RDP 7702: Inflation and Economic Stability in a Small Open Economy: A Systems Approach Appendix B. Simulation Procedures

Section 3 of the paper discusses five simulations of the model of the Australian economy set out in Appendix A. The period for the simulations, which are all dynamic non-stochastic simulations, is 1966(3) to 1975(4).

(1) The Control Solution

This has all exogenous variables taking on their historical values, and all lagged endogenous variables are the solution values from the previous period. Any errors are thus allowed to cumulate, in contrast to their treatment in a non-dynamic simulation.

(2) Managed Float

In this simulation, the dummy variables in the exchange rate equation (no. 21) which represent the timing of the major changes in parity which occurred in the sample period (QER, QUS) are set equal to zero, and the authorities are assumed to react continuously, rather than at discrete times, to the level and changes in the money stock, international reserves and to the level of prices relative to world prices.

It is assumed that if such a managed float had been adopted, the capital controls imposed in 1972 would not have been necessary, and hence the dummy variable (QF) representing these is also set to zero. It is also assumed that the behaviour of exchange rate expectations would have been different, and in particular the dummy variable indicating the degree to which expectations built up prior to each of the major changes in the exchange rate (QE) is also set equal to zero.

(3) Managed Float with Tighter Honey

The changes made in (2) are made, and in addition higher interest rates on government securities are imposed in 1972 and 1973, and the aggregate bank liquidity ratio is not reduced from its value at the start of the simulation period. It is assumed that a policy with higher official and bank interest rates early in the inflationary period would have prevented bank lending from responding to demand, and accordingly, the parameter on in equation 14 is set equal to zero. This package represents a somewhat clumsy way to simulate a tight money policy within the traditional institutional framework, and it might eventually be more convincing to simulate such a policy by introducing explicitly an interest rate on bank lending, which would be market determined. There are of course some difficult issues to be resolved before such a major change of structure could be imposed on the model.

(4) Managed float, tighter money and tighter budgetary policy

The changes made in (2) and (3) are made, and in addition government outlays are assumed to increase steadily from their 1966(3) values as follows:

The growth rates in these aggregates are approximately those achieved in the second half of the 1960's.

(5) Managed float, tighter money, tighter budgetary policy and steady growth of award wages

All of the changes in simulations (2) to (4) are again made, and real award wages are assumed to grow at the rate of growth of productivity indicated by the estimate of λ₃.