The objectives of this study are: a) to identify the fundamental economic structure (FES) in Indonesian regional economies; b) to develop a hybrid method for constructing regional input-output tables using the fundamental economic structure approach; c) to evaluate the performance of the fundamental economic structure approach; and d) to apply the model for constructing regional input-output tables in small regions to two districts in South Kalimantan Province.
This study tried to identify the fundamental economic structure (FES) developed by Jensen, West and Hewings (1988). The FES framework is used to capture the main features of regional economies in Indonesia. However, a slightly different approach and functional forms are used for the study. A revised approach of previous FES approach developed by Van der Westhuizen (1992) was established. This approach is used as the first approximation of regional input-output tables. The FES approach is relatively efficient in terms of using superior data.
The results of the study suggest that the fundamental economic structure in Indonesia has similar pattern with other studies such as Australia and South Africa. However, unlike Australia, Indonesia not only has significant cells in the secondary and tertiary sectors, but some of the primary sectors as well. Subsistence agriculture, Central Government food self-sufficiency policy at the regional level, island economy and limited infrastructure may be the explanations of these phenomena. In a 21-sector aggregation level, the quadratic and linear models gives the largest proportion of statistically significant cells. The independent variable (economic indicator) providing the largest proportion of statistically significant relationships is TSGO (total sectoral gross output). The quadratic model except population as an independent variable gives the highest proportion of adjusted R2.
The evaluation of the performance of the model is measured by applying several statistical methods to provide some indications of the error patterns. The results of the evaluation using different tests appear to be consistent across provinces. In most cases, the errors are relatively low on average using different measures. The FES approach is applied to construct input-output tables for Banjarmasin, a typical urban economy, and Hulu Sungai Tengah, a typical rural economy, in South Kalimantan. Banjarmasin is smaller than Hulu Sungai Tengah in terms of area. However, Banjarmasin has a greater population and economic size than Hulu Sungai Tengah.
The performance of FES approach used to construct Banjarmasin and Hulu Sungai Tengah input-output tables is satisfactory in the sense of holistic accuracy because these tables capture the main features of the economies.
The work carried out in this thesis highlighted a number of limitations to the FES approach. For example, a limited number of regional input-output tables available for the study is a major constraint to the FES approach. It is difficult to determine the characteristic of the distribution of the sample tables.