Modelling and forecasting the demand for automobile petrol in Australia, and its policy implications.

This paper presents different econometric modelling systems to estimate the demand for petrol in the Australian road transport sector, emphasizing the effects of national income and petrol price. Quarterly time series data for Australia over the period 1977-2006 are employed to capture the adjustment process associated with responses through time to changes in those factors. Eight modelling systems for petrol demand are used to compare the forecasting performance of different approaches. The most appropriate model is selected to predict automobile petrol demand in Australia from 2007 through to 2020. The prediction of a 14-year forecasting horizon shows that Australian automobile petrol consumption will continuously increase under the business-as-usual scenario, and increased greenhouse gas emission (primarily CO2) will be produced and emitted into the environment. Effective policy instruments need to be implemented to contain and then reduce the emissions from automobiles. TRESIS, an integrated transport, land use and environmental strategy impact simulation program, is used to estimate the impacts on CO2 of several policy instruments. Given the findings from the evaluation of different policy scenarios, the appropriate strategies are suggested in order to contribute to reducing greenhouse gas emission in Australia. (a) For the covering entry of this conference, please see IRRD abstract no. E216383.