Despite the importance of the tradable/non-tradable division to economic theory, there has been little empirical analysis reflecting this division due to a lack of reliable data. This working paper seeks to address the lack of adequate information on the classification, output and prices for the tradable and non-tradable sectors of the Australian economy, using a classification system developed by Dwyer (1992) as a starting point.
In most previous analyses, the classification of output as tradable or non-tradable is based upon untested a priori reasoning. The output and price measures developed in this paper have attempted to address the acknowledged weaknesses of previous measures and the adopted methodology is an improvement upon previous methods. The improvements are due to the use of an objective and consistent classification system, the disaggregated level of the analysis, the ability to capture changes in the composition of sectors, the breakdown of the tradable sector into its importable and exportable components, the consideration of the choice of price index formula, and the detailed matching of price data to input-output industry.
Several important features have been revealed in Australian data. The size of Australia's tradable sector is significantly smaller than many previous estimates have indicated. The composition of the tradable and non-tradable sectors has changed significantly between 1977- 78 and 1989- 90, and the importance of the exportable and tradable sectors has increased over the past decade. The relative importance of individual industries within the tradable and non-tradable sectors has also been identified. The development of price indexes for the tradable and non-tradable sectors of the Australian economy has revealed the relative volatility of the prices of tradable (especially exportable) commodities compared to non-tradable commodities, and a significant improvement in Australia's internal competitiveness (as measured by relative domestic prices) since 1980.
Valuable information has been obtained about the size, composition and price movements for the tradable and non-tradable sectors in Australia over the period 1977 to 1995. To the extent that the adopted methodology provides an accurate representation of the tradable and non-tradable sectors, the data presented in this paper can be applied to the empirical analysis of a range of topics to which the tradable/non-tradable dichotomy is relevant.
Given the importance of this division to modelling the features of a small, open economy such as Australia, the development of this dataset opens up empirical research opportunities which the lack of adequate data had previously served to limit. Consideration is being given to releasing a tradables database, containing the series outlined in this working paper, on a regular basis. However, this is dependent upon sufficient interest being demonstrated in the sectoral profile, output and price index data compiled in this project.