# Who creates money? Central banks or commercial banks? [closed]

Which of these is a better description of money creation:

Scenario 1: Central bank creates money Customer A deposits \$10 at his bank. Customer B wants to borrow \$100. The bank is in good standing with the Central bank and is allowed to borrow 10x the amount of deposits. The bank borrows \$100 from the Central bank and lends it to customer B.

Scenario 2: Commercial bank creates money Customer A deposits \$10 at his bank. Customer B wants to borrow \$100. The bank simply loans him the \$100 thereby creating money.

I have always believed that scenario 1 is how money is created but scenario 2 is more common on the Internet. There are two issues with scenario 2 that make me doubt it. First, how is the loan to customer B recorded in a double entry bookkeeping system? An asset is created on one side, but where does the asset come from? Thin air? Second, doesn't scenario 2 imply a 0% interest rate for the bank? Since the bank isn't borrowing money, the Central bank interest rate shouldn't influence customer B's interest rate, but clearly it does.

Edit/Revision: I have googled some more but haven't gotten closer to the truth. Instead of two scenarios I present to you three alternatives. We have customer A, central bank CB and customer D (to avoid confusion). Customer A deposits \$100. Bank's balance sheet looks like this:

```+-----------+------------------+
|  Assets   |   Liabilities    |
+-----------+------------------+
| cash \$100 | A's account \$100 |
+-----------+------------------+```
Customer D wants to borrow \$80 from the bank. What happens then?
Alternative 1: The bank lends A's money to D
```+--------------+------------------+
|    Assets    |   Liabilities    |
+--------------+------------------+
| cash \$20     | A's account \$100 |
| D's loan \$80 | D's account \$80  |
+--------------+------------------+```

Alternative 2: The bank lends money from the Central Bank and gives it to customer D

```+--------------+------------------+
|    Assets    |   Liabilities    |
+--------------+------------------+
| cash \$100    | A's account \$100 |
| CB money \$80 | debt to CB \$80   |
| D's loan \$80 | D's account \$80  |
+--------------+------------------+```

Alternative 3: The bank simply creates money out of thin air:

```+--------------+------------------+
|    Assets    |   Liabilities    |
+--------------+------------------+
| cash \$100    | A's account \$100 |
| D's loan \$80 | D's account \$80  |
+--------------+------------------+```

Alternative 1 doesn't add up. Alternative 2 seems plausible but with an unnecessary step. Alternative 3, which I was skeptical to earlier, is beginning to make sense.

• Money is created by Central Bank. I haven't seen the scenario 2 you are mentioning. Jun 12, 2012 at 12:14
• Trivially 1 is the closest to being correct but not really true. How money is produced also varies by currency, and whether the currency is backed or a fiat currency. I voted to close as too local but was teetering on not constructive/not a real question. I think the question could be closed for any of those 3.
– user4127
Jun 12, 2012 at 13:17
• Its Alternative 1. But you have got the entries wrong. When Bank has lent \$80, they have just moved asset from Cash to Loan. There is no liability incurred.[ie Bank does'nt have to pay], So just remove the D's account \$80 on the liabilities and you are fine. Techincally you would need 3-4 steps to get to this level. ie first create a liability of \$80 that you need to give, and a matching asset for \$80. On payment by cash, you reduce cash by 80 and reduced the liability by 80. Jun 13, 2012 at 4:06
• Opened a discussion on meta about why this question was closed. IMHO it's an interesting one. Could we discuss more. Caouldn't we accept econmic questions as an interesting subject that adds value to this community? meta.money.stackexchange.com/questions/528/… Jun 15, 2012 at 12:23
• Note: We will be working to clarify the question of what economics discussion are considered appropriate on the site. Jun 27, 2012 at 16:05

A central bank typically introduces new money into the system by printing new money to purchase items from member banks. The central bank can purchase whatever it chooses. It typically purchases government bonds but the Federal Reserve purchased mortgage-backed-securities (MBS) during the 2008 panic since the FED was the only one willing to pay full price for MBS after the crash of 2008.

The bank, upon receipt of the new money, can loan the money out. A minimum reserve ratio specifies how much money the bank has to keep on hand. A reserve ratio of 10% means the bank must have \$10 for every \$100 in loans.

As an example, let's say the FED prints up some new money to purchase some office desks from a member bank. It prints \$10,000 to purchase some desks. The bank receives \$10,000. It can create up to \$100,000 in loans without exceeding the 10% minimum reserve ratio requirement.

How would it do so? A customer would come to the bank asking for a \$100,000 loan. The bank would create an account for the customer and credit \$100,000 to the customer's account. There is a problem, however. The customer borrowed the money to buy a boat so the customer writes a check for \$100,000 to the boat company. The boat company attempts to deposit the \$100,000 check into the boat company's bank. The boat company's bank will ask the originating bank for \$100,000 in cash. The originating bank only has \$10,000 in cash on hand so this demand will immediately bankrupt the originating bank.

So what actually happens? The originating bank actually only loans out `reserves * (1 - minimum reserve ratio)` so it can meet demands for the loans it originates.

In our example the bank that received the initial \$10,000 from the FED will only loan out \$10,000 * (1-0.1) = \$9,000. This allows the bank to cover checks written by the person who borrowed the \$9,000. The reserve ratio for the bank is now \$1,000/\$9,000 which is 11% and is over the minimum reserve requirement. The borrower makes a purchase with the borrowed \$9,000 and the seller deposits the \$9,000 in his bank. The bank that receives that \$9,000 now has an additional \$9,000 in reserves which it will use to create loans of `\$9,000 * (1 - 0.1) = \$8100`. This continual fractional reserve money creation process will continue across the entire banking system resulting in \$100,000 of new money created from \$10,000.

This process is explained very well here.

• Typically, central banks cannot purchase whatever they chose. While the US system is an exception, the rule is that the governments decide how much money is going to be printed and for what purposes. Jun 13, 2012 at 10:04
• The Bank of England started purchasing private sector assets in 2009 (bankofengland.co.uk/markets/Pages/balancesheet/default.aspx#qe). The Bank of Japan started purchasing stock ETFs in 2010 (articles.marketwatch.com/2012-05-08/markets/…). But I guess the FED, the Bank of England, and the Bank of Japan are all "exceptions".
– Muro
Jun 13, 2012 at 20:36
• Muro: from your link re bank of England: "under a remit from the Chancellor of the Exchequer". I.e.: Governmental decision, not the bank's initiative. Re bank of Japan from your link: "But Japan’s monetary authority is almost unique among its peers in the major developed economies" - yes, it is an exception. Your own link says so. Jun 13, 2012 at 20:39
• This answer is wrong: "In the intermediation of loanable funds model of banking, banks accept deposits of pre-existing real resources from savers and then lend them to borrowers. In the real world, banks provide financing through money creation." see bankofengland.co.uk/working-paper/2015/… title says it all... May 29, 2018 at 20:48

Scenario 1 is typically the better description. If commercial banks were allowed to simply "create" money, they wouldn't be in the mess they're in now.

In the U.S., the central bank is the Federal Reserve or Fed, and is the only entity (not the government, not the banks, not the people) that is allowed to create money "out of thin air". It does this primarily by buying government debt. The government spends more than it takes in, and so to come up with the deficit, it issues bonds. The Fed buys a certain amount of these bonds, and simply prints the money (or more realistically authorized the electronic transfer of \$X to the Treasury) which the government then spends. That places money in the hands of corporations and the people, who turn around and spend it. However, long-term, the interest charges on money borrowed from the Fed will actually remove money from the economy. The central banks, therefore, have to constantly make marginal changes to various monetary policy tools they have when the economy is just humming along. If they do nothing, then too much of a short-term increase in money supply will result in there being "too much money" which makes an individual monetary unit worth less (inflation), while making money too hard to get will reduce the rate at which it's spent, reducing GDP and causing recessions.

The exact scenario you describe is typically seen in cases where the government is running with a balanced budget, and the central bank thus can't give its "new money" to the government to spend when it wants to increase the money supply. In that situation, the central bank instead lowers its lending rate, the percentage interest that it will charge on loans made to other banks, thereby encouraging those banks to borrow more of the money created by the central bank. Those banks will then use the money to make loans, invest in the market, etc etc which puts the money in the economy.

In the U.S., the Fed does have this tool as well, but increases or decreases in the "Federal Funds Rate" are typically used to influence the rate that banks charge each other to borrow money, thus encouraging or discouraging this lending. A lowering in the interest rate makes banks more likely to borrow from each other (and from the Fed but the amount of money "created" this way is a drop in the bucket compared to current "quantitative easing"), and thus increases the "turnover" of the existing money in the economy (how many times a theoretical individual dollar is spent in a given time period).

Empirial evidence for the second scenario: Can banks individually create money out of nothing? — The theories and the empirical evidence. Excerpt:

It was examined whether in the process of making money available to the borrower the bank transfers these funds from other accounts (within or outside the bank). In the process of making loaned money available in the borrower's bank account, it was found that the bank did not transfer the money away from other internal or external accounts, resulting in a rejection of both the fractional reserve theory and the financial intermediation theory. Instead, it was found that the bank newly ‘invented’ the funds by crediting the borrower's account with a deposit, although no such deposit had taken place. This is in line with the claims of the credit creation theory.

Thus it can now be said with confidence for the first time – possibly in the 5000 years' history of banking - that it has been empirically demonstrated that each individual bank creates credit and money out of nothing, when it extends what is called a ‘bank loan’. The bank does not loan any existing money, but instead creates new money. The money supply is created as ‘fairy dust’ produced by the banks out of thin air.

• How can someone downvote an answer (within 3min) which sources to a peer reviewed journal? Mar 26, 2015 at 20:07
• Welcome to Personal Finance & Money! As with the rest of the StackExchange network, the goal of this site is to provide a compendium of knowledge in the form of concrete answers to specific questions. Simply posting a link to another site is not the goal. Consider amending your answer to include your own explanation of the thesis of that article (making sure to include the link as a citation).
– dg99
Mar 26, 2015 at 20:09
• You probably got the downvote when your answer was merely 6 words and a link, i.e. smelled spammy. Mar 26, 2015 at 20:19
• This is the only contribution relying on peer-reviewd empirial evidence so far...Anyways thanks for the formatting and I'm curious about your opinions.. Mar 26, 2015 at 20:42
• @MikeSchärer - this is not a discussion board. An answer to an old question with an accepted answer isn't likely to attract too much feedback. The question itself is on the edge of off-topic. It's more economic than PF. Mar 26, 2015 at 22:06