AbstractsEconomics

The Factors Affecting the Long Run Supply of Rubber from Sarawak, East Malaysia, 1900-1990: An Historical and Econometric Analysis

by Timothy Purcell




Institution: University of Queensland
Department: Department of Agriculture
Year: 1993
Keywords: rubber; rubber supply; Sarawak; econometric analysis
Record ID: 1032262
Full text PDF: http://espace.library.uq.edu.au/view/UQ:206177


Abstract

The factors affecting the supply of rubber from Sarawak, East Malaysia, were identified and reviewed in an historical framework. A methodical framework for the general analysis of economic relationships between variables was reviewed and a practical application of the methodology to the supply of rubber from Sarawak was carried out. An econometric analysis of the long run factors affecting the production of rubber was carried out. (1) Two log-differenced autoregressive models of the rubber supply were formulated. (2) The models were tested for parameter constancy to identify structural breaks in the time series and for structural invariance to determine whether they were suitable for policy analysis, forecasting and backcasting. (3) The variables were tested for bivariate Granger Causality to determine the relationships between the factors of production and the output of rubber. (4) Forecast Error Variance Decomposition analysis of multivariate Granger Causality was carried out using a Vector Autoregressive Model. The results confirm the a priori economic theory that long run changes in supply have been affected primarily by changes in area under rubber production and long term price trends. The area planted to rubber has depended upon price incentives and the availability of scarce labour resources. Prices have been affected by the supply of rubber from Sarawak but this is posited to be a reflection of global supply trends affecting prices. While the results generally confirm the economic theory, caution is urged when interpreting the results. The severe inadequacies of the data used highlights the need for more accurate time series and the mainly methodological approach of this study.