Richard Werner discussed three different views of banking during his appearance on the Tucker Carlson show:
- The financial intermediation theory;
- The fractional reserve theory (which includes the “money multiplier”);
- The credit creation theory
Werner claims to have empirically verified the credit creation theory by observing a bank making a loan. He rejects the other two theories.
The financial intermediation theory says that banks take in funds from depositors and savers and then use them as a basis for making loans to borrowers in the form of business loans, mortgages, etc. Thus, banks act as intermediaries between savers and borrowers. Werner claims that this view is dominant and has led to the economics profession, broadly speaking, improperly disregarding banks and their influence on the economy.
The fractional reserve theory starts with the financial intermediation theory but then asks, “What happens when those who receive bank loans spend it and the money is deposited in other banks?” Answer: other banks now have the ability to extend credit. Since banks keep a fraction of deposits in reserve, new deposits enable new lending. The “money multiplier” refers to the extent to which the banking system can increase the money supply based on the reserve ratio.
The credit creation theory says that banks are not mindful of reserves when making loans. Instead of receiving deposits first and then extending loans based on some desired reserve ratio, banks create new money by extending loans based only on the prospective profitability of the loan. In this way, banks create credit (and money) ex nihilo.
So, which theory is right and which two are wrong?
Unfortunately, that’s the wrong question. It assumes that the theories are mutually exclusive. A better question, if we don’t assume mutual exclusivity, is, “Which theory gives us the fullest picture or explanation of what’s going on in the banking system?”
Exhibit A
Consider an individual bank, Bank A, that receives a new deposit in the amount of $100 from Richard. Immediately, Bank A has new assets (+$100 in reserves) and liabilities (+$100 in demand deposits). The next person in line at the bank is Bob, who is looking for a loan to buy a $90 chess set. Bank A—bolstered by Richard’s deposit—considers Bob’s creditworthiness, the loan rate Bob is willing to pay, and the probability that Richard (or other depositors) will want to withdraw their deposits before Bob pays Bank A back. Bank A makes the loan.
So far, we’ve described a situation that looks like the financial intermediation theory. There are some quibbles (and unanswered quibbles, still more unanswered quibbles, and more than quibbles) between Rothbardian full reservists and fractional reserve free bankers regarding the nature of demand deposits and whether they represent depositors’ savings, but this is beside the point we’re making here. Setting these issues aside, Bank A is using Richard’s deposit as a basis for extending a new loan to Bob.
Exhibit B
The story continues: Bob takes his borrowed $90 and purchases a chess set from Tom. Tom banks at Bank B, and so he deposits the $90 there. Bank B is happy to receive Tom’s deposit, because now it can make new, profitable loans to others. If both Bank A and Bank B have decided that keeping 10 percent reserves is a prudent cushion for depositors’ withdrawal demands and interbank clearing, then Bank B will extend new loans up to $81 to borrowers. Those borrowers spend the money and deposit it in their respective banks and the cycle continues. With each step, the money supply (which includes demand deposits) grows by a smaller amount, but the growth can reach a theoretical maximum of ten times the original deposit amount. (The money multiplier in this example is ten, calculated by taking the inverse of the reserve ratio).
We haven’t introduced any outlandish assumptions. We’ve only considered what can happen as borrowers take the money and use it. The result: the fractional reserve theory of banking.
These two “theories,” then, aren’t theories and they aren’t mutually exclusive. They just describe different aspects of the way depositors, banks, and borrowers interact. The financial intermediation view looks at one depositor, one bank, and one borrower. The fractional reserve view looks at what happens when another bank receives funds borrowed from the original bank and then makes another loan—multiple depositors, multiple banks, and multiple borrowers are in view.
Exhibit C
Somewhere else in the banking system is Bank C. It is not involved in the unfolding story of Bank A, Bank B, Richard, Bob, Tom, and all the other borrowers and depositors that might be affected by Richard’s original deposit. Suppose Joe walks into Bank C looking for a $300,000 mortgage. Bank C considers Joe’s creditworthiness, the loan rate Joe is willing to pay, and the probability that their depositors will want to withdraw their deposits before Joe pays Bank C back. Bank C makes the loan.
This act by itself was not necessarily preceded by somebody making a deposit and then Bank C deciding to return to its desired reserve ratio by extending the loan to Joe. It could be that Bank C decided that their prior reserve ratio was more than sufficient to cover withdrawal demands and interbank clearing and so they made the loan simply by making new liabilities on themselves ex nihilo.
Now we have a situation that looks like the credit creation theory (or view).
Narrow and Systematic Views
Remember, however, when Bank A made the loan to Bob? That singular act looks exactly the same! All we did was insert Richard the depositor in line before Bob the borrower. We could have easily switched their positions in the queue, and it would tell a very similar story:
Consider an individual bank, Bank A, that receives a new request for a loan from Bob, in the amount of $90. Bob wants to use it to buy a chess set. Bank A considers Bob’s creditworthiness, the loan rate Bob is willing to pay, and the probability that other depositors will want to withdraw their deposits before Bob pays Bank A back. Bank A makes the loan. The next person in line at the bank is Richard who deposits $100 into his checking account…
Thus, the credit creation view has the narrowest field of vision—it only looks at the extension of a loan. It’s obvious that these three views are not incompatible with one another. They only describe different parts of the story.
Just because the credit creation view is narrower in focus does not necessarily mean that it is wrong. As Bob Murphy pointed out in our Human Action Podcast episode on this topic, the issuance of loans by fractional reserve banks is how the money supply expands. It’s worth spending some time focusing on that singular act because that is where the (dark) magic happens. The fractional reserve/money multiplier view takes a more comprehensive view and considers what enables and inhibits bank money creation systematically. See, for example, Rothbard’s Mystery of Banking (p. 97), for an example of the money-supply-increasing act of credit creation by a bank. While Rothbard highlights the fact that the fractional reserve bank’s loan creates money ex nihilo, he then takes a broader look at the banking system to explain the money multiplier in chapter eight.
Werner’s “Experiment”
In Richard Werner’s empirical test, he observed a bank’s balance sheet before and after obtaining a loan from them, testing this hypothesis:
Should it be found that the bank is able to credit the borrower’s account with the loan principal without having withdrawn money from any other internal or external account, or without transferring the money from any other source internally or externally, this would constitute prima facie evidence that the bank was able to create the loan principal out of nothing.
This, however, does not test the theories. All three theories can explain such an observation. If you narrowly focus on the singular loan, as Werner did, you will only see credit creation. If you disregard the board-level decision-making regarding anticipated withdrawal demands, interbank clearing needs, what level of reserves are sufficient, what kinds of borrowers and projects to lend to, etc., then you will never see what the financial intermediation proponents say is going on inside individual banks. If you disregard what borrowers do with the funds and how other banks end up with claims on each other and how banks may extend loans in view of a desired reserve ratio (and what happens when reserve ratios change), you will never see what the people talking about the money multiplier are talking about.
Moreover, right before the conclusion of his paper, Werner laments that “other bank transactions” transpired over the course of his observation period of two days:
The evidence is not as easily interpreted as may have been desired, since in practice it is not possible to stop all other bank transactions that may be initiated by bank customers (who are nowadays able to implement transactions via internet banking even on holidays).
But these muddling “other transactions” are critical for the other views. If you disregard Richard in Exhibit A, you “confirm” the credit creation theory. If you disregard Tom’s deposit and the follow-up lending by later banks in Exhibit B, you “confirm” the credit creation theory.
Werner constrained his experiment from the start:
A written agreement was signed that confirmed that the planned transactions would be part of a scientific empirical test, and the researcher would not abscond with the funds when they would be transferred to his personal account, and undertakes to immediately repay the loan upon completion of the test.
Before the “test,” Werner promised that he would immediately repay the loan. Of course the bank wouldn’t care about reserves! Of course they wouldn’t worry about withdrawal demands or interbank clearing! This is like “testing” your friend’s willingness to lend you his car right after signing a contract that says, “I won’t drive your car. In fact, one second after you give me your keys, I’ll hand them right back.” You haven’t tested anything except your friend’s willingness to participate in a silly “experiment.”
Moreover, the limited observation window prevented Werner from being able to confirm or reject the financial intermediation or fractional reserve “theories.” To do that, he’d have to cash the check or otherwise spend the lent money from the bank he observed (to see financial intermediation) or deposit the lent money into another bank and observe their follow-up lending activity (to see the money multiplier in action).
Werner Skips Over Some Inconvenient Text
Werner quotes Paul Samuelson in his review of the literature on the fractional reserve theory:
“Can it expand its loans and investments by $4000…?”
“The answer is definitely ‘no’. Why not? Total assets equal total liabilities. Cash reserves meet the legal requirement of being 20 per cent of total deposits. True enough. But how does the bank pay for the investments or earning assets that it buys? Like everyone else it writes out a check — to the man who sells the bond or signs the promissory note. … [*] The borrower spends the money on labor, on materials, or perhaps on an automobile. The money will very soon, therefore, have to be paid out of the bank. … A bank cannot eat its cake and have it too.
Samuelson (for all his faults) is making a good point. In the days of legal reserve requirements, banks couldn’t lend to an extent that would result in them having insufficient reserves. He isn’t saying that banks literally pull money out of reserves to make loans—he’s saying that as soon as a bank makes a loan, it expects the borrower to spend it, which means the bank must be ready to settle (by drawing down reserves) with the other banks that receive the borrowed funds. We don’t have legal reserve requirements these days, but banks still keep reserves to clear with other banks and because the Fed pays interest on reserve balances. In Samuelson’s text, you can switch out “legal requirement” for “desired level” and the logic remains.
Here’s the funny part: in the ellipses I marked with an asterisk, Werner skipped over this little sentence from Samuelson: “If all such people would promise not to cash the bank’s check—or what is the same thing, to hold all such money frozen on deposit in the bank—then, of course, the bank could buy all it wants to without losing any cash.”
To paraphrase: “If some guy walks in and asks for a loan but credibly promises not to cash the check but instead pay it back immediately, of course the bank wouldn’t care about dipping below its desired reserves and it wouldn’t need to move funds around at all.”
Samuelson anticipated Werner’s fake experiment back in 1948.
Here we find a rare point of agreement between Murray Rothbard and Paul Samuelson. Rothbard also showed that if all the lending by a bank simply remained as deposits in the same bank and if its depositors don’t run on the bank, “it can continue to expand its operations and its part of the money supply with impunity.”
Conclusion
Werner’s experiment is dubious at best. He strawmanned the alternative theories and set up the experiment in such a way that only his preferred theory would be confirmed. Even so, he admitted that the evidence wasn’t as clear as he desired due to “other bank transactions.”
These other bank transactions, including those that occurred before and after his two-day observation window, are critical parts of the theories he tried to reject. His favored theory, interpreted charitably, isn’t incorrect but it narrowly focuses on one loan by a fractional reserve bank. The fractional reserve theory, along with the money multiplier, encompasses Werner’s credit creation theory and provides a much fuller explanation of the way the banking system operates.
