[1]https://www.federalreserve.gov/monetarypolicy/reservereq.htm
While this is technically true, the statement does not do justice to the real economical case. You start pulling out the banks funding (i.e. deposits, interbank loans etc.) and the bank must start to liquidate its assets (i.e. loan book). In practice it is perfectly fair to say that bank uses deposits to fund its loan book, and thus saying that bank loans out deposits is not that wrong.
It's a bit like we having a lunch but you forgot your wallet. I pay your lunch and when we get back to office you pay me back. Then I start running around and telling that I paid your lunch. Yes, technically true, but not quite representative of practical economic reality.
I agree that in practice statements like the above are acceptable, but in an erudite paper like the Economist, when the article is discussing banking licenses, I expect a higher level of precision.
Part of the trickiness here is that even electronic reserves[1] and deposits are really different data types. There is something a lot like automatic boxing and unboxing between the two types that greatly confuses casual understanding of the operational realities of the banking system. When one withdraws paper notes from a bank, not one but two accounts are debited[2]: one is the deposit account, which is the one the customer sees, and the other is the institution's reserve account. Getting a grasp on these mirrored transactions was hard, and communicating that is even harder. I'm afraid I'm probably failing here too.
Edit: Similarly, creating deposits is a lot like a malloc(), and extinguishing those liabilities is a lot like a free(). I'd bet money the banking system actually is Turing complete.
Edit edit: It's also interesting that banks usually need reserves to clear inter-bank transactions rather than to meet reserve requirements. This is because inter-bank transfers (like depositing a check from another bank with yours) are also dual operations that involve both reserves and deposits. While a bank can create a million dollar loan without any reserves at all, they can't clear a $20 check drawn on their bank without at least $20 in their reserve account at the end of the business day.
[1] Reserve notes are the only form of reserves the vast majority of citizens ever hold. However Fed members can hold reserves in an online account.
[2] Informal usage
That is an important distinction to grasp. When people say that banks can create money from nothing, that means banks can create liabilities from nothing. But that can anyone do. I can write on a piece of paper that beefield pays 100 dollars to anyone holding this paper, and I need no reserves, no money, nothing but a pen and piece of paper to make that. And that is exactly what happens when bank "creates money". Bank says that some future day bank pays account holder money. And exactly as little as I can create the money that I would use to actually pay my liabilities, banks can't create money to pay their liabilities off.
Wait, what? I thought a bank cannot hand out any (book) money without backing [1]. I.e. if they want to loan me a million dollars by crediting it to my account, they have to balance that by "borrowing" the million from another bank or the central bank.
In other words, my bank account does not really sit at my bank. They cannot create money or a loan by just writing to a database [2]. And the other way around is also true: they cannot take (electronic) money. If I wire transfer money into my account, they don't just make an electronic note of it. They have to, in the same instant, deposit the money at another bank. It's always the case that another third party C makes the note "customer A deposited money into bank B's account at bank C". In a sense, the whole banking system is built on distrust.
This is the reason why, at least in Europe, banks "have to" "deposit" money overnight at the ECB, even if it costs due to negative interest rates. It's not a thing they conciously can or cannot do, but taking or lending (book) money always involves an upstream bank - unless it is cash of course.
[1] That might not be the right term. Financial English is not my strength :-)
[2] It would be an interesting model if individual banks could actually create money and anti-money, like the Dirac see model where an electron and a positron are spontaneously taken from the vacuum. Although you would have the same problem with the infinite vacuum energy, which would correspond to inflation I guess...
Loans are 'backed' by capital (paid up shares, retained earnings, etc). They have to have enough capital to cover at least 8% of the amount of the money they have created by lending with the recent Basel 3 rules.
What the bank is doing is managing IOUs. You do not store money in the bank, and the bank does not loan stored money. When you deposit money in the bank, they take it and give you an IOU (bank credit). When you loan from the bank, you make an IOU to them (loan), and they give you an IOU (a new deposit in bank credit). When you make a withdrawal, you're exchanging one of those IOUs (credit) to cash. When you make a transfer, they reduce their IOU to you, and then either increase an IOU to someone else, or settle that transaction between banks with some asset (such as bank reserves).
The bank does hold some of their assets at the central bank of their country (or ECB for Eurozone banks). But they actually try to keep as little as possible in there, and they don't need that much compared to the amount of bank credit on the bank's balance sheet.
I'm not familiar with the mechanics of the European banking system so I can't speak to that.
Loans and deposits do not have any other direct connection than handling them together being a convenient and sensible way to handle things. You can make a loan without making a deposit (by asking bank to pay the loan out in cash) or bank can make a deposit without there being a respective loan created (E.g. bank paying dividend to shareholder account)
Further, if the banker is really dumb, bank can put money on the account with no economic reason whatsoever by just stating that I owe now to this customer a million dollars. No loan needed. But usually, of course, bank wants the customer to pay the money back, i.e. make a loan agreement.
Also, if the customer is really dumb, he can obviously "take a loan" and promise to pay bank a million dollars in a years time without ever receiving any money on any deposit account. A loan created without any deposit activity. Of course, usually when a customer agrees to pay a million dollar in a years time, he wants to get the money first...
And if they know they won't be able to get the funding, they won't lend. This seems like an entirely pointless distinction to draw.
This describes every financial institution that isn't a national bank, e.g. broker-dealers, insurance companies, hedge funds, et cetera.
They're still loaning out deposits, though. They're just borrowing someone else's deposits to do it.
I'm not the GP here but his point is that actually the bank in general does not loan out deposits[0].
Instead, the bank creates money at the moment a counterparty agrees to borrow it (in countries with a certain type of financial system, aka most countries today, with some interesting exceptions).
The name for an institution that loans out deposits is "credit union," but most people think that's what a bank does.
Most people are very surprised by this claim that ordinary banks create money, and deny it when you tell them. But really there's no coverup of this fact, it's written e.g. literally on the websites of most central banks such as e.g. [1], wherein the critical passage (in boldface on the first page) is:
"Whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower’s bank account, thereby creating new money."
This seems unjust to most people, since in our daily experience, money is not a thing which is very commonly created. I take no position on that alleged injustice here, and instead leave it as an exercise for the reader to reason out the consequences of a private-bankers-printing-money themed financial system (of which consequences there are many) and to ponder the identities of the people doing most of the printing.
[0] Sometimes it actually does turn out that a bank loans out a deposit, as the GP mentions, but this is by construction a small fraction of bank business.
[1] https://www.bankofengland.co.uk/-/media/boe/files/quarterly-...
And is there a special license for being allowed to create money?
The difficulty is not in letting people make accounts and setting balances on those accounts. The difficulty is getting everyone to not immediately withdraw the entire balance as cash.
There are regulatory hurdles, but they're very much secondary.
If everyone immediately took loans in cash form and they never got deposited back to banks, then banks wouldn't be able to create money via loans either.
Banks count on (depend on!) there not being a run on the banks, and they count on borrowers to not default at a higher-than-expected rate. As long that is the case, the banks thrive, the borrowers thrive, and the savers thrive. If however, the economy tanks hard enough, then too many borrowers go broke, savers get scared and ask for their money back, and the banks go bust.
Banks lever capital (paid up shares and retained profits) to create money for loans, not deposits. Existing deposits are funding for interbank settlements (for transfers).
The amount of reserves the bank has doesn’t actually affect lending one bit - it does affect liquidity but most countries don’t even have reserve requirements because banks can just borrow reserves from other banks or the lender of last resort (central bank) if they need to. The main bank regulation now is capital adequacy ratios.
Their reserve, deposits, determine the size of loan they can create.
Without deposits banks couldn't loan out anything since they couldn't meet their reserve requirements and they'd basically be insolvent.
So, true, banks don't loan out 100% of deposits but they certainly do loan out deposits.
The 1000% figure comes from the fact that the money that the bank loans out is in turn held in another deposit account somewhere, and so it has a cascading effect... I deposit $100, my bank loans out $90 to person B, that money gets saved in a bank which can then loan out $81, then $72.9 gets created somewhere else. 1000% is just the limit of the sum 1 + 0.9 + 0.81 + ...
Sure. So we have a 10% reserve requirement.
Logically that means we have 90% loans, and 90% is nine times as much as 10%.
So loans are 900% as big as reserves.
Is that supposed to be surprising?
(I believe you, just want to learn more.)
NVM, I was confused over another statistic. ~97 percent of all new money is created in local banks, the government only creating ~3 percent.