FDI, Exports, and GDP in East and Southeast Asia — Panel Data versus Time-Series Causality Analyses

Foreign direct investment (FDI) discussed in Chapter 2 is only one of the growth factors in economic development. Other major factors are exports and gross domestic product (GDP). What are the causal relations among these three factors? Economists often take two of the three variables at once and find the effects of one variable on the other using time-series analysis, presumably because that simplifies the analysis. Thus, each country has different relations and no unified theory emerges. In this chapter, we consider all three factors simultaneously and compare the results from time-series analysis and panel data analysis. For the first time in the literature, we present the theoretical framework of the vector autoregression (VAR) form for the Granger causality tests. We then use time-series and panel data from 1986 to 2004 to examine the Granger causality relations between GDP, exports, and FDI among China, Korea, Taiwan, Hong Kong, Singapore, Malaysia, the Philippines, and Thailand and the eight rapidly developing East and Southeast Asian economies. After reviewing the current literature and testing the properties of individual time-series data, we estimate the VAR of the three variables to find various Granger causal relations for each of the eight economies. We find that each country has different causality relations and, therefore, the analysis does not yield general rules. We then construct the panel data of the three variables for the eight economies as a group, and we use the fixed effects and random effects methods to estimate the panel data VAR equations for Granger causality tests. The panel data causality results reveal that FDI has unidirectional effects on GDP directly and indirectly through exports, and there also exists bidirectional causality between exports and GDP for the group. Our results indicate that the panel data causality analysis has superior results over the time-series causality analysis. Economic and policy implications of our analyses are then explored in the conclusions.