Read on Twitter
Thread: 3 wrong theories of bank money creation: (1) An ordinary bank must wait for reserves (deposits) to come its way in order to make loans; if they seem to maintain an 5% reserve ratio, then they can only lend 95% or deposits received.
(2) An ordinary bank can create money equal to a multiple of reserves (deposits) received. So if it receives X dollars and maintain 5% reserves, it can lend 20 times X.
(3) An ordinary bank can create money "out of thin air," with no need to either possess or to acquire funds by which to finance loans it commits to make, and hence no need to worry about either its actual reserve or the cost of securing funds from elsewhere.
Alas, all three theories have achieved some degree of popularity. But all are equally mistaken. The true state of things rests on several points. First, when a bank makes a loan, it first credits the borrowers account, which costs it nothing. BUT
As the borrower draws on the account, funds must be transferred from it to persons who (w/ high probability) bank elsewhere. Other things equal (or, assuming we speak of a "marginal" loan), the lending bank must have a CB balance sufficient to finance the transfer.
It need not have had this balance on hand before it arrange the loan. It need not be a mere "passive intermediary (false theory #1). It might commit to the loan, and then seek the necessary funds. It only has to expect the loan to yield a profit net of the costs of the funds.
(If reserves bear interest, it won't lend even if it has plenty on hand unless the loans yield more than the reserves themselves do.)
But the bank has to fund its lending somehow, or else it will default at settlement time--again, other things equal. It may indeed have to rely on a last-resort central bank loan or overdraft; but then it has usually miscalculated, as this is usually a costly source of funding.
It follows that, while ordinary banks aren't _passive_ intermediaries, they are intermediaries nonetheless--go-betweens between funding suppliers and funding seekers. They do not, therefore, create money "out of thin air." Theory (3) is therefore wrong.
No can they lend a multiple of funds acquired, for that, too, would expose them to settlement default. So theory (2) is also wrong.
The situation for fiat-M issuing central banks is fundamentally different, because they alone have the power to create final settlement media, aka bank reserves. To them alone, theory (3) applies.
