Fractional Reserve Banking – Two Approaches to Regulation
The meaning of a “fractional reserve” Banking system seems simply enough – the first way to understand it would be simply place it in opposition to a full reserve banking system. A full reserve bank is a bank where assets are placed in a vault, and held there (presumably against them). Owners of assets are given promisory notes which all them to return and withdraw their deposits when they please. For this protective service, the owners of they money held at the bank pay a fee. On the other hand, in a fractional reserve bank, we might assume that deposits are made by the owners of assets, the Bank holds a fraction of those assets are a reserve and loans the rest out at a rate of interest. They then pay some of this interest back in proportion to the deposits – and in a free system a tension between security (your assets will be there when you return for them.
On the surface, the fractional reserve system seems just as safe as the full reserve system because the possibility of a loan not being repaid (loosing a depositors money) is evened out by the interest paid on the loan (i.e. I pay interest to conver someone else’s actual default, which is in fact the possibility that I might default). However, this seems to hold true even when banks are making deposits in other banks. Imagine BankA makes a deposit of 100$ into BankB – the 100$ must have been taken out of moneys deposited into BankA, from the portion other than the reserve fraction, so in loaning the money from a bank to another bank, Bank A looses 100$ of money it could loan to someone else (it’s oppertunity cost), and Bank B gains 100$ more it can loan out, but must pay for it by an interest rate. This system has a high degree of protection from systemic risk – BankB might fail, but unless Bank A had made many loans to Bank B, Bank A will take a reasonable hit. The key here is that Bank A has not issued loans on the basis of money it has already loaned.
However, the fractional reserve system that we have is a bit different from this when it comes to deposits made by a central bank. Whereas when a depositer places money in the bank, the money, say 100$, is simply divided into the 20$ fractional share, and the 80$ loan share, when a central bank deposits 100$ into a bank there is a third thing created. In addition to the 80$ that can be loaned out and the 20$ that is kept as a reserve, there is another 80$ in the form of a promisary note, which is effectively the treatment of the loan as an asset. Since the bank has an extra 80$, it can loan this to another bank, who will take 16$ of that and place it in the reserves, and loan the 64$, and then loan the 64$ (again, this is not a 64$ bill from the reserve, but a promisory note for such a bill) again to another bank. This repetition means 100$ of money placed into a bank from the federal reserve creates 500$ (although not in a single bank, but spread across the banking system), whereas when 100$ is placed in a bank by an investor, only 100$ is created.
There is no way to tell between the “original” 80$ that is loaned out to the public, and the 80$ that is subsequently loaned to another bank – except when there is a run on the banks. When there is a run on the banks, people want federal reserve notes, not the promisary notes banks right to each other based on federal reserve notes – these are worth nothing to anyone except bankers – they do not represent actual M1 dollars, but rather are a mechanism to allow banks to lend more money out than there exists in M1.
The importance of this system in the U.S. has become great, although that is no argument that the system is neccesary. M1 in the U.S. is apparently 900 billion dollars – roughly half the size of the combined stiumulus packages to deal with the current crisis. However, the actual money supply is much greater than M1 – the amount it is greater depends on the reserve ratio (i.e. if it is 20%, the actual supply is something like 4.5 trillion).