Abstract: This research studies the effects of Free Trade Agreements on Intra-Industry Trade. The textile sector of Turkey which has an important role on Turkey`s external trade is based on while the effects of Free Trade Agreement on Intra-Industry Trade are studied. The textile sector has been examined in three subgroups as classified by SITC within the 1996-2003 time internal. These subgroups are SITC-26 which is made up of textile fibers (other than wool tops) and their wastes, SITC-65 made up of textile yarn, fabrics, made up articles, n.e.s related products and SITC-84 made up of apparel and clothing accessories. As a consequence of all studies done, it can be said that agreements have a positive effect on Intra-Industry Trade based on textile sector. And it also can be said that this effect is more intense on SITC-65 products.
INTRODUCTION
For developing countries, the conditions under which development is achieved bears importance. According to the competitive advantage principle, it has suggested that developing countries specialize in certain industrial goods, produce them at low cost and that these goods be traded at IIT (Intra-Industry Trade) level (Balassa, 1966). The major effects of IIT are issues such as product differentiation, economies of scale, border trade and re-export.
IIT is defined as the simultaneous import and export of the same product group between two countries. For that reason, it makes up a large portion of world trade for developed market economies (Culem and Lundberg, 1986). The concept of simultaneous import and export has first been mentioned by Ohlin (1933) and Hilgerdt (1935) before the second world war. IIT had been largely ignored until the 1960’s, when it was empirically studied by Dreze, Verdoorn and Balassa (Dreze, 1961; Verdoorn, 1960; Balassa, 1965).
The theoretical and methodology studies that the empirical studies about IIT have been based on were started by Grubel and Lloyd (1975).
After Grubel and Lloyd (1975), IIT has began to be analyzed in context of many countries and industries. Besides the increasing importance of IIT in world trade, there are only a few research papers concerning Turkey (Schüller, 1995; Çepni and Köse, 2000; Erk and Tekgül, 2001) calculated IIT levels for Turkey’s total trade. Gönel (2001) measured the IIT levels for Turkey’s traditional export industry, textiles. Whilst these papers measured IIT levels, none of them attempted to test the determinants of IIT for Turkey. So this study aims to fulfill this shortage in the literature. The aim of this study is to analyze the IIT of Turkey and to test empirically the determinants of IIT among countries and across industries. Emirhan (2002) measured the IIT levels for various industry and country groups for Turkey and to test the determinant of these calculated levels empirically.
In this study measured the IIT levels for textile sector and the country with which Turkey has a Free Trade Agreement (FTA).
MATERIALS AND METHODS
In this study, the IIT levels are calculated by using the Unadjusted Grubel and Lloyd (1975) Index which is the most appropriate method for measuring IIT. The Grubel-Lloyd Index (GL index) measures the share of IIT in an industry:
|
GLIIT = 1 - [|Xi - Mi|/(Xi
+ Mi)] |
Where, Xi and Mi are exports and imports of industry I, respectively. This index GL takes values between 0 and 1, where 1 represents all trade being IIT. This index was adjusted for trade imbalances by Grubel-Lloyd and some other researchers but Vona (1991) suggested that index does not need to be adjusted.
In the literature, it has been accepted that the determinants of vertical IIT and horizontal IIT are different. Vertical IIT is a term that is used to describe the IIT of commodities that are at different qualities; whereas horizontal IIT is used to define IIT of commodities that are at the same qualities. In order to test whether the above proposition is valid in Turkey or not, the determinants of these two types of IIT are tested in different regression equations. For disentangling IIT into vertical and horizontal components, quality differences within exports and imports of an industry are used. Quality differences in trade are measured by export and import unit values per ton. Stiglitz (1987) argued that prices of commodities are reflections of their qualities so, prices can be used as an indicator of quality.
An industry is subject to horizontal IIT when the unit value of exports relative to unit value of imports lies within a specified range. This condition can be shown with the equation
| 1-α ≤ (Uvxij/UVmij)
≤ 1+α |
where, UVxij is country j’s unit value of exports in commodity I and UVmij is country j’s unit value of imports in commodity I. When relative unit values lie outside this range, the industry is facing vertical IIT. The condition for vertical IIT is shown as follows:
(Uvxij/UVmij)
< 1-α or (Uvxij/UVmij)
> 1+α |
where, α = 0,15. The transportation and freight costs are usually assumed to account for ±15% of the value of the product, so α is taken as 0,15.
Horizontal or vertical differences have not been taken into consideration while the level of IIT is calculated in this study.
The study textile sector has been examined in three subgroups as classified by SITC within the 1996-2003 time internal. These subgroups are: SITC-26 which is made up of textile fibers (other than wooltops) and their wastes; SITC-65 made up of textile yarn, fabrics, made up articles, n.e.s and related products; and SITC-84 made up of apparel and clothing accessories. Data received Under secretaries of the Prime Ministry for Foreign Trade of Turkey (Anonymous, 2004).
The GL indices have been calculated for these groups with the aim of displaying the developments concerning IIT within the textile sector, in which Turkey has competitive advantage.
EMPIRICAL EVALUATION AND CONCLUSIONS
In the first phase of the study, the following table has been formed of GL indices between 1996 and 2003 of several sectors to demonstrate the position of the textile sector.
In the Table 1 the GL index for Turkey’s agricultural trade has been calculated to be 1,00 for 2003. The presence of high values for other years shows that full specialization exists in agricultural IIT. It is worth noting that IIT has risen in industrial goods in recent years. High GL values for icon and steel and other raw material means that IIT is a common in industry sub-sectors.
In textiles, the IIT values are high in SITC-65 subgroup, medium in SITC-26 subgroup and very low in SITC-84 subgroup.
The poor performance of SITC-26 subgroup and especially SITC-84 subgroup with relevance to IIT demonstrates that trade is being carried out in one direction (import or export) in these two groups. When examined, it is seen that import figures are five times as large as export figures in SITC-26 subgroup and export figures are thirty-five times as large as import figures in SITC-84 subgroup. As a result, it can be concluded that Turkey has shown poor performance in IIT apparel although it is among countries with biggest competitive advantage in this sector.
In this study, the annual average percentage rate of change in the textile trade of Turkey between the year 2003 and the years when the agreements were started to be applied by the countries with whom Turkey had signed FTA and all the data is given in Table 2.
In this Table 2, the FTA effective data and average changes in the foreign values are presented. FTAs has affected Turkey’s textile trade in different ways and different rates for each country. In Table 1, it is especially noted that the annual average rate of change in textile trade between Latvia and Bulgaria is significantly high. Of the two countries, export-based trade was done with Latvia whereas import-based trade was done with Bulgaria in the years studied. On the other hand, when the results in Table 1 are generally studied, it can be said that free trade agreements were affected Turkey’s textile import much more and more positively.
In the last part of the study, the IIT concept which expresses the event of Turkey’s both importing and exporting synchronically in textile sector.
| Table 1: | GL index for main sectors |
| Table 2: | Rates of change in foreign trade after FTA (SITC-26, SITC-65, SITC-84 products subgroup) |
| Source: Anonymous, 2004 | |
| Table 3: | GL index for textile sectors with countries party to an FTA |
| * It could not find for Foreign Trade value of Iceland and Norway. ** It could not find for Foreign Trade value of Textile fiber and their wastes of Latvia and Turkey | |
It is studied one by one for each country party to an FTA and for each sub-product group and the results are given in Table 3.
The presence of IIT with any of the countries in Table 3 depends on the condition GLIIT>0.5. In the case of GLIIT<0.50 or GLIIT = 0, the presence of IIT can not be mentioned. The absent values on the table denote that foreign trade was not carried out for that year in the given product category. It can be seen that the effect of FTA’s on IIT varies from one country to another, one year to another and one product category to another. The IIT values with Romania have improved since 1998 (the year the FTA come into force) in SITC-26 subgroup, while for Hungary, Poland and Czech Republic these values have only risen in the year of the FTA coming into force and subsequently have fallen. The reason for this drop is the skew towards one way trade. The FTA has increased exports to Poland and decreased exports to Hungary and Czech Republic in SITC-26 subgroup. It is noted that the effects of the FTA with Norway have been observed after 2001 and that they have been bi-directional. With Switzerland it is seen that IIT exists in SITC-26 subgroup and best performance has been 2001.
In SITC-65 subgroup, IIT exists with Switzerland, Bulgaria and Slovakia regardless of the FTA’s. The impact of the FTA’s with the Czech Republic and Latvia can be observed. In SITC-84 subgroup IIT has been done for several years with Iceland and Latvia and in 2003 only with Bulgaria. Among countries Turkey has signed on FTA with exports have grown to Estonia, Iceland, Latvia, Norway and Slovenia and imports have grown faster from the Czech Republic, Israel, Switzerland, Lithuania, Hungary, Poland and Romania. Although exports to Bulgaria have been strong, the sharp rise in imports from this country in 2003 has caused IIT to improve.
Naturally, it would be fallacious to claim that Turkey’s IIT performance in SITC-84 subgroup has been impressive for the years studied.
When all the results are considered generally, it can be said that Turkey’s free trade agreements have positive effects on Intra-Industry Trade.
Turkey shows a very high performance in textile export. In this study, it is noted that the agreements have positive effects especially in textile import of Turkey. So it enables the Intra-Industry Trade which is done synchronically and containing the phenomenon of both import and export to rise. By the rise of IIT, while taking advantage of different markets in textile sector, Turkey will give opportunity of a wide variety of products to consumers, also can produce all the products more effectively, at low cost and in big quantities by producing less variety of products.