Empirical Analysis of the Determinants of FDI in Thailand : A Case Study of FDI from Singapore

Abstract: The thesis analyzes the determinants of Singaporean foreign direct investment in Thailand during the years 1981-2009, taking into account the relevant previous studies. The research comprises of analyzing the patterns of Singapore’s OFDI to Thailand, literature reviews of theories and empirical studies of Singaporean OFDI in general, and also an econometric analysis of the determinants of Singapore’s total FDI in Thailand. Singapore is the largest overseas investors in Thailand when compare to other countries in ASEAN. More specifically, Singaporean investors make the decision to invest abroad, in this case is Thailand, base on home country factors and host country factors. In terms of home country factors, Singaporean FDI in Thailand is caused by the limited Singapore’s domestic market and also rapid changing in the comparative advantages of Singapore. While, for host country factors, the econometric approach is utilized to observe in this study. In form of econometric analysis, the two regression models with time series data between the years 1981 and 2009 are employed to analyze. To be exact, in the first regression model, the Singaporean FDI in Thailand (RFDI) is treated as the dependent variable, while the growth rate of Thai domestic market (GRGDP), the Thai real wage rate (RWAGE), the relative price of Thailand and Singapore (PTS), the nominal relative exchange rate (EXR) and the dummy variable of the Asian crisis (AC) are treated as dependent variables. While, in the second model, RFDI is also the dependent variable but PTS and EXR are combined to be the real relative exchange rate of Thailand and Singapore or RER (PTS'EXR). However, in order to reach the conclusion of observing the determinants of Singaporean FDI in Thailand, OLS technique is utilized in two regression models. The OLS results present that GRGDP, PTS and RER have an influence on RFDI. Further, the long run relationship between the variables is also observed in this study by using the Johansen cointegration test. The results of the test indicate that there are conintegrating equations or the long-run relationships among the variables in model 1 and model 2. Apart from that, the Error Correction Mechanism (ECM) is also employed with the purpose of illustrating that when RFDI of two models deviate from their long- run equilibrium, an adjustment to pull the actual RFDI to the long-run equilibrium will take place. However, empirical evidences present that the speed of adjustment is rather rapid in both of models.