RDP 8301: Financial Innovations and Monetary Policy, A Preliminary Survey IV Problems of Definition and Controllability

The upshot of disturbances to the shape and position of the LM function might be to weaken the case for using intermediate money-supply targets, in contrast with earlier decades when the LM schedule was thought to represent more stable and predictable behaviour than the real goods market behaviour behind the IS schedule. Such disturbances raise the thorny issues of what should be defined as “money” and what policy-makers should seek to control.

We observe that when policy presses to decrease the money supply, innovations arise to offset it. (Or on another plane, we could just as well speak of the decrease and increase of the rate of growth of the money supply.) Many economists regard this correspondence between monetary regulation or control and financial innovation to be almost as inevitable as mushrooms after rain.[9] Charles Kindleberger put it in the following way:

My contention is that the process is endless: fix any Mi and the market will create new forms of money in periods of boom to get around the limit and create the necessity to fix a new variable Mj. (Kindleberger, 1978, p. 58)

In 1936, Henry Simons commented on the feasibility of controlling the quantity of currency and current deposits in a similar tone:

The fixing of the quantity of circulating media might merely serve to increase the perverse variability in the amounts of “near moneys” and in the degree of their general acceptability just as the restrictions on the issue of bank notes presumably served to hasten the development of deposit (checking-account) banking. (Simons, 1936, p.5)

It becomes clear that when institutions or instruments are developed which “monetise” credit in new ways but are excluded from the definition of money, the quantity of money as defined might not grow although its velocity could rise. Therefore, in spite of the comforting “accounting” view that innovations will not disrupt monetary policy since almost everything has to end up in M3 again, we must be alert to the structural changes beyond our definition. Such changes displace the effective LM curve, alter velocity and disturb monetary control.

In response to such structural change a new variable Mj could be fixed, as Kindleberger suggests. New assets which act as very close substitutes for that which was previously defined as money might now be included in the monetary aggregate. In this way, M1 in the United States expanded to cover NOW accounts. Similarly, the U.K. authorities have adopted monetary redefinitions based on a characteristics approach rather than along institutional lines. Three characteristics highlighted in the U.K. exercise were term to maturity, size of deposit and purpose of the deposit. One might also want to consider the risk of default of the deposit-taking institution as a relevant characteristic although this last factor might complicate and reverse the aggregation and categorisation achieved under the first three characteristics.

However, the caveat relating to redefinition concerns the extent to which new institutions and instruments can be influenced by policy so that control over the total money supply is maintained. To begin with, as the definition of money widens to maintain stability in velocity, the strength and directness of control by the central bank over the aggregate declines. To counter this problem, it has been suggested that reserve requirements be imposed on all institutions offering close substitutes for money. So that further circumventive innovation is discouraged, market rates of interest should also be paid on both required reserves and demand deposits (see Cagan, 1979). Nevertheless, the costs of data collection, processing and time lags naturally increase as monetary aggregates encompass more institutions and instruments and ultimately this might begin to impair the effectiveness of Policy. Some trade-off with coverage must be admitted.

Finally, it is pertinent to repeat that – according to the earlier discussion of the second factor altering the slope of the LM curve – as more financial assets pay market-related rates of interest, the effectiveness of monetary policy diminishes. In other words, monetary redefinitions might correct for shifts in the LM curve if the shift factors could be predicted and estimated;[10] but the factors changing the slope of the LM schedule cannot be so reversed. Of M1 in the US, to take the most advanced example, David Lindsey argues that the development of interest-bearing current deposit forms “has made savings-oriented motives a more important influence on M1 demand than they have been in the past. Thus M1 demand in the future could respond differently to movements in economic variables than it has historically” (Lindsey, 1982, p.261; see also Morris, 1982, p. 11 n. 5). Therefore merely redefining money is an insufficient response to the impact of financial innovations.


Kane (1981) has suggested such a pattern and labelled it the regulatory dialectic. [9]

An example is the likely boost to M1 in the U.S. during October 1982 when the issue of All Savers Certificates matures. We know from this case that estimation of the impact on money is virtually impossible. [10]