1) Fractional reserve banking has caused numerous financial crisies.
2) In the US, banks are required by law to carry deposit insurance (FDIC). While it is theoretically possible for the FDIC to become insolvent, that is far less likely than a bank becoming insolvent. Further, in practice, the FDIC can't fail because the government will just fund it directly if it's actual funds ran out.
3) While banks have less cash on hand than the sum of their debts; they are still solvent. That is to say, they are capable of paying back all of their creditors, they just need to be able to collect from their debtors in order to afford it.
When a bank takes a deposit for $100, they have a $100 liability, and a $100 asset in cash. Net worth $0, technically solvent.
They then loan out $90. Considering the likelihood of repayment, interest rate, and loan term, this debt is an asset that is worth (hopefully) >$90, so the bank remains solvent. The bank can take some of that surplass worth they have to do things like pay salaries, build offices, fund FDIC, give executive's their bonues, etc.
In contrast, when a stablecoin takes a deposit of $100, they have a $100 liability and $100 asset in cash. If they take $90 to pay salaries, they now have only $10 dollars in assets and are insolvent. In theory, a stablecoin could loan out that $90, or use it to invest is something else. If they do this, they are in roughly the same situation as the bank where they are hoping that their asset ends up being worth as much as they expect. They are also less regulated then banks, so the odds of something going wrong (either mallisiously or not) is greater.
In practice, even the normal banking system runs into some issues:
* Deposits into banks are much more liquid then loans out of banks. In theory, everyone can decide to withdraw all their money tomorrow. But if you took at a mortgage, the bank may have to wait 30 years to be fully payed back. This can be mitigate by selling debt to other institutions, or taking out loans backed by the debt, but you can still run into issue when these markets experience problems.
* Debtors do not always pay back what they owe to the banks. In theory this is accounted for, but if banks overestimate the actual value of the debts, they can end up in a situation where they are actually insolvent.
Banks don't lend out deposits though - a $100 loan creates both a $100 liability - the value in the customer's account - and a $100 (plus interest) asset - the loan itself - which also balances out to $0. The process literally creates money, meaning a bank could theoretically lend money even if it had no deposits at all. The constraints on making loans come from central bank requirements and regulations, not deposits made.
The fractional reserve model makes sense for commodity-backed money or things such as stablecoins that are backed 1:1 in some way, just not for banking in the modern economy where physical money is only a small part of the money supply.
>The process literally creates money, meaning a bank could theoretically lend money even if it had no deposits at all
If the assets and liabilities balance, why describe it as "literally creating money"?
It makes it sound to me like someone gained something, like banks have a special privilege, and like money is a physical resource.
And being able to lend money without deposits is possible whenever a bank has money from another source. Are you suggesting that there is something non-obvious going on that permits bootstrapping from nothing?
> If the assets and liabilities balance, why describe it as "literally creating money"?
1) Say you have an account at FooBank with a balance of $1000, and I have an account there with a balance of $0. The total supply of money at this point is $1000.
2) I apply for a loan of $1000 and get it. My account now has a balance of $1000, and your account still has a balance of $1000. The total supply of money is now $2000!
Adding another layer:
1) I have an account at FooBank with a balance of $0. You have $1000 in cash in your pocket. FooBank has $0 of assets and $0 of liabilities. The money supply is $1000 of cash.
2) You open an account at FooBank and deposit your $1000 of cash. FooBank now has $1000 of assets (your cash) and $1000 of liabilities (your deposit), balancing out still. The money supply is also still $1000, now made up of $1000 of bank deposits and $0 cash.
3) I apply for a loan of $1000 and get it. My account now has a balance of $1000, and your account still has a balance of $1000. The bank now has $2000 of assets ($1000 cash plus my $1000 loan) and $2000 of liabilities ($1000 in each of our accounts), which still balances out. However the money supply is now $2000, made up of $2000 of bank deposits and $0 cash.
The money supply here is M1 or "broad money" which consists of the total of coins and bank notes plus deposits in checking/current accounts, which you can roughly think about as the total amount of money held by households and companies.
4) I withdraw $1000 in cash from my account at FooBank, leaving me with $1000 cash and $0 in my account. FooBank has assets of $1000 (my loan) and liabilities of $1000 (your account). The money supply is still $2000 but now comprises $1000 of cash and $1000 of bank deposits.
The bank's accounting of assets and liabilities always match up, as do ours - you started with $1000 of assets and no liabilities and ended up with exactly the same, while I started with $0 assets and $0 liabilities and ended up with $1000 of assets and a $1000 liability. However the total of our cash and bank deposit assets has gone from $1000 to $2000 i.e. the money supply has increased.
So yes, banks really do have a special role in the economy in that most "money" is in the form of bank deposits and most transactions involve moving money between bank accounts. It's not super-intuitive and doesn't quite fit the "common sense" view money as a set of tokens that just get moved about that's ingrained in our thinking - that works for cash, commodity money or cryptocurrencies, but not when viewing modern economies and fiat currencies where money is an IOU between two parties.
"1) Say you have an account at FooBank with a balance of $1000, and I have an account there with a balance of $0. The total supply of money at this point is $1000.
2) I apply for a loan of $1000 and get it. My account now has a balance of $1000, and your account still has a balance of $1000. The total supply of money is now $2000!"
The money supply didn't change as far as I can see.
At first:
Me: $1000, FooBank: $-1000, You: $0
After your loan:
Me: $1000, FooBank: $0, You: -$1000
The sum was $2000 before and $2000 after. Do you have an issue with any of the six numbers above?
You could, I suppose, say "I prefer to count the proceeds of my loan, rather than the debt, and not use a negative sign". Ok, but since we're adding the absolute values to get the total, it's still the same number.
After working it out painfully and slowly, it seems to me where you're going wrong is you think there is a fundamental difference between deposits and loans, other than the direction they're going in. I believe that's an error.
When I deposit money in the bank, that's just a type of loan. To the bank, rather than from the bank.
I'm pretty sure millions of people believe in and spread your meme about creating money, so feel free to explain why the majority is right.
Counting money is complicated. When I deposit $1000 in the bank, then even if the bank loans out a portion of it, my account still has $1000, and I still act like I have $1000.
Put another way, there are essentially 2 types of money: central bank money and commercial bank money.
When you deposit $1000 of real cash into a bank you exchange $1000 of central bank money, for $1000 of commercial bank money.
When the bank loans out $1000 to someone else, they credit that person's account with $1000, creating $1000 of commercial bank money out of nothing. The recipient of the loan is entitled to redeem that for $1000 of central bank money, but they probably wont, because cash is inconvenient and you can do most things with commercial bank money anyway.
I suppose the difficulty is more in defing money than counting. Apart from the M0 component (included in all counts) counting should just be a matter of aggragating data from all banks and the fed.
> "SomeoneElse" who got the loan has $1000, right.
Define $. "SomeoneElse" has $1000 of bank money. But the bank still has the $1000 of cash you deposited.
It might be easier to look at this through the definitions:
M0: Total amount of physical money. This is $1000 throughout the story. Only the Treasury is allowed to create M0.
M1: M0 + checking and savings accounts.
M0 is unchanged at $1000. However, there are 2 checking accounts with $1000, so M1 is $3000.
> The money supply didn't change as far as I can see.
Your numbers (1) look like a attempt to assess net assets and liabilities rather than money supply and (2) are wrong even for that.
Money supply (I’ll use M1 here because it's simple, but M2 or M3 wouldn't differ in any way material to the example, since the example consists entirely of cash and demand deposits) is the sum of all cash, demand deposits, travelers checks, and certain other highly liquid deposits in the economy.
At first: You have $1,000 in demand deposits. M1 is $1,000.
After the loan, you still have $1,000 in demand deposits and the loan recipient has $1,000 in cash (or, more likely, a demand deposit, because they probably get and deposit a check.) M1 is $2,000.
I think I agree, if you divide assets into "money" and "other stuff" buckets then when you move something between them, "money" is created.
I don't think people intend to say this and only this, when they talk about creating money. Because it's trivial.
But I can't really engage with you about stuff other people wrote and what they might mean. I belatedly realized there were a bunch of people tag-teaming me.
> I think I agree, if you divide assets into "money" and "other stuff" buckets then when you move something between them, "money" is created.
A new loan doesn't move assets between buckets it creates a new asset and a new liability. The asset is money, the liability is not money (nor is it negative money).
> I don't think people intend to say this and only this, when they talk about creating money. Because it's trivial.
Money creation is fairly trivial if somewhat counterintuitive, but people talk about it because it is important in its economic effects.
> The sum was $2000 before and $2000 after. Do you have an issue with any of the six numbers above?
Well clearly because that's not what I wrote. You're mixing liabilities with assets and also redefining the money supply to include the bank. Not sure how borrowing $1000 leaves me with "-$1000" in assets or what that's supposed to mean.
The money supply is the total cash and bank deposit assets of all individuals and corporations. Central bank money is different and isn't part of it.
>Well clearly because that's not what I wrote. You're mixing liabilities with assets and also redefining the money supply to include the bank
Wait, I thought the entire point is that FooBank is creating money? Do you want to pretend they don't exist?
>Not sure how borrowing $1000 leaves me with "-$1000" in assets or what that's supposed to mean
Ok, fine, you can have the $1000, it's not a loan after all.
Lemme update it:
At first:
Me: $1000 in deposits
FooBank: $1000 in assets - $1000 in deposits, total of $0
You: $0
After your non-loan:
Me: $1000 in deposits
FooBank: $1000 in liabilities, no assets
You: $1000
But! FooBank can never return my deposit at this point.
So the cold hard truth is...
Me: $0 in deposits
FooBank: (bust)
You: $1000
And so in the end, there is the same $1000 that there was in the beginning, no money created.
In sort-of real life, the FDIC would bail me out, I guess, and then I would have $1000, but that certainly would be the opposite of the bank creating it.
That's not a great analogy, you can't redeem currency at a bank.
My understanding is that stable coins are stable because they claim to be backed 1:1 by some other currency. If they are not actually doing that, then thats a fraudulent statement, isnt it?
Not all stablecoins claim redeemability for USD. Some claim each unit's redeemability for an amount of some digital asset that is worth 1 USD, with the amount redeemed fluctuating in proportion to the value of the digital asset, to maintain the stable value.
Maybe I am missing something, how is this different than banks employing fractional reserve banking?