Right, but I’m not tracking your previous statement that banks don’t lend out existing deposits. They do! Sometimes up to 100% of existing deposits! I suspect we’re saying the same thing though. For example, a bank might take a $100 deposit, keep $10, and loan out the additional $90. That creates $90 because they original depositor still has the $100 in their account.
They don't lend out deposits, if a bank had zero deposits it could still loan you that $100 just by adjusting your balance and adding the loan to their books. It's a subtle difference, and honestly, the BoE explainer I linked to is by far better at explaining this than I am.
Yes, at the point where you try to withdraw actual cash the bank will have to use any cash it has as its assets, but nowadays the majority of transactions are digital and all that needs to happen is to adjust the balances of the payee and payer accordingly (potentially this can involve the interbank settlement network but the end result is the same).
Once you stop thinking of money in terms of tokens and bank deposits as those tokens getting stored in individual buckets - or there being any buckets at all - you can think of it in terms of assets and liabilities it makes sense. Also that commercial bank money (deposits) and central bank money (cash) are different because they are liabilities to different entities - commercial banks and the FED respectively.
Or more fundamentally, modern money is just I-O-Us being moved around.
Yes, I agree with all of this. But you’re still not going to have a bank with zero capital. We don’t have infinitely leveraged banks, for good reason :)
I’m still missing how this negates the money multiplier though. If banks are subject to reserve requirements, doesn’t the money multiplier still give a reasonable upper bound for the amount of money that can be created?
Also, this doesn’t negate the money multiplier. Assuming non-zero reserve requirements, the money multiplier is still an upper bound on the amount of money that can be created.
