Money multiplier makes no sense

November 14, 2005 10:28 PM

Any economists here? I have a logic / semantics puzzle. An article about money creation says that when the Fed 'gives' the banks $1000, the total amount somehow becomes a lot higher through something called a money multiplier. But the mechanics of this make no sense.

Basically if banks have a 12% reserve requirement, they have to keep 12% but can loan out the rest. So if Bank A gets $1000, it can loan out $880 to Bank B. Bank B gets the $880 and can loan out $774.70 to Bank C, ad infinitum.

Supposedly this is creating $7333.33 of additional money. A lot of articles and books talk about how this creates additional money, e.g. "each dollar of new reserves enables the banks to create ten dollars of new money".

But it seems to me that all that's happening is $880 is being divided out into smaller pieces. If someone goes to Bank B to take out a loan, it won't have the money because it loaned out its $880 to Bank C. So you can't add it as part of the total. I don't see any new money being generated; just the original amount being sliced and diced.

What am I missing here? Or am I looking at this the wrong way?

posted by rolypolyman to law & government (16 comments total)

I think your mistake is that the money multiplier effect isn't when the money is going directly from bank to bank. Instead, it is actually entering the economy between steps.

That is, Bank A lends $880 to some business/individual, they buy $880 worth of product/service, and the $880 finds its way into the seller's bank account. Then, the seller's bank can now loan out $774, etc., etc...

(IANAEconomist)
posted by kickingtheground at 10:53 PM on November 14, 2005

Did the article actually state the Fed gives the banks the initial $1000? I can only see this sort of thing happening with deposits from individuals. Rather than keeping that money as is, the bank is free to invest all but the reserve requirement. So if it keeps a certain amount in another bank, probably a larger one that doesn't even deal with individuals, a certain amount would have to be kept before that larger bank could then invest it, etc..

So depositer A has $1000 on their investment ledger, earning interest as we'd expect. Bank A takes that and keeps the reserve requirement ($120) and invests $880 in Bank B. They try to turn a higher interest on it than their paying the depositor, which is partly how banks stay in business. Anyway, Bank A invests the $880 with bank B, which now has $105.60 in reserve and invests $774.40, etc.

With these alone there would be $1000 on the individual's investment ledger, $880 on Bank A's (since that is what they borrow from the individual and invest), and $774.40 with Bank B. So $2654.40, even though it is really just $1000 at the begining.

The reserve requirement is what brings order to the chaos that this appears to be. The higher it is the less risky the whole enterprise is. I believe 10% is what the Swiss do. The other day someone told me that was what the US had as well, but we've been reducing it to 1% over the last 6 decades, in affect creating a whole lot of money which would displace inflation over those decades. I haven't had a chance to look into that though.
posted by jwells at 6:37 AM on November 15, 2005

Oh, and I'm not an economist but I work at a business school, and have a business degree from the school in computer information sciences. And yes, that is as strange as it sounds.
posted by jwells at 6:39 AM on November 15, 2005

Wikipedia tells me that the US reserve requirement is actually 10% like the Swiss.
posted by godawful at 6:44 AM on November 15, 2005

Basically if banks have a 12% reserve requirement, they have to keep 12% but can loan out the rest. So if Bank A gets $1000, it can loan out $880 to Bank B.

I think this is the source of your confusion. The bank has to have 10% or 12% of a loan amount in the vault according to the reserve requirement. Therefore, if they come by $1,000, they can loan out $10,000 or $8,333.33 depending on the reserve requirement.
posted by yerfatma at 7:04 AM on November 15, 2005

If bank A gets $1000 and has to keep 12% of its loans on hand, then it can keep that $1000 on hand and loan out $8333.33 of other people's money. So the money that's added to the non-bank economy is $7333.33.
posted by nicwolff at 7:10 AM on November 15, 2005

As yerfatma just said. I previewed, I swear!
posted by nicwolff at 7:11 AM on November 15, 2005

But it seems to me that all that's happening is $880 is being divided out into smaller pieces. If someone goes to Bank B to take out a loan, it won't have the money because it loaned out its $880 to Bank C. So you can't add it as part of the total. I don't see any new money being generated; just the original amount being sliced and diced.

In your example above the person who deposited their cash into Bank B still has a right to the whole deposit back (traditionally instantly for a current account or with a delay for a savings account) even though Bank B has loaned it out to another person. In that way both the initial depositor and the borrower have the cash.

The money "creation" is easier to see once you spot that you still have a right to your entire deposit even though the bank doesn't have the whole amount in its vaults.

(On preview, in this context yerfatma and nicwolff are wrong. It's 10% of the deposit that must be held. Wikipedia's explanation is as adequate as ever and the worked example, er, works.)
posted by patricio at 7:13 AM on November 15, 2005

Nuts, I was wrong. And it seemed so intuitively right! I hereby retract my name for consideration for Chairman of the Federal Reserve.
posted by nicwolff at 7:38 AM on November 15, 2005

Ah well. At least we were wrong together.
posted by yerfatma at 7:41 AM on November 15, 2005

I took the basic macroeconomics class in collage, and I remember something like this.

I don't think what rolypolyman has anything to do with the fact that the bank can loan out more money it has in reserve (which is what nicwolff and yerfatma are talking about).

Rather he's talking about the fact that ever dollar put into the economy goes through multiple hands, and each step increasing the economy in general.

Think about the GDP, the GDP is the sum of all the things produced in the country.

So if I build a chair and sell it, that value gets added to the GDP.

If bob buys one of my chairs on his credit card (a loan from the bank) for $100, that $100 gets added to the GDP. but now I've got to pay some taxes, save some money, and buy some more wood. So lets say the tax rate is 10%, and so is my savings rate. That means I've got $80 left over.

I'll use that $80 to buy some wood. now, the $80 worth of wood I just bought to make another chair also goes into the GDP.

The woodsmith then saves some money and pays some taxes, and is left over with $64. He spends that money on gas for his obnoxious H2 with 22" chrome rims. That $64 then goes into the GDP.

So in just three steps from $100, the GDP has increased $244. But it's going to keep going. Not forever, because at each step taxes and savings are taken out. (Of course, banks can loan out people's savings and start the whole process over again).

It's a convergent series, and that means that it has an absolute sum.
posted by delmoi at 7:47 AM on November 15, 2005

On post, it seems I have a different definition then wikipedia. If what I've described isn't called the "money multiplier" what is it called?

That's what I remember from econ...
posted by delmoi at 7:53 AM on November 15, 2005

delmoi - that's the other multiplier: the spending multiplier, which is a feature of the Keynesian model rather than a monetary effect.
posted by patricio at 8:31 AM on November 15, 2005

William Greider has a chapter in his book Secrets of the Temple that explains this whole money creation process and demystifies The Federal Reserve system.
posted by hortense at 8:44 AM on November 15, 2005

Just for reference, a more detailed explanation of Modern Money Mechanics from the Federal Reserve Bank of Chicago. It's a bit out of date on the history of reserve requirements changes (I think they've since been lowered to zero for the "low reserve tranche", which is now larger than it used to be, or something like that). But it's got lots of nifty charts.
posted by sfenders at 9:56 AM on November 15, 2005

Money multipliers blew my goddamn mind the first time I learned about them.

It's like our whole economy is built on a fourth-grade math trick.
posted by Yelling At Nothing at 12:44 PM on November 15, 2005