Three essays on international trade

This doctoral thesis consists of three independent chapters. However, those three chapters discuss a similar big issue, i.e. international trade. In depth, they talk about an empirical case of trade in Southeast Asian area (Chapter 1), and Indonesian manufacturing industries (Chapter 2 and 3). Chapter 1 will put its intention in the bilateral trade relationship among ASEAN (Association of Southeast Asian Nation) countries itself and between ASEAN and their major trading partners, Japan and the US. Chapter 2 will do an investigation on the level of efficiency of Indonesian manufacturing industries and study the determinants of export in Indonesian manufacturing. Finally, Chapter 3 observes the decision of doing export in a specific Indonesian manufacturing industry, i.e. textile and apparel firms. In fact, we specifically put our intention on the theme of export. The discussion of it unifies those 3 chapters. However, each chapter has different emphasizes on its export idea. We stress our analysis on the effect of change in real exchange rate, real domestic income, and real domestic cash balance on the real bilateral trade balance in the first chapter, while in the second chapter; we observe the influence of capital intensity, size, export diversification and technical efficiency on the export performance, then for the last but not the least, third chapter will study the decision to export of Indonesian textile apparel firms. Furthermore, efficiency also obtains our great concern, as we believe that efficiency is one of the important aspects that can positively affect country’s export performance (which is stated by self-selection hypothesis), thus country’s economy as well. When a firm or industry could achieve a higher level of efficiency, it should be able to enjoy a better export performance or a higher probability of becoming an exporter. More exports could be nearly related to more profit which can be collected by a firm, since export can be used by firm as a way for generating profit. Assuming firms will not do exports if they can not get any profit, then through the accumulation of the exporter’s profits, the country’s trade balance and economy will also positively flourish. Efficiency is measured by technical efficiency in the second chapter and by total factor productivity (and labor productivity) in the third chapter. To explain shortly the discussion of the thesis, a concise description about each chapter can be shown as follows. The first chapter examines the short run and long run effects of real exchange rate, real domestic (foreign) income, and real domestic (foreign) cash balance on the real bilateral trade balance in ASEAN region. Based on quarterly data set 1980q1 to 2007q3, investigations are carried out using VECM method. However, the impulse response functions, the variance decompositions, as well as the OLS are analyzed in order to capture further the dynamic perspective of the research. Expanding our analysis, we do test and analysis for a possibility of the presence of structural break. The Chow test is applied when a structural break with a known break point is considered. While, Zivot and Andrews (1992) unit root test, Gregory and Hansen (1996) cointegration test, and two steps Engle-Granger method are performed when a structural break with an unknown break point is measured. Results show that in the long run: (i) income effect is found to be dominant in determining the change in trade balance, either when a structural break is omitted or allowed; (ii) the cash balance effect do influence the bilateral trade; (iii) the exchange rate effect significantly plays a role only in the period before the regime shift where it is absent in the post regime shift, indicating that the structural break carries a significant impact in removing the positive long run effect of exchange rate on trade balance. With respect to that, the small economy effects are suspected to be present in ASEAN region. Meanwhile, in the short run: (i) the cash balance effect relatively plays a major role in influencing the improvement of trade balance, either when a structural break is omitted or allowed, (ii) compared to cash balance effect, the income effect is present with slightly less difference in contribution; (iii) the exchange rate effect is always observed in all types of analysis and is more discernible when a structural break is omitted, while a J-curve phenomenon is present in minor cases. On the contrary, once a structural break is allowed, we provide lack of evidence for the presence of the phenomenon. The second chapter examines the technical efficiency of Indonesian manufacturing industry by estimating a stochastic production frontier (SPF) and the constant returns to scale (CRS) output-oriented DEA approach. In addition, this chapter also analyzes the determinants of export performance of the industries using panel analytical model. The results show that the estimated mean technical efficiency of Indonesian manufacturing industries which are found by those approaches are lower than 0.5. It indicates that there is substantial inefficiency problem in the industry. Comparing the scores obtained by SPF and DEA, we identify there are six relatively prominent industries in term of efficiency achievement, namely iron & steel, tobacco, transport equipment, food products, industrial chemicals, and machinery, electric. Our results which correspond to the technical efficiency indicate that the estimated mean technical efficiency in the DEA approach is larger than those obtained from the SPF. Utilizing Fixed Effect Model (FEM), we find that all export determinants, i.e. capital intensity (CAP), number of labor (SIZE), diversification (DIV) and technical efficiency (TE) shows the appropriate significant sign. The highest elasticity coefficient is provided by diversification (DIV) variable. With respect to the relationship between efficiency and export performance, we provide evidence for the presence of self-selection hypothesis. Then, the third chapter is devoted to the examination of factors which influence the decision to export of firms in Indonesian textile and apparel industry. Using a panel of firm-level data and a panel probit model, we test for the role of heterogeneous characteristics of firms in determining firms’ probability of exporting. We use two different definitions of firm productivity in our model, i.e. TFP and labour productivity. Besides executing the general estimation for whole observations, we execute as well the disaggregated specifications concerning the firm size (middle and large size). Mostly, the main findings in this study are in line with related previous works. In particular, we find that productivity increases the probability of exporting. Likewise, firm characteristics related to size and foreign ownership has a positive influence on the probability of exporting. On the other hand, variables accounting for capital intensity and Java region dummy affect negatively the decision to export. However, in the general estimation results, we find no significant effect of labor quality on our model. Concerning the estimation results of disaggregated specifications, a consistent finding with the general estimation results is basically provided, except for labor quality variable which is significantly negatively related with exporting in middle-size firms’ case. This condition seemingly reflects the actual state of the textile and apparel industry in the developing economy, such as Indonesia. In which, exporters are generally inclined to be more labor intensive, and ‘cheap’ labors are specifically dominant to be employed among middle firms. Besides, the reason of less competitive in the international market also should be taken into account. Although still showing positive signs, the coefficients of productivity lose their statistical significance when the models are applied for large-sized firms. However, in general, our findings corroborate the self-selection hypothesis.