Technology spillovers of FDI in ASEAN sourcing from local and abroad

Purpose – This paper aims to empirically analyze the role of FDI technology spillover effects in the development patterns of ASEAN. Design/methodology/approach – The paper is based on the analytical framework of Borensztein by utilizing time‐series data between 1990 and 2008 in ASEAN countries. Models (1)‐(3) estimate parameters by adopting OLS, and Model (4) uses dummy saturation methods in PC Give, Oxmetrics 6 to determine significant dummy variables automatically. By judging the overall significance of each variable and the model, Model (4) has a substantial improvement, and is finally selected to judge the education threshold of each country. Findings – The empirical results render support to the existence of technology spillovers in ASEAN. The regression results also suggest that it is necessary to combine FDI technology spillovers with domestic human capital, that is, in the process of promoting economic development. China's FDI in ASEAN requires lower education threshold, and also has positive effect on economy growth in six countries of ASEAN. Research limitations/implications – As this paper investigates data from many countries, there are some unavoidable differences in the statistical sources and the calibre of data. It is inevitable to have some inaccuracies. This paper tries to process the data into similar calibres and units. Practical implications – The paper examined whether total FDI, intra‐ASEAN FDI and FDI from China have played exactly the same role in ASEAN countries. Accordingly, the paper put forward suggestions after combining the characteristics of FDI from China with different local situations of ASEAN host countries. Originality/value – The paper, respectively, takes data of total FDI, intra‐ASEAN FDI, FDI from China into the FDI variable in the model, and get three sets of results for each country. The year dummy variable is newly added to the original framework of Borensztein in order to test whether the dependent variable can be affected to a large extent by potential great changes in macro‐economies.