RDP 8811: Monetary Transmission in a Deregulated Financial System 3. Deregulation in Existing IS/LM Models

Having outlined the directions in which financial innovation and deregulation are taking the economy, the paper now reviews how one can incorporate such changes within standard macroeconomic models.

The open economy IS/LM model (in logs) can be described by the following equations:

where y is real output, r is the interest rate and m the nominal money stock. The exchange rate (e) is measured as the domestic price of foreign currency (i.e. a rise represents a depreciation). A dot over a variable is a time derivative, a star represents foreign variables and a bar signifies long-run steady state values.

Equations (1) and (2) are the IS and LM equations with standard assumptions concerning parameter signs. Equation (3) is a version of the interest parity condition.[2] The open economy extensions are not necessary for the discussion that follows, but do make the model resemble the Australian case more closely. Prices are held fixed at this stage, so the analysis should be thought of as short-run.

A monetary policy expansion in such a framework is illustrated in Figure 1. The exact solution (in a deterministic world) is given by;

Figure 1
Figure 1

An increase in the money supply directly shifts right the LM curve and lowers the interest rate. Lower interest rates stimulate the interest sensitive component of domestic demand, bringing forth a rise in output. The lower domestic interest rate also depreciates the exchange rate (which shifts out the IS curve) and reinforces the rise in income. If rational expectations are imposed on the model (with the exchange rate allowed to overshoot), then income may rise even further on impact, putting upward pressure on interest rates. As shown in Dornbusch (1976), it is possible that all of the transmission process goes through the exchange rate, and the interest rate may actually rise following a monetary expansion. However, this result requires perfect capital mobility and a very large short-run response from the trade balance to exchange rate movements; assumptions that are not supported by empirical work.[3] It is generally accepted that the short-run response to a monetary expansion is a fall in the interest rate and a depreciation of the domestic currency. Both of these, in turn, act to raise domestic income, by stimulating interest sensitive domestic spending and providing a competitive boost to net exports.

While this IS/LM result has been criticized in a number of areas, the fundamental transmission mechanism it embodies remains the standard workhorse in most textbooks and policy debates.[4] Indeed, the attraction of monetary targeting derives largely from the above framework, under the assumption that shocks to the LM curve are smaller (or more predictable) than those affecting the IS curve, and that the LM curve is not flat.[5]

How does financial market deregulation complicate the standard IS/LM result? Recent work by Tobin (1987) summarizes the standard response.[6] As he pointed out, the major effect of deregulation has been the payment of market-related interest rates on almost all forms of transactions balances. This renders the usual definitions of “money” insensitive to changes in interest rates. In the extreme, the interest rate drops out of the LM curve (λ=0), which becomes vertical. The “money” supply side is fixed by reserve requirements and the demand for “money” depends only on income. The classical quantity theory is reinstated and the case for monetary targeting would appear to be even stronger. The solution for a monetary expansion is now very simple;

As shown in Figure 2, a monetary expansion again shifts the LM schedule to the right and now directly raises income in proportion to the rise in “M”. There are no longer problems of allowing for unwanted shifts in velocity because the fall in interest rates does not feed back onto the demand for money. According to this model, the monetarist's quantity theories should be strengthened via the process of deregulation, yet all around the world central banks are giving up the strict reliance on monetary targeting as an operating procedure. Where is the explanation?

Figure 2
Figure 2

One explanation is that theoretical economists have tended to be somewhat vague in their definition of exactly what they mean by “money”. In the strict textbook sense, money is a liquid asset that pays no nominal return. Its quantity can easily be controlled either by changing its supply relative to the bond stock, or moving the rate of interest on other assets. (Both actions are identical from an open market operations point of view). In fact, the sort of assets that are classed as “money” in the Tobin-type framework do not resemble at all the textbook definition. The vertical LM curve is built on a broad definition of “money”, (intermediated assets that return market rates of interest) with reserve requirements constraining the supply side. In a deregulated world such aggregates would not be under the direct control of the central bank. Nor would they necessarily have a stable relationship with nominal income.[7]

Tobin argues that such a broad aggregate could be brought under the control of the central bank as long as reserve requirements are placed on all financial intermediaries. Even if this were possible (it was argued above that it is unlikely), the vertical LM schedule is still not the most appropriate framework for analysing the transmission mechanism in a deregulated financial system. The reason being that it is not the supply of required reserves that is the crucial variable under the control of the central bank that causes changes in money and then activity.[8] Rather, in a deregulated system it is the price that banks pay for excess reserves (i.e. the interest rate) that is the variable being controlled by the central bank and the variable that ‘causes’ changes in activity (and, in turn, the demand for some broad definition of money).[9] Even those central banks that have institutional arrangements more suited to base money targeting have been quick to point out that interest rates are still the crucial control mechanism. To quote the Bundesbank:

“.......it is the nature of the complex process of money creation in which the central bank, credit institutions and non-banks are all involved that the Bundesbank can work only indirectly towards ensuring that the central bank money stock develops along the envisaged lines, by the appropriate fixing of interest rates and the other terms on which it constantly makes central bank balances available.” (Bundesbank: 1982, p. 88)


Using the Dornbusch (1976) definition of exchange rate expectations; that is E Inline Equation. [2]

While a rising interest rate is not likely following a monetary expansion, it is also true that the exchange rate should be playing an increasing role in the transmission process. A recent IMF report concludes: “In comparison with earlier periods, the growing responsiveness of capital flows and exchange rates to domestic policy actions implies that … (monetary policy).. actions tend to be felt relatively less on domestic activity and more on external variables (e.g. exchange rate and current and capital account flows). Within the domestic economy, it is, therefore, possible that the incidence of monetary policy has tended to shift principally from such sectors as housing and fixed investment (typically with an important share of non-traded goods) more toward the export- and import-competing (tradable goods) industries”. (Watson, et. al., 1988, p. 46.) [3]

There are at least two further transmission channels that have been seen as important in traditional models. The first is a real balance effect. This allows real money to directly enter the IS curve via its effects on private spending, i.e., an increase in the real money stock raises private sector real wealth and, therefore, private consumption. Laidler (1984) reviews the origins and underpinnings of this approach to the transmission mechanism. A second effect is credit rationing. In a world in which banking interest rates are controlled, those sectors that rely heavily on credit are largely at the mercy of the central bank in regards to the quantity of funds available. Up until recently, the credit rationing mechanism was seen to be of crucial importance to the implementation of monetary policy in many countries (Bank for International Settlements: 1984). Both of these mechanisms reinforce the interest rate and exchange rate channels discussed above, but are likely to be of little relevance in a deregulated financial system (Morris, 1988). [4]

See, for example, Poole (1970). [5]

See, also less formal discussion of these issues in Moses (1983) and Akhtar (1983) and Akhtar and Harris (1987). [6]

Tobin (1987) does recognise the likely instability in income velocity and does not recommend strict monetary targets. [7]

See Macfarlane (1984) for a full discussion of the role of reserve requirements in Australia and Battellino and Macfarlane (1987) for some overseas examples. [8]

Rogers (1987) points out that there has always been an academic school of thought supporting the central banker's arguments on monetary control. This goes back to Wicksell, but can be traced through the writings of Keynes. [9]