Credit divisor

When the monetary base is endogenous, the direction of causality between that variable and loans or deposits is reversed. Hence, the orthodox concept of the money multiplier (see Phillips, 1920; Cannan, 1921; Crick, 1927) is replaced in most heterodox theories with the credit divisor, a concept developed by Le Bourva (1962 [1992]), a prominent French monetary theorist. As pointed out by Lavoie (1992b), who translated Le Bourva's original 1962 paper into English, the specific phrase diviseur de credit, or "credit divisor", first surfaced in a later article by Lévy-Garboua and Lévy-Garboua (1972, p. 259). Nonetheless, these authors attribute their turn of phrase to a suggestion by Le Bourva (1962 [1992], p. 259) himself.

Le Bourva wrote mostly about "overdraft economies", to wit, systems typified by companies that were always in debt to banks, and banks that were perpetually in debt to the central bank. In such economies, banks supply credit to creditworthy customers on demand at a fixed rate of interest, up to given credit limits. Lavoie (1992a, pp. 174 and 207-10), Renversez (1996) and others have further generalized the divisor concept to a more "financialized" economy.

Following Renversez (1996, p. 477), one version of the credit divisor relationship is:

\( \displaystyle M=d \times L = \left( \frac{\left( 1-g \right)}{1- \left( e+f \right)} \right) L \)

where d is the credit divisor, L is bank loans, M is a narrow monetary aggregate (bank- notes and demand deposits held by the public), e is the ratio of "vault cash" to retail bank deposits, f is the ratio of (required) reserve balances to deposits, and g is the household sector's ratio of banknotes to deposits (see also, inter alia, Goodhart, 1995). In general, the above equation can be read as an identity. If the parameters e, f and g are assumed constant, it can also be read as the reduced form of a simple model. Then, defying the "base money multiplier" approach, the equation above is read from right to left, making credit the endogenous variable. Even in more complicated models in which the relevant coefficients are functions of rates of return, liquidity preference and so on, this divisor approach comports well with the adages according to which "loans make deposits" and "deposits make reserves". The crux of the matter is the endogeneity of the monetary base.

Renversez (1996) describes the model-based interpretation of the above equation with constant parameters as the "strong form" of the divisor theory. The behavioural parame- ters e, f and g of course change over time, implying that the divisor itself changes, but the French overdraft economists believed that they were approximately constant in a system such as theirs over an appropriate run. Moore (2006, p. 204) claims that, in practice, "there is considerable week-to-week variation in the money multiplier (m ) over the short run. But for periods of one quarter or one year (m) is empirically highly stable". Some Institutionalists, Minskyans and post-Keynesians emphasize that the equation above is merely an identity, insisting on a strong proviso that the relationship between bank lending and the monetary base varies over time with financial innovations and changes in institutions (see Niggle, 1991; Palley, 1996).

Hence, Renversez's (1996) "weak form" of the divisor, which is most relevant to an asset-based system, allows for asset demands that depend on interest rates and other variables. In general, the analysis embodied in the credit divisor remains relevant in an asset-based system, in part because open-market operations provide short-term paper on demand at a policy-determined rate of interest. Indeed, central bankers (see for instance Constâncio, 2011) have occasionally gone on record in favour of the "divisor" interpre- tation. While the originators of the "divisor" looked to credit controls as the primary means of preventing excessive credit growth, modern-day policy regimes in central banking often place reliance on interest-rate rules.

Despite the reservations insisted upon by sceptics, Le Bourva's analysis is still applica- ble in some form to all modern monetary systems. As Moore (2006, p. 208) categorically states,

[i]n the real world, CBs [central banks] do not exogenously increase or decrease the supply of credit money by expanding or contracting the high-powered base as the money-multiplier analysis asserts. CBs continually smooth security prices to ensure system liquidity at their tar- geted level of interest rates. The CB sets BR [bank rate] and the supply of credit money varies endogenously with changes in the demand for bank credit.

Recent history has illustrated that, for example, "quantitative easing" and other "uncon- ventional" policy measures change the composition of private sector balance sheets, but nonetheless any undesired balances thereby created return to the banking system, as dictated by the so-called "law of reflux".

See also:

Endogenous money; High-powered money; Interest rate rules - post-Keynesian; Money multiplier; Open-market operations; Quantitative easing; Reflux mechanism.