Peter, Paul and M0.
Absolutely nothing. This fact is part of the logic behind the creation of central banks, which can act as a lender of last resort. Also why there are government insurance schemes to protect depositors. This is the only valid criticism of fractional lending- it can exacerbate periods of economic distress and cause bank runs in extreme cases.
I will leave that one alone for the moment. Bank runs of the past (before 1913) are different in character from bank runs now. So, let me just move to the next parts.
Well that might be my vernacular chosen in attempts to explain the same subject a trillion ways, hehe. And it's more or less true- like you asked above, if depositors/borrowers try and withdraw the $19,000 ($10,000 deposits and $9,000 loan) from the banking system they will find there's only $10,000 actually available.
I see no proof in the diagram or in anything stated so far for the last sentence above. In fact, there is something of a proof against this idea in stage 6. When a borrower deposits the $9,000 in that stage in another bank the assets and liabilities both go down by equal amounts in the bank holding the original $9,000.
In the model I advocate there is an exacting correspondence between new deposit money and new loans, as is shown even in the page 11 table. The difference is that new loans is not new money, but new debt, which is covered by the new money deposited in the deposit liabilities. With respect to the system as a whole, there is a one to one correspondence (again, as is shown I would argue by the table on page 11 of MMM).
No different to if I took your money, recorded that fact in an Excel spreadsheet, and then lent it to someone else. Sure, the spreadsheet still records the money I owe you, but when you come asking for it back I tell you I don't have it. But, in reality this is rarely a problem because that 10% covers daily usage easily.
This sounds like one of those stories people say to little kids. I am going to address this further below.
Good, we can use this as a fiducial point of understanding.
Because money circulates repeatedly. The sum total of M0 in the system at any given stage isn't that relevant because loans are not paid off all at the same time. You pay your loan back to your bank, the bank pays me to clean its toilets, I spend my wage at your grocery store. You pay more off your loan. Repeat.
I agree with psionl0 on this one. M0, at least according to wikipedia, which can be wrong, is not checkbook money, but coins and notes. If said money is stored at a bank, it does add to reserves of that bank though (and deposits!).
Aside from the M0 thing, the sentence above is very handwavy, as have been the arguments thus far for the ideas of the Begging Peter to Pay Paul model of how banking is supposed to operate. I will call, Sceptic-PK, your model, the Peter-Paul model (unless you want to call it something else, but please do not be conceited by calling it the Fractional Reserve Banking model, because that is after all what is being debated about). My model I am going to call the Shell model.
Quick synopsis of each model to the best of my abilities without supporting arguments. I am also not going to address how Central Banks work.
Shell Model:
There are three main types of actions. One action, called a transfer, decreases the assets and liabilities of a given bank (to the exact same amount) and increases in exactly the same previous amount the assets and liabilities of another bank.
The other type of action is a loan. In this action the assets and liabilities of a given bank both rise by the same amount. A bank can do the loan action as long as it does not go over 90% (or other relevant fraction as the case may be) of the difference of deposits and already existing loans.
The third type of action is repayment. In this action assets and liabilities go down by the same amount, but in specific, deposits and loans go down by the same amount. This can happen so long as what you would probably expect to pay a bill, the loan amount is not already zeroed and the account in the deposit has the loan repayment amount in the first place.
Peter-Paul Model:
Money circulates repeatedly. The repayment and transfer actions in this model are roughly analogous to the previous model in at least some senses. The loan process is very different. When the bank has X and it loans Y the bank now has X - Y money left. The bank acts though like it still has X, however that may be construed. A series of loans would (in differing banks) look like then
Actual Bank Money Has After Loan
vs.
Loan Amount
vs.
Amount Bank Acts Like It Has:
X1 - X2 vs. X2 vs. X1
X2 - X3 vs. X3 vs. X2
etc.
Comments.
Shell Model: So far as I can tell it comports with what is written in MMM. There is quite a bit missing to give even a rough picture of actual banking (such as interest, operating costs, assets that do not count as liabilities, etc.) but it follows well to what is written in MMM.
Peter-Paul Model: Undoubtedly there will be complaints as to my characterization of this model. I did try and be as faithfull to the model as I could, but the problem is I do not see the logical consistency of this model so that made it hard very hard to write out (you try stating someone else's case you do not agree with and you will understand why! It is harder then it looks!).
There is no problem per se in a mathematical sense with the idea of zero-sum loans as to how banks supposedly operate. The problem is what does "Amount Bank Acts Like It Has" mean? It either has it or it doesn't, why the charade? I hope Sceptic-PK you do not think I am setting up a Strawman here because words to the previous effect were written by you. I will find them if you disagree. The model so far as I can tell does not match up with what is written about in MMM either, my own canonical source on these matters.
This reminds me of when I was debating a fellow on the MySpace physics forum. There are these things called invariants, just a fancy term for formulas that stay the same when the underlying variables are transformed in some way. The invariant of Special Relativity is s^2 = t^2 - x^2, or a difference. Forgeting about the squares for the moment, I hope everyone will realize that if you want s^2 to stay the same, but want to use an x that is greater than before, say x1, you have to use a t1 that is greater than the t from before. So in this case s^2 = t1^2 - x1^2 = t^2 - x^2, t1 > t, x1 > x.
This is roughly analogous to how banking works. You have reserves (s^2) and you want to make a loan (a greater value in x), then you will need a greater value in deposits (t^2) to offset the increase in loans (x^2). The analogy of course breaks down in a way because while in SR you can increase t without limit (this does not break the c-limit in case you are wondering, and remember t^2 is deposits in this analogy), you can not increase deposits/loans without limit due to the geometric sum nature of FRB.
Sceptic-PK, you are like that fellow who misunderstood SR. He could not get his head around the idea of an invariant having a difference because he always thought the invariant had to be a sum! In a sum, such as z = x + y, if you increase x, you have to decrease y to keep z the same. Look at page 11 table of MMM. The invariant is the total reserves (note, it never changes!) which is the difference of Deposits and Loans (something you have already agreed to). That is the true invariant.
Hope that helps.
All the best to you all!
