Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Not really. The process is as follows.

You go to a bank with an asset. In this case the new house. The bank then creates a new financial asset called a 'mortgage' that goes on the asset side of their balance sheet. Against that they create a liability called an 'advance' that they give to you.

You then give the advance to the builder in return for the house, where it becomes a deposit of the builder than they can then transfer to anybody else in payment for goods and services.

If any of them happen to be at another bank then all that happens is the deposit is transferred to the bank and it becomes an 'inter bank loan'. The target bank then simply creates another deposit in its books to balance that loan and that is allocated to the customer being paid.

It's all just debits and credits. What limits bank money creation is the assets they are prepared to discount and the creditworthiness of the individuals they choose to advance to and whether the bank believes they can pay the current price of money.



I think you’re glossing over some important details regarding the inter-bank loans. As far as I understand, they are not automatic and must be negotiated in the same way as any other bank loan.

Modulo some amount of transaction batching, when one bank’s customer initiates a payment to an account at another bank, reserves are actually transferred between the two banks (these days, in the form of central bank account transfers).

Each bank’s main concern is managing its daily cashflow— They need to have sufficient reserves each day to cover that day’s net outflow of payments. But holding reserves is expensive; they would much rather use that currency as an advance payment for a new loan if they can, so that somebody else is stuck with the currency.

And here’s where the inter-bank loans come into play: Because most of a bank’s outflows happen on an irregular schedule as directed by their depositors, and they have an incentive to hold as few reserves as possible, they may find themselves in a position where they would need to suspend payments. When this happens, they borrow reserves from another bank in order to continue operating. If the borrowing bank is considered trustworthy by the banking community, they generally have no trouble securing such a loan on standard terms.

Of course, if a bank systematically misjudges the default risk of their borrowers, they may have a bigger problem that their assets (future loan repayments) won’t cover their liabilities (future withdrawals by depositors). If that imbalance isn’t corrected quickly, they will eventually find themselves unable to secure future inter-bank loans because they will have become an unacceptable default risk to the other banks.


"As far as I understand, they are not automatic and must be negotiated in the same way as any other bank loan."

Not in the slightest.

If the bank doesn't take over the deposit in the source bank, then their customer doesn't get paid. At which point they will close their account and move to the source bank. That's a deposit outflow they won't want to have.

Both the source and the destination bank have an interest in ensuring the payment clear. Therefore it does.

Inter bank loan negotiation is about shuffling the risk around for a price afterwards.

"reserves are actually transferred between the two bank"

Reserves don't exist in aggregate. Central banking is a collateral optimisation of correspondent banking.

There was no such thing as reserves in the UK banking system for three hundred years, for example.

Correspondent banking is the base for payments, and is still used for most international currency payments - even with the advent of the CLS central clearing mechanism.


> If the bank doesn't take over the deposit in the source bank, then their customer doesn't get paid. At which point they will close their account and move to the source bank. That's a deposit outflow they won't want to have.

The destination bank can demand to withdraw that balance. And they typically do that with the net flows at the end of the settlement period.

(And they don't withdraw physical currency these days. The 'withdrawal' comes in the form of a transfer of electronic central bank reserves.)

> Reserves don't exist in aggregate. Central banking is a collateral optimisation of correspondent banking.

No? This is easiest to see in a gold standard system: your physical gold are your reserves.

In typical modern fiat system, the reserves are account entries at the central bank (and vault cash, but that's a small-ish amount).

Private commercial banks can create 'money' via their normal commercial operations. But they can't create central bank reserves that way, nor are they allowed to print central bank notes.

> There was no such thing as reserves in the UK banking system for three hundred years, for example.

Citation needed. And also an explanation of exactly what you mean by reserves.

(Btw, talking about the 'UK banking system' over the last three hundred years already makes me very suspicious of your argument. During most of that time, Scottish banking differs remarkably from English banking. So there are not many blanket statement about the 'UK banking system' that are true.)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: