When you deposit a $1000 check, they mark up your account, an added liability for the bank. At the Fed, $1000 in reserves are transferred from drawer's bank's reserve account to depositor's bank's reserve account, the matching asset to the check. There is nothing tangible to loan out to anybody. But since only $100 of those reserves are necessary to cover the $1000 deposit, the bank has $900 in excess reserves, which allows it to create another $9000 in loans. Hopefully, we agree up to this point.

The new loan is created by the bank expanding its balance sheet, because there is nothing tangible to loan out. The bank can create a loan for $9000 by marking borrower's account up by $9000 and executing a promissory note for $9000 (plus interest), and the necessary reserves are already in place. M1 increases by $9000, and MB stays the same.

Those papers don't back you up at all. They both confirm the credit creation version of banking.

Look, I am tired of arguing the obvious fact to MMTers. Imapeg covered it better then I could have. M1-MB is $800b. What you are saying is factually incorrect.

 

[imagep]

Because investor money would show up as equity on the bank's balance sheet. If the bank had $1 million in equity and I took out a loan for $1000, the bank's equity would drop by $1000.

No it wouldn't. The owners equity would be the same that it was before because making a loan isn't the same as spending money, the banks money on hand would drop by $1000, but it's assets would go up by $1000 (or more) because loans are assets to banks, the net change would be zero, or possibly even a few dollars profits if the loan was worth more than face value (which is common).

And that is exactly what the Werner paper showed does not happen. Nobody's account goes down to make my loan happen.

Exactly! If a bank lends out a thousand dollars, and that money is deposited back into the bank, no accounts would drop because it's a net neutral transaction. That proves MY point, not yours.

What we mean by "money is fungible" is that it makes no difference if we hold a government-created paper dollar or a dollar in a bank account, they are interchangeable for spending purposes. But the paper dollar is MB, while the bank dollar is most likely M1.

Yes, I agree. And what is your point?

But M1 is greater than MB. A bank cannot cover everybody's deposit with a hard asset like reserves. Most of a bank's assets are promissory notes.

Wait...you have been arguing that banks don't lend from deposits. No you are admitting that banks don't have enough reserves to cover the deposits. So are you waving the white flag?

That doesn't mean you have two types of reserves, it just means that banks can (and do) hold excess reserves. Required reserves are, simplified, equal to 10% of the bank's total liabilities.

You are playing semantic games again. Reserves can be categorized into to parts, required, and excess. Banks can lend from excess, but not required. I'm glad to see that you are admitting that required reserves are a percentage of LIABILITIES (in the past you have claimed that reserves were a percentage of assets/loans), but not all deposits even have a reserve requirement, only "on demand" deposits.

The reason banks consider your account balance a liability is because it represents dollars that the bank owes you. If you withdraw $100, you take $100 of the bank's reserves with you.

Yup, and $10 of that was a required reserve and the other $90 was excess reserves.

When you make a deposit, the bank gets both assets (reserves) and liabilities (your balance) from the transaction. But in reality, the assets and liabilities are separate - reserves are reserves, and your balance is your balance. If they were not separated, then banks couldn't create money by expanding their balance sheet and creating new dollars. That is when we would have 100% reserve banking.

We would have 100% reserve banking if banks held all time deposits like you WERE claiming they did. I assume that since you now admit that banks don't hold all time deposits and that they lend from them, you are just now realizing what "reserve banking" means.

No, I never said anything like that. What did I say to make you think that?

That's what it means when you claimed that banks don't lend from reserves.

No, the reserve ratio is based upon the bank's total liabilities - our bank balances. If I said anything else, then I misspoke.

That's still not exactly correct, banks don't have to keep reserves on all liabilities, and the percent is actually different based on the particular type of deposit account, but I'm glad that you are admitting that you may have been incorrect about this very important detail.

The promissory note is the bank's asset; the deposit (which probably left the bank soon after the loan) is the bank's liability.

Yup. And the bank only has a required reserve for the deposit (liability), not the promissory note (asset).

 

[imagep]

(cont.)



But banking is still done that way, that is the whole point. (As is was during the gold standard, too.) Reserve accounts are basically settlement accounts, and we have set laws in place to specify how much a bank needs in its settlement account. That's it. That's also the point when we point out that lots of other countries don't even have reserve requirements - in that case, banks are fine as long as they have enough reserves for settlement at the end of the day.

Most of the money we use is bank money, created the same way the Old West Bank created it - bank I.O.U.s, borne of loans to people and businesses. The only thing that is different with a central bank is that the central bank handles settlement. Instead of banks crediting each other with settlement accounts, our central bank creates MB, and that is what fills out settlement accounts now.

Back then, banks weren't regulated like they are today, and yes, there were times that banks could issue their own money. During a 130 year period of time we had at least 4 different monetary systems, and we have had three different monetary systems since that time. This caused huge problems, which is why some of the most successful bankers got together and came up with the concept of the federal reserve - a stable money supply and the management of our money supply was in the best interest of honest bankers and all Americans.

 

[imagep]

no, 9 holes can be dug at one time since the bank can lend 9 people shovels after one is deposited.

How? What would happen if the original depositor took his deposit out of the bank to bury in his back yard? Wouldn't the bank then have to re-aquire that same amount of money from another source?

For all practical purposes, the banking system depends on always being able to re-aquire any money that is withdrawn from individual banks instantly. We do have a system worked out where this can happen. But 8 of those nine claims to the same money can't use the money at the same time without the bank having to acquire more money from an outside source.