RDP 2001-02: Changes in the Determinants of Inflation in Australia Appendix B: First Stage Import Price Equation

The first stage pass-through relationship is estimated from the March quarter 1985. Results from an error-correction equation are shown below. In the equation, linear homogeneity is accepted and has been imposed.

Table B1: Explaining Import Prices(a)
Estimated from 1985:Q1 to 1999:Q3
  Coefficient Std error
Independent variable
Constant 153.2712 40.2754***
Import prices (lag 1) −0.3345 0.0889***
World prices (lag 1) 0.3345 0.0889***
Exchange rate (lag 1) −0.3345 0.0889***
Change in the exchange rate (lag 0) −0.6572 0.0303***
Change in the exchange rate (lag 1) −0.0998 0.0462**
Change in world prices (lag 0) 0.5669 0.2115***
Change in world prices (lag 1) 0.3887 0.2113*
Trend(b) −0.1077 0.0261***
Dummy 1998:Q2–1999:Q3(c) −2.7964 0.6993***
Long-run elasticities
World prices 1  
Exchange rate −1  
Adjusted R2 0.9199  
Residual autocorrelation LM(4)   {0.0715}
Breusch-Pagan heteroscedasticity test   {0.0022}
Jarque-Bera normality test   {0.0302}
Linear homogeneity(d)   {0.9843}

Notes: (a) Equation is from Beechey et al (2000). ***, ** and * represent significance at the 1, 5 and 10 per cent levels. Numbers in braces {} are p-values. All variables in log-levels are multiplied by 100 (so growth rates are in percentages). Given the evidence of heteroscedasticity, the standard errors reported are White heteroscedasticity-consistent standard errors.
(b) The trend captures the shift in imports towards lower-priced goods from non-G7 countries.
(c) The dummy captures price undercutting by Asian exporters following the Asian crisis.
(d) Linear homogeneity implies that the coefficients of the world price and exchange rate are 1 and −1, i.e. the PPP restriction holds.