This article has always bugged me because I get hung up here:
> When a bank makes a loan, for example
to someone taking out a mortgage to buy a house, it does not
typically do so by giving them thousands of pounds worth of
banknotes. Instead, it credits their bank account with a bank
deposit of the size of the mortgage.
That's certainly not how it worked for me in the US. My bank account went down dramatically in the process of buying a house because I had to wire the down payment to an escrow company, and the bank gave either the escrow company or seller the rest of the funds (I assume, I had no visibility into the process. But at no point did me-as-borrower get an increase in my deposits!
Is that super meaningful? I wouldn't think so, except for that if the seller wants cash, or wants to deposit that money in a different bank (or puts it into the stock market, or whatever) then it requires my lending bank to have something other than just numbers in their own internal database - they have to convince that other institution that they're good for the money they just lent out. And that's the part where I'd assume consumer deposits would come back into play - unless the banks have another source of currency on hand.
Think of a car loan, then, in which case the bank can just increment the number in your deposit account (until you spend it on a car).
The bit of information your second paragraph alludes to is the fact that all banks have accounts at the Federal Reserve. The Fed has the single database that the banks use to clear with each other. And the Fed and other regularity agencies audit the banks to make sure their internal databases are consistent, their loans are backed by assets of sufficient quality, etc.
Something to note is that in the US (and most modern economies), the federal government creates a 1:1 exchange rate between private bank money (e.g. money created through loans) and central bank money (numbers in the Fed database and physical cash) via deposit insurance (e.g. FDIC in the US).
Sure, though it's rare that you'd borrow money without intending it to leave the bank (cases like paying off higher-interest stuff with lower-interest borrowing aside), but that aside, yeah, all the bits about how the banks have to have their accounts reconciled with the Fed and backing assets and all is really the core of my disagreement with the "banks can basically just print money infinitely" claims. Which I've sometimes seen people cite that BoE article as support of leaning on that "they just add a number in their computer when you take out a loan" bit. If it were that simple, I'd love to just make myself a bank and print myself some money, after all. ;)
My understanding is also that these discussions of "money" ignore things like investments or non-liquid assets, which I think is another big source of fuzziness. E.g. borrowing against other assets, including stock, that might have appreciated incredibly rapidly which gives you more purchasing power (the ability to "spend more money") without requiring anyone else to actually have given you money for anything specific.
See the part of my comment about the Fed and other entities regulating that the bank has claims on assets of sufficient quality. So the banks can’t just totally create loans willy nilly. They liabilities must be matched by assets of sufficient quality.
This does not preclude, however, asset bubbles as we saw in Japan in 1991 and globally in 2008. The banks create loans which drives up the price of assets. Those assets, now appearing to be worth more, enable the banks to create bigger loans, because hey, the asset is worth more! This is a positive feedback loop and a major failure mode of this system. Regulation tries to tamp it down but does not always succeed.
> My understanding is also that these discussions of "money" ignore things like investments or non-liquid assets, which I think is another big source of fuzziness.
It definitely is. When central banks think of the money supply, they take into account different aggregate which are sorted by liquidity.
There’s no reason (other than cost and your current wealth and bank risk management) why you couldn’t get a personal loan for the full purchase price of the house, and pay the seller the cash price. You’d then see your bank account balance go up by the house’s cost, withdraw the cash and then pay the seller.
I’m not sure what bank would give a normal person that kind of money unsecured but you could secure it with e.g. another house you own. Most people don’t have a spare house, so the banks optimise the process for the everyday scenario where the buyer doesn’t need to see their bank balance go up. But money is still being created in there somewhere
> When a bank makes a loan, for example to someone taking out a mortgage to buy a house, it does not typically do so by giving them thousands of pounds worth of banknotes. Instead, it credits their bank account with a bank deposit of the size of the mortgage.
That's certainly not how it worked for me in the US. My bank account went down dramatically in the process of buying a house because I had to wire the down payment to an escrow company, and the bank gave either the escrow company or seller the rest of the funds (I assume, I had no visibility into the process. But at no point did me-as-borrower get an increase in my deposits!
Is that super meaningful? I wouldn't think so, except for that if the seller wants cash, or wants to deposit that money in a different bank (or puts it into the stock market, or whatever) then it requires my lending bank to have something other than just numbers in their own internal database - they have to convince that other institution that they're good for the money they just lent out. And that's the part where I'd assume consumer deposits would come back into play - unless the banks have another source of currency on hand.