Yes it does. A bank needs $1.1 billion to lend $1 billion, or they're out of compliance with Federal Reserve regulations.
The interesting part is that once they lend the $1 billion, and it's spent on aircraft (etc) and ends up in the bank accounts of Boeing and its contractors and its employees and the raw materials companies... then there's $1.1 billion in the original bank's accounts and $1 billion in all the Boeing accounts and there you go, they've turned $1.1 billion into $2.1 billion. And Boeing-etc's banks' can loan out up to about $909 million with it, and so on and so forth. Which may be what you're thinking of.
Now, the bank can borrow some or all of that $1.1 billion. And sometimes they can borrow that from the Fed. But the Fed isn't too big on that, and in non-2008esque-crisis situations tries to discourage it.
This is a huge commonly held misunderstanding of fractional reserve banking.
'The capital ratio is the percentage of a bank's capital to its risk-weighted assets. Weights are defined by risk-sensitivity ratios whose calculation is dictated under the relevant Accord. Basel II requires that the total capital ratio must be no lower than 8%.'
With $1 billion in deposits a bank may lend upto $12 billion under Basel II assuming a risk weight of 1.0
With $1 billion in assets and $12 billion in liabilities the capital ratio is 8%.
Where does the bank get this extra $11 billion? They borrow it from the Federal Reserve, unless they are lending it to another of their accounts in which case they just credit the account.
There is a disturbing amount of mis-information in this thread. As a sibling commenter writes, deposits have nothing to do with capital.
The Basel regulations are, as the term "risk-weighted assets" which you quoted implies, about the riskiness of assets of the bank.
NB: Loans made by a bank are assets of the bank, and they are risky, that is why Basel regulations are relevant. The corresponding liabilities of the bank are the money that is created in the debtor's accounts when the loan is made. But those liabilities are not part of the Basel computations, because Basel is about risky things. Risks do not come from liabilities, because liabilities are known, certain quantities. Risk only comes from assets.
When a risky asset has to be written off (e.g. loan goes bad), then the asset side of the bank's balance decreases. This is offset by an equal decrease on the liability side of the bank's balance. To be precise, the bank's capital is reduced (yes, capital is a liability).
This makes sense because capital represents the "liability" that the bank has towards its owners. When the bank makes bad decisions, the owners are supposed to pay for it in properly implemented capitalism.
When capital goes below zero, the bank goes bankrupt. Therefore, the ostensible goal of the Basel regulations is to ensure that capital never goes below zero (or, at least, that a lot has to go wrong before that happens).
This is why a risk-weighted sum of the bank's asset (the things that can go bad) is compared to capital (the only liability that can be legitimately decreased).
There's one of two possibilities, either I'm wrong and capital requirements don't work like that, or every bank in the world is insolvent. Given Quantitative easing 1,2,3,etc which do you think is true?
You take the blue pill – the story ends, you wake up in your bed and believe whatever you want to believe. You take the red pill – you stay in Wonderland, and I show you how deep the rabbit hole goes. Remember, all I'm offering is the truth – nothing more.
So from a bank's perspective, a loan is an asset and a deposit is a liability.
Say someone deposits $100 cash (federal reserve notes) into Bank A. Let's say the reserve ratio is 20%. It takes $80 and loans it to someone, who deposits in the same bank. It then takes $64 of that and loans it to someone who deposits in the same bank. It then takes $51 of that and loans it to someone who takes out cash and holds it.
The bank has the following assets: $20 + $16 + $13 in reserve, plus loans of $80 + $64 + $51 = $244.
It has the following liabilities: $100 + $80 + $64 = $244.
Now, if those loans don't get repaid, the bank might not remain solvent, but that has nothing to do with fractional reserve banking. Any entity that is solvent on the books can be rendered insolvent by loans going bad.
The accounting rules for banks are different. They book cashflow differently and I believe balance sheets are presented in a way that essentially neutralises the deposits and loans in order to focus on what the bank itself owns or owes.
Not true. There is only one Fed regulation about how much central bank money a bank must have, and that is the minimum reserve requirement, which is currently 10% for sufficiently large banks.
This is a required ratio between accounts of clients of the bank and money that the bank itself has in its account at the Fed (or as cash in its vaults). As such, it does not even have anything to do with loans in the first place.
The only way it has anything to do with loans is that as a loan of e.g. $1000 million is created, the bank creates a new account or marks up an existing account to the extent of $1000 million.
The minimum reserve requirement then increases by 10% of the newly created money, i.e. by $100 million. If the bank does not already have a sufficient amount of central bank money in its accounts, it must obtain this money within the next two weeks or so.
This is what happens in practice: Banks create loans based on creditworthiness of potential borrowers. An institutionally separate department of the bank then ensures sufficient central bank money to satisfy regulations.
tl;dr: Your number of $1.1 billion is completely wrong. The number $100 million would be somewhat less wrong, but is still not correct. In reality, banks do not need any money to make loans. They do need to satisfy minimum reserve requirements, but if necessary, they can obtain the required money after the loan is made.
> I think that's backwards. A bank is required to hold $100mm cash if it has $1b of demand deposits.
Yes. So it can loan out the other $900 million. Or, if it holds $1,111,111,111.11 it can loan out $1 billion.
> But it can hold all sorts of no-reserve-requirement deposits (like CDs) ... From that it can lend out whatever it wants
Yes, but in either case, it needs to have that $1 billion available to it in order to cut a $1 billion check. In neither case does a bank with only $100 million available get to loan out $1 billion, which is what it sounded like someone was suggesting.
Well... The bank could start with half a billion, and lend $450,000 of it. Then whichever bank that gets put into could lend 90% of that again, and so on. After a few times, there's a billion dollars lent from half a billion.
I fail to understand your reasoning: 0.5e9 * 0.9^n < 0.5e9 < 1e9. In other words: No one is ever receiving $1B even if banks' balance sheets sum to more than $1B lent collectively. On par, each bank is holding a small part of the 0.5e9, and then the final individual is holding the rest.
So? If you chain the loans, the bank might have the final balances: ($50M, -$450M) and ($45M, -$405M) and ($405M). The bank's net loan after settling internally is still $500M. Even if spread across banks, the sum of "real" money loaned out is still only $500M. If the loan at the end of the chain defaults, the loss is only $500M net.
The interesting part is that once they lend the $1 billion, and it's spent on aircraft (etc) and ends up in the bank accounts of Boeing and its contractors and its employees and the raw materials companies... then there's $1.1 billion in the original bank's accounts and $1 billion in all the Boeing accounts and there you go, they've turned $1.1 billion into $2.1 billion. And Boeing-etc's banks' can loan out up to about $909 million with it, and so on and so forth. Which may be what you're thinking of.
Now, the bank can borrow some or all of that $1.1 billion. And sometimes they can borrow that from the Fed. But the Fed isn't too big on that, and in non-2008esque-crisis situations tries to discourage it.