![Sample our Global Development journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5929/TOC-Subject-Sample-2021-Global-Development.jpg"})
Abstract
Inflows of foreign direct investment generate externalities that spill over to domestic firms and raise their productivity. This article examines the extent of spillover effects of foreign direct investment for firms in the highly disaggregated garment (ISIC 3221) and electronics industries (ISIC 3832) in Indonesia. Both are export-intensive industries, but differ greatly in technological sophistication and labour intensity. Changes in both the productivity level and rate of growth in each industry are decomposed into the effects of technological change, technical efficiency change and scale efficiency change and then the impacts of spillovers on each component and on total productivity are estimated. The findings suggest that foreign direct investment generates a positive effect on total productivity change, technical efficiency change, technological change, and scale efficiency change in the garment industry. In contrast, foreign direct investment contributes significantly negatively to total productivity, technological change and scale efficiency change, but has no significant effect on technical efficiency change in the electronics industry.
Acknowledgements
The authors are grateful to all anonymous referees as well as the editor of this journal for helpful comments and suggestions which tremendously improve the quality and presentation of this article. However, the usual disclaimer applies.
Notes
1. The garment sub-sector in this study is defined as an industry that produces clothes from textiles (ISIC 3221).
2. The OPEC fuel prices are converted from US$ values to Indonesia rupiah (IDR) using average yearly exchange rates published by the central Bank of Indonesia in Statistics of Economic and Finance Indonesia (Statistik Ekonomi dan Keuangan Indonesia or SEKI).
3. One might suspect that there is multicollinearity between FDIHorizontal variable and Age variable, as these two variables are plant-invariant variables. Pearson's partial correlation test is applied to test the multicollinearity. It is found that the partial correlation between the two variables is 0.149 for garment industry and 0.349 for electronics industry. These results imply a possibility of a slight multicollinearity between the two variables in electronics industries. However, slight multicollinearity is not a problem as long as the correlation between independent variables in a model is lower than the correlation between each of the independent variables and the dependent variable (Klein, Citation1962; Gujarati, Citation2005).
4. The technical efficiency (TE) indexes are calculated from the estimates for all samples of garments in . Each firm at each time period has a unique TE index, and the averaged TE index for foreign and domestic firms are calculated using an arithmetic mean.
5. To check the robustness of the estimates in , the authors estimate an alternative model of the stochastic frontier method based on a two-stage approach of Cornwell et al. (1990). The estimates are consistent with the results in the main analysis, both for the full sample and for the sub-sample of domestic firms. However, the level of significance is lower in this two-stage approach, demonstrating that the power of estimations falls when the two-stage approach is applied. The two-stage approach has a well-known problem related to the assumption of one-sided TE, which may lead to inefficiency in estimation (Wang and Schmidt, Citation2002). The detailed results of this robustness check are available upon request.
6. Only estimates for all firms are shown. Separate estimates for domestic firms only, which are not statistically different, are available from the authors.
7. As argued in Liu (Citation2008), the level effect of FDI spillovers reflects the long-run impact of FDI on firms' productivities, while the growth effect pictures the short-run impact.