Critical Remarks on R. Werner's Typology of Banking Theories
including an explanation why bankmoney is separate, but not independent from central-bank money
Richard A. Werner 2015: A lost century in economics - three theories of banking and the conclusive evidence, International Review of Financial Analysis, 1 – 19. http://eprints.soton.ac.uk/384540/?_ga=1.209791644.866290598.1476449053.
Richard A. Werner 2014: Can banks individually create money out of nothing? The theories and the empirical evidence, International Review of Financial Analysis 36 (2014) 1–19. http://eprints.soton.ac.uk/372866/?_ga=1.244861452.866290598.1476449053
Werner distinguishes three types of banking models or banking theories:
(1) the financial intermediation theory of banking, which is linked to the
loanable funds model
(2) the theory of fractional reserve banking or reserve circulation, and
(3) bank credit creation out of nothing.
Werner rejects (1) the loanable funds and financial intermediation models as well as (2) the theory of fractional reserve banking, while (3) credit creation out of nothing is considered the only correct one. This is incorrect. Even though the typology can be endorsed under various aspects, the theories under (2) represent an incomplete and partially inadequate understanding of reserve circulation, while model (3) on bankmoney creation out of nothing is all too unspecific and overstated, blinding out the connection of bankmoney with reserve circulation.
Rejecting the loanable funds model and the financial intermediation theory of banking is not controversial. Simply, a number of related aspects could be made clearer. This includes, for example, that
- interbank circulation of reserves (i.e. central-bank money on a bank
account with the central bank) and
- public circulation of bankmoney among nonbanks
represent two different circuits where reserves and bankmoney represent two different classes of money that never mingle, even though the public circuit is tied to the interbank circuit. Reserves cannot be transferred into a customer bank account, as in the reverse the deposits in a customer bank account, which is bankmoney, cannot be transferred into a central bank account.
Furthermore, banks cannot make use of customer deposits for funding bank loans or making other bank payments, and banks cannot on-lend customer deposits to other customers, or use customer deposits in another way. The reason is that a deposit is a bank liability to a customer, not a liquid asset such as solid cash and excess reserves (payment reserves) which is what a bank needs either to cash out notes and coins, or to transfer deposits (bankmoney) to customers at other banks.
It is different, however, with traditional solid cash. Banks can on-lend cash they have received from customers, as banks can pass on cash from the central bank to customers. Solid cash, however, is no longer constitutive for the modern money system which, at source, builds on cashless credits-on-account, both in the forms of reserves and bankmoney.
Beyond the banks' primary credit creation, however, there are secondary credit markets where customer deposits (bankmoney) are on-lent among nonbanks, or invested in nonbank financial institutions such as trusts or investment funds. On secondary capital markets, the loanable funds model still applies. It also applies to the interbank money market on the basis of reserves. Finally, although commercial banks are definitely no financial intermediaries, they are, so to say, monetary intermediaries, mediating the cashless payments among nonbanks by way of interbank clearing and settlement in reserves.
Werner's Model (2) does not reflect the entire truth about fractional reserve banking. The model combines reserve banking with the loanable funds model and equates that combination with the multiplier model and the reserve position doctrine. Even if inconsistent assumptions in this vein may have been present in various renowned authors, this does not justify reducing fractional reserve circulation to that mishmash and then presenting fractional reserve banking and bank credit creation as 'mutually exclusive views', while in actual fact both elements – bank credit creation, and fractional reserve circulation – go hand in hand. The existence of reserve circulation in connection with bankmoney circulation and the banks' proprietary business can simply not be denied as a matter of fact.
Similarly, Werner's type (3) – credit creation out of nothing – is unspecific and overshooting the mark. It is unspecific in that modern token fiat money, as it has no intrinsic value, is generally taken 'out of thin air', no matter whether it is about treasury coin, central-bank notes and reserves, or bankmoney. The value of modern money, its purchasing power, is conferred value relating to the prices of things money can buy, and is ultimately covered by, or anchored in, real economic output.
Serving as empirical evidence of credit creation out of nothing, Werner refers to the widespread representation of bankmoney creation by balance sheet expansion of a single bank, when that bank makes a pairwise asset-and-liability entry on its balance sheet, as a credit claim on the customer (left side) and an overnight liability to the same customer (right side). The related bookkeeping practice, however, can be different depending on the country and historical period, and even if that representation is not wrong in itself, it still does not make sense and can be misleading, because a customer does not take up money for keeping it on account, but for making payments due; and as soon as the customer withdraws the bankmoney in cash or transfers the bankmoney to somewhere else, the balance sheet expansion of the respective bank is reversed, in that the liability to the customer is closed out, and the cash account or reserves account of the bank is debited.
This reflects the fact that balance sheet expansion by bankmoney creation is not an individual act of a single bank, but a cooperative process of the entire banking sector, in that a credit claim or other asset is added on the balance sheet of a credit-creating bank, while the related liability (bankmoney) is added on the balance sheet of the recipient bank. All banks have to accept each others' liabilities transferred to them, otherwise bankmoney creation could not work.
A balance sheet expansion, both collectively and individually, is resulting from continued credit creation and mutual acceptance of bankmoney all the same. This is due to the fact that the velocity or use frequency of bankmoney and cash in the public circuit is many times slower than that of an amount of reserves in the interbank circuit; or put in the reverse, the velocity or use frequency of reserves in interbank circulation is many times faster than that of bankmoney. This is what enables the fractionality of reserve banking, that is, what enables banks to create pro-actively large amounts of bankmoney that need to be re-financed by a central bank in cash and reserves only to a very small extent, re-actively upon or after the monetary facts the banks have created beforehand.
For making a credit-and-liability entry on its balance sheet, a bank does of course not yet need reserves or cash. Would the bank in the Werner example be very huge and represent, say, half of all customers within a currency area, then about half of all cashless payments were carried out by simple internal rebooking of overnight liabilities among the internal customers. To that extent the Werner example would be right.
In the real world, however, such gigabanks do not exist. The vast majority of domestic and international cashless payments include interbank transfers among different banks. And when transferring a customer deposit into an external account with another bank, the sending bank will need to have or obtain reserves which are transferred to the recipient bank that will re-credit a deposit (bankmoney) to the recipient customer. This is all the more evident in today's real-time gross-settlement payment systems (Fedwire, CHIPS, CHAPS, TARGET2, CLS)[1] all of which are reserves-based.
Extending credit without the credit being used does not make sense, and in this regard credit creation 'out of nothing', if taken literally, is misleading. Money creation 'out of nothing' is a catchy metaphor regarding the nature of modern money. But this should not obscure the fact that creating money denominated in a specific currency, and ensuring the validity and value of that money, has many prerequisites which a bank cannot meet 'out of nothing'. One of these many requirements is the availability of a fractional amount of reserves and cash on which the banks' credit and deposit creation is still dependent. Bankmoney creation and fractional reserve circulation go hand in hand indeed, even if today's bankmoney regime is bank-led and entirely determined by the banks' pro-active primary credit and deposit creation, whereas the central banks have become anytime re-financers of the banks and their debt dealers of last resort, regularly under conditions of business as usual, and all the more so in times of crisis which inevitably occur in a bank-led reserve system.
The paper as a PDF >
[1] Fedwire = Federal Reserve Wire Network (USA; RTGS). CHIPS = Clearing House Interbank Payment System (USA; combines clearing with intraday final settlement). CHAPS = Clearing House Automated Payment System (UK; RTGS). TARGET2 = Trans-European Automated Real-Time Gross Settlement Express Transfer System (Euro/EZB). CLS = Continued Linked Settlement System, for international payments (combines continued clearing with final settlement).
The paper as a PDF >
Below a video On the Functioning of the Present Bankmoney Regime Basedon Reserve Circulation, registered at the American Monetary Institute's 12th Annual Conference in Chicago, Oct 2016