RDP 2000-06: Inflation Targeting and Exchange Rate Fluctuations in Australia Appendix B: Equations for the Exogenous Variables in the Model

This Appendix sets out the estimated equations for each of the exogenous variables in the model. The specification of these equations is based on de Brouwer and O'Regan (1997).

The same general definitions apply as before (see Appendix A). Further details on the measurement and sources for each of these exogenous variables is contained in Appendix D.

All of the equations were estimated by OLS using quarterly data over the sample period: 1982:Q1–1998:Q4. Where necessary, linear trends in the exogenous variables were estimated over the slightly longer sample period: 1980:Q1–1998:Q4. Absolute values for t-statistics are contained in parentheses below the coefficient estimates.

Foreign output equation

Terms of trade equation

Farm output equation

Foreign short-term real interest rate equation

World price inflation equation

As with the endogenous variables, the constants in the equations for the exogenous variables were calibrated to place the model in equilibrium in the initial period (αwy = 0.0017, αtot = 14.5806, αfy = 0.0063, αwr = 0.0012 and αwp = 0.0019). These calibrated coefficients also ensure that the exogenous variables, except the world real interest rate, are equal to 100 in the initial period. The equilibrium world real interest rate is set equal to its historical long-run average of 1.97 per cent.

For the purposes of the simulations, potential output (yp) was treated as exogenous and non-stochastic, growing at a constant rate consistent with the notion of potential output implicit in the error-correction term of Equation (A1), that is, 1.212 times the long-run trend real output growth rate in the US (which was estimated to be 2.7 per cent per annum over the sample 1980:Q1–1998:Q4).