Original Paper Research on the Trade Effects of China’s Direct Investment in ASEAN

Based on the data of China’s direct investment and trade in ASEAN countries from 2004 to 2018, this paper established a variable parameter model, variable intercept model and constant coefficient model with panel data, and investigated the heterogeneous impact of China’s direct investment in ASEAN countries on their import and export trade and the impact mechanism. It is found that China’s direct investment in ASEAN will be deposited in the trade between China and 10 ASEAN countries, creating effects for trade, and the size of the effect varies with the host country. On average, when other factors remain unchanged, every 1 percentage point increase in China’s direct investment flows to ASEAN countries will increase exports to ASEAN countries by 0.54%, imports will increase by 0.44%. Further studies have found that our country to Brunei, Indonesia, the Philippines, Singapore and Thailand’s export create effect is greater than the imported create effect, on the whole, China’s investment in the five countries promoted the net exports, mainly on its investment in China, led to many mechanical equipment and other related products exports, and imports mainly import countries have their comparative advantages of products, variety is less.


Introduction
In 2018, world economic growth was basically the same as that of the previous year, and the growth of trade in goods slowed down. Global FDI outflows shrank, falling for three consecutive years from 2016 to 2018. The Chinese economy has made steady progress, the level of opening-up has been constantly www.scholink.org/ojs/index.php/ibes    China's OFDI flow to ASEAN, exports to ASEAN will increase by 0.54% and imports by 0.44%. In general, China's investment in the five countries promotes the net export, mainly because China's investment in them drives the export of many related products such as machinery and equipment, while the import is mainly due to the small number of products with the comparative advantages of their countries.

Literature Review
Direct investment flowing into the host country from the home country will have substitution or complementation effect on the trade between the home country and the host country according to its inflow mode and industry, which can be divided into export creation effect, import creation effect, export substitution effect and import substitution effect.
In theoretical studies, Mundell (1957) first studied the relationship between direct investment and trade, believing that direct investment has substitution effect on import and export trade. Later, Xiao dao-qing (1987) put forward the marginal industry expansion theory, believing that there was a complementary relationship between direct investment and trade. Markusen and Svensson (1985) put forward the contingency relationship between direct investment and trade, believing that the relationship between factor flows and commodity trade is related to the cooperation between trade and non-trade factors, while cooperation is complementary and non-cooperation is substitution.
In the empirical study, Niu Xi-fei (2012) and Hao Feng-feng (2014) discussed China's influence on asean's OFDI on import and export trade, and empirically found that OFDI created trade effects on import and export. Wang Chang-yi and Chen Li-xia (2013) analyzed that: in the long term, FDI and export are complementary, while import is substitution. Duan Zhen Yuan (2015)  However, when it comes to the regional economic cooperation between China and ASEAN, only a few studies have involved the relationship between OFDI and import and export trade. Therefore, this paper tries to systematically investigate the trade effects of China's direct investment to ASEAN countries through empirical analysis.

Analysis of OFDI's Export Effect on China
The export creation effect of OFDI is mainly shown as follows: foreign direct investment enables Chinese enterprises to increase the export of related equipment, raw materials and other products to ASEAN. For example, China's investment in ASEAN's communication industry will drive China's communication equipment export. Export substitution effect of OFDI: it means that OFDI reduces the export of Chinese enterprises to ASEAN, mainly because China's investment in ASEAN manufacturing industry leads to the decline of China's export of manufacturing products to ASEAN. When many manufacturing enterprises make cross-border investment, they mostly adopt a gradual approach and usually start trade with them first. With the development of trade, companies will set up subsidiaries in foreign countries, and then invest in factories to produce and sell locally, which will replace some of our exports to them.

Analysis of Import Effect of OFDI on China
In terms of import trade, the import creation effect of foreign OFDI is a kind of reverse import effect: mainly referring to the fact that China's direct investment to ASEAN leads to the product originally exported to ASEAN being imported from China. Such situation is easy to occur in the manufacturing and natural resource industries. For example, Chinese enterprises used to export washing machines to ASEAN countries, but now they set up factories in ASEAN countries, and the domestic output will decrease accordingly. At this time, Chinese enterprises will import the product from the subsidiaries of ASEAN countries. In contrast to the import creation effect, import substitution effect is caused by the decrease of imports from ASEAN due to OFDI, which often leads to the change of trade structure.

Mechanism Analysis and Summary
The industry of China's investment in ASEAN, the two countries' policies and economic environment will have an impact on the trade effect of OFDI, and different results will be generated according to different situations, which cannot be generalized. China's direct investment in ASEAN is unevenly distributed among countries and industries, so China's direct investment in ASEAN may have substitution or creation effect on trade, and it is difficult to judge its overall and individual influence from theory and experience. Therefore, this paper will establish two panel data models for practical verification, and determine the size and direction of China's import and export effect on ASEAN through quantitative analysis.

The Data Source
By searching the websites of China's Ministry of Commerce, ASEAN statistical yearbook, the world bank and the United Nations trade database, this paper found the stock data of China's direct investment in 10 ASEAN countries from 2004 to 2018, the import and export trade data of China and 10 ASEAN countries, and the GDP of China and 10 ASEAN countries.

Panel Data Model Settings
In this paper, two panel data models will be established to study the export trade effect and import trade effect of China's direct investment in ASEAN respectively. Import and export volume are taken as dependent variables, while the stock of OFDI is taken as independent variables. Although the degree of political stability, tax and other factors of ASEAN countries all affect import and export trade, due to the difficulty in obtaining relevant statistical data of the ten ASEAN countries and different statistical standards, FDI stock and GDP were selected as independent variables. In order to reduce the heteroscedasticity of the data, logarithms of all scalars were taken in this paper. The regression form of the model is as follows: LNEX= α 1 +β 1 LNFDI +γ 1 LNGDP ASEAN +δ 1 (Export model) LNIM =α 2 +β 2 LNFDI +γ 2 LNGDP China +δ 2 (Import model) a 1 , a 2 is a constant, δ 1 ，δ 2 is a random disturbance. LNGDP ASEAN and 2 LNGDP China are control variables, respectively representing the economic development level of ASEAN countries and China, and γ 1 and γ 2 are respectively the influence coefficients of GDP on exports and imports. β 1 is the LNFDI coefficient of export and β 2 is the INFDI elasticity coefficient of import. If the LNDFI coefficient of the above two models is greater than 0, it indicates that China's direct investment in ASEAN has a creative effect on import and export. If the coefficient of LNDFI in the above two models is less than 0, it indicates that China's direct investment in ASEAN has a substitution effect on import and export, and the larger the coefficient is, the more obvious the substitution effect is.
There are three scenarios in the panel data model. According to the import model and the export model, they are set as follows: The first: α i ≠α j ,β i ≠β j ,γ i ≠γ j (Variable parameter model) LNE X it = α 1it +β 1it LNFDI it +γ 1it LNGDP ASEANit +δ 1it (Formula 1) LNI M it = α 2it +β 2it LNFDI it +γ 2it LNGDP Chinait +δ 2it (Formula 2) The second: α i ≠α j ,β i =β j ,γ i =γ j (Variable intercept model) LNE X it = α 1it +β 1 LNFDI it +γ 1 LNGDP ASEANit +δ 1it (Formula 3) LNI M it = α 2it +β 2 LNFDI it +γ 2 LNGDP Chinait +δ 2it (Formula 4) The third: α i = α j ,β i =β j ,γ i = γ j (Invariant parametric model) LNE X it = α 1 +β 1 LNFDI it +γ 1 LNGDP ASEANit +δ 1it (Formula 5) LNI M it = α 2 +β 2 LNFDI it +γ 2 LNGDP Chinait +δ 2it (Formula 6) In the above equation, i represents the individual of the cross section and t represents observation periods of the t-band edge. The first is variable parameter model, which is not only affected by individual spatial factors, but also causes corresponding changes in its coefficients due to structural changes. Second, the variable intercept model is not affected by individual factors and structural changes in the cross section. Third, the invariant coefficient model is not affected by individual factors and structural changes in the cross section, and the sample data comes from spatial data of different time series.

Panel Data Unit Root Check
Before the empirical analysis, the stationarity of data should be tested first to prevent spurious regression. For stationarity test, we can use the ADF unit root test, the null hypothesis is unstable.
LNEX, LNIM, LNFDI, LNGDP ASEAN and LNGDP China were respectively tested, and the test results were shown in the following Figure. During the horizontal variable test of unit root, only the P value of LNIM was less than 0.05, while all other variables accepted the null hypothesis and there was unit root.
Then the first-order difference test was carried out. As can be seen from  of the random effect model is rejected. Then, according to the LR test results, all P statistics are 0, so the hypothesis that the mixed effect model is more effective than the fixed effect model is strongly rejected at the significance level of 1%. Therefore, the fixed effect model should be used for all models in this paper.

Regression Analysis of China's Export Effect on OFDI of 10 ASEAN Countries
The results show that the panel data in this paper should adopt the fixed effect model of variable parameter number. In the regression analysis of export and OFDI, R 2 was 0.976293, indicating that the model fitted well. As can be seen from Table 4, P statistics of 10 ASEAN countries are all less than 0.001, which indicates that explanatory variables have a significant impact. in different ASEAN countries has different impact on China's export. The LNFDI coefficients of Singapore and Vietnam were as high as 0.8, while the elasticity coefficients of Indonesia and the Philippines were relatively small for the first few countries. Third, China's direct investment in ASEAN countries with rich natural resources has a great export creation effect, which is basically because Chinese enterprises need relevant technical equipment and parts and components when developing their resources, so it will drive Chinese products to export to the host country. Fourthly, the LNFDI coefficient of Singapore is close to 0.9, that is to say, the coefficient of this kind of Singapore is close to 0.9, that is to say, every additional unit of investment in Singapore by China will enable China to export 0.892496 more units of products to Singapore. This is mainly because most of the local people are Chinese and there is no great cultural difference, which is conducive to the trade between the two countries. China for its main industry (manufacturing, mining, leasing and business services, power/heat/gas and the production of water supply industry, finance, construction) have a browse, nearly $30 billion investment, the corresponding brings China a lot of related machinery and equipment, trade in services, power/heat/gas and water production technology exports. For Thailand, which had the lowest coefficient, China invested $2,019 billion mainly in its manufacturing sector, with fewer other sectors, mainly to drive exports of related machinery and equipment.

Regression Analysis of Import Effect of China on OFDI of 10 ASEAN Countries
In the regression analysis of China's import effect on OFDI of 10 ASEAN countries, the goodness of fit R 2 is 0.967664, indicating that the explanatory variable LNOFDI in the model has a high degree of interpretation of the other explanatory variable LNIM. By observing the significance of the regression results in Table 5, it can be found that the P statistics of 10 ASEAN countries in the regression results are all less than 0.05, and even less than 0.001, indicating the significant influence of explanatory variables. China's relatively large investment in the tertiary industry of Singapore. For example, wholesale, retail, leasing and business services are more conducive to China's export of ordinary products to Singapore.
China's imports from Singapore are mainly high-tech products, and China's investment in Singapore's high-tech industry is less than 5%, so the import drive is relatively small.
According to Table 4 and Table 5, we can see that China's export creation effect on brunei, Indonesia, the Philippines, Singapore and Thailand is greater than that on imports. In general, China's investment in the five countries promotes net exports. The main reason lies in China's investment in it, leading to the export of a lot of machinery and equipment and related products, while the import is mainly to import the products that their countries have a comparative advantage, with a few types. For Vietnam, China's direct investment in Vietnam has obviously boosted China's import and export trade with it.

The Economic and Trade Relations between China and ASEAN Are Close on the Whole but Uneven in Distribution
Due to the establishment and development of China-ASEAN free trade area and the implementation of China's "One Belt And One Road" strategy, the economic exchanges between China and ASEAN have been continuously strengthened. In terms of investment, China's direct investment flow to ASEAN is on the whole on the rise, and the investment stock is growing steadily, gradually surpassing the EU.
However, the uneven distribution of China's direct investment to ASEAN countries is mainly due to the different levels of economic development and comparative advantages of different countries. In terms of investment, China and ASEAN to more than 50% of the investment stock is concentrated in Singapore and Indonesia, the two countries and most of the ASEAN countries of ASEAN in China accounted for less than 10% of the total stock of OFDI, and especially in the least amount of investment of the two countries, brunei and the Philippines to brunei and China investment in Singapore stock gap totaled more than 200 times, so the Chinese investment in ASEAN countries lack of balance. In terms of trade, the trade scale between China and the 10 ASEAN countries is on the rise, with a large gap between countries and trade is relatively concentrated. In the long run, China's import and export to Malaysia, Singapore, Indonesia and Thailand are very large, accounting for about 80% of the import and export trade between China and ASEAN. In the short term, the trade volume between China and Vietnam has grown rapidly, becoming China's largest trading partner in southeast Asia.

China's OFDI to ASEAN Generally Creates Effects for Trade
The empirical results show that China's direct investment in ASEAN has a creative effect on China's import and export trade. Moreover, the export creation effect for Brunei, Indonesia, the Philippines, Singapore and Thailand is greater than the import creation effect, and the direct investment for Vietnam is the biggest promotion for China's import and export trade.