Luận văn Models to analyze the impact of foreign direct investment on economic restructuring in Vietnam

is characterized by higher wages, advanced techniques production and positive marginal productivity of labor. 1.2.2. A model of stages in economic growth by Walt Rostow Summary of 5 stages of development of Walt Rostow over time is shown in figure 1.3. Figure 1.3 Diagram of Rostow's 5 stages of development 1.3. Theory of impacts of foreign direct investment on economic restructuring 1.3.1. Theory of the Investment Development Path The theory of Investment Development Path (IDP) hypothesizes that there is a relationship between a country's net of foreign direct investment and its level of economic development. The premise of this theory has two sides: it sees economic development as a process of structural change and this structural change affects the morphology of both foreign direct investment in and outbound, and thus, affecting the position of the net of foreign direct investment. In the stylized form of IDP theory, countries develop through the five stages. This evolution took place in response to the changing conditions of specific advantages of the country’s location, and thus, on the one hand, its appeal to foreign investors, on the other hand, it gradually upgrades the characteristic advantages of ownership of domestic enterprises. According to IDP theory, when moving along the stages of Time D ev el o pm en t le ve l Traditional social period (livelihoods, bartering, agriculture) Period of Transition (specialization, surplus, infrastructure) Period of taking off (Industrialization, increased investment, regional growth, political change) Mature period (diversification, innovation, less dependence on imports and investment) Period of high level of consumption (consumer oriented, prosperous durable goods, dominant service sector) 5 development, the economic structure of both the investment country and the FDI recipient country has shifted. In the process of outward investment as well as accepting investment, the comparative advantage of these countries will shift to respond to and to create new advantages to attract FDI. In the opposite direction, the participation of the FDI inflow and outflow also adjust the economic structure of the accepting investment country and country implementing investment. 1.3.2. The "flying geese" model The "flying geese" model showed that differences in the importance of production factors in the stages of development of products gave rise to the law of advantage shift. In order to restructure the economy, innovate the technical structure, modernize the economy, focus on research and development of key industries to dominate and control the world economy, the leading countries always have a need to shift “outdated industries”, outdated techniques, and “aged products” to the following countries in order to focus capital on developing new industries and technologies, at the same time, extending the "lifespan" of our technologies and products, thereby generating high profits. Similarly, the following countries also have a need to shift investment in technology and products for which they have lost their comparative advantage to less developed countries. This process of transferring technology takes place in many different forms, which may be through the form of FDI. And it has changed the face of economies at a lower level of technology, accelerating the process of regional assignment and cooperation and regional linkage. In addition, it also created a shift in the economic structure of the countries participating in the "flying geese", in particular a shift towards industrialization and modernization of the economy. 1.3.3. Model of upgrading the sector An improved version of the "flying geese" model was the industrial upgrading model, in which it described the structural change of Japan since 1950 and clearly demonstrated the role of FDI as a moderator of structural change. According to this model, a country's industrial structure undergoes four successive transitions. In each stage, there is a specific group of industries that dominate the industrial structure by possessing comparative advantages. Ozawa also found that there was a specific correspondence according to the stages between structural upgrading and forms of FDI that lead to a model of similar phases in FDI activity. 6 1.4. Overview of empirical studies on the effects of FDI on economic restructuring Studies in the world on the effects of FDI on economic restructuring have been carried out for groups of countries (Kalotay, 2010; Gui-Diby and Renard, 2015; Ssozi and Bbaale, 2019, ...) as well as individual countries (Blomström et al., 2000 for Japan; Zorska, 2005 for Poland; Liang and Bing, 2010 for Korea; Huang Na, 2011 and Jiang, 2014 for China,). These studies have been based on various research methods from qualitative research to quantitative research with the use of the models of time series regression and the panel regression model. The results obtained largely support the assumption that FDI positively supported the economic restructuring of the host country in the direction of increasing the proportion of the industrial and service sectors, and gradually reducing the proportion of the agricultural sector, in both output and labor. Besides, a number of other studies have also examined the impact of FDI on the restructuring of the industry and the structure of international trade of the host country. Research on the effects of FDI on economic restructuring in Vietnam is still limited. In terms of methodology, the studies available were classified into two groups. The first group mainly focused on theoretical basis as well as on descriptive statistics to make judgments about the effects of FDI on economic restructuring. The second group provided results based on quantitative models. The conclusions of these studies all had a unified point that confirmed the positive effects of FDI on the process of the economic restructuring. Most of the aforementioned studies all used the proportion of industries representing economic restructuring, the proportion of which only shows the economic structure at a static time and the sectors were independent of each other. This seems unreasonable when the economic structure is often shown as a dynamic object and needs to be considered the economic restructuring in the general context of all sectors in the economy. Therefore, it is necessary to continue to clarify this relationship under different approaches to get more accurate conclusions for the dynamic relationship between FDI and the economic restructuring. 7 Chapter 2 METHOD OF RESEARCH 2.1. Measure the economic restructuring - Proportion of the sectors - Absolute value index (NAV) = 0,5. | |, = 2, , ; = 1, , 1 - Modified Lilien index = . . , > 0; > 0 - Measure the speed of restructuring based on vector method cosθ = ∑ aa ∑ a . ∑ a - Some other approaches + Use a component in the disintergration of overall productivity, considering it represents the change in labor structure = = + + The index "changing structure efficiently " of Vu (2017) = 0.5 ∑ | |∈ , = {} sao cho > 0 in which, iTS and 0iS are the labor proportion of industry i at time T and 0; Ci is identified by = ̅ + 2.2. Measure the imbalance of the economic structure Let and respectively the added value and labor of industry i Ando and Nassar (2017) defined the following values: ∑ ∑ , 8 Thus, each reflects the difference between the proportion of labor and the proportion of industrial productivity of the industry i. The distance d = 0 equates to equal productivity across sectors. 2.3. Several panel models 2.3.1. Static panel data models = + + The thesis in turn examines the model with estimates including POLS, FE, RE and FGLS to choose a suitable model for static panel data model. 2.3.2. Dynamic panel data models = , + + + The dynamic of restructuring is considered under dynamic panel data model. The appearance of the lag variable of the dependent variable as an independent variable giving rise to endogenous problem. The thesis uses the method of estimating GMM to overcome this problem including IV-GMM, D-GMM and S- GMM. 2.4. Spatial econometrics models 2.4.1. Static spatial panel models = ∑ + + + () = + + , = ∑ + () = ∑ + + ∑ + + () 2.4.2. Dynamic spatial panel models Dynamic spatial panel models under consideration include the SAR and SDM model. = , + + , + + + () = , + + , + + + + () 9 Chapter 3 THE SITUATION OF FDI ATTRACTION AND THE PROCESS ECONOMIC RESTRUCTURING IN VIETNAM 3.1. The situation of FDI attraction in Vietnam Since opening to attract FDI, Vietnam has been very successful in attracting this capital to support domestic capital to promote economic development. The total number of projects which are licensed by the end of 2018 is 27,454 projects with a total registered capital of 340.85 billion USD, in which they mainly focus on manufacturing and processing industries (accounting for 57.7%). ), followed by real estate (17%), electricity, gas, hot water, steam and air conditioning are in third place with around 6.8%. The remaining industries account for a relatively low proportion, ranging from 0.2% to 3.5%. 3.2. The process of economic restructuring in Vietnam During the research period, the economic structure has always shifted towards modernity in both labor and GDP. The proportion of GDP in the agricultural sector decreased from the highest level of 46.3% in 1988 to 14.7% in 2018. The remaining two sectors had a gradual increase in GDP, but the growth rate was still quite slow. The industrial sector only increased by 11.5 percentage points from the lowest point (22.7% in 1990) to the end of the period (34.2% in 2018). The service sector also had an equivalent increase from the lowest level of 29.7% in 1988 to 41.1% in 2018, which means an increase of 11.4 percentage points. For the labor structure, the transformation took place more strongly. The agricultural sector, from a very high proportion in 1986 (73.9%), has sharply decreased to 37.7% in 2018. The proportion of industry and services increased from 13.9% and 12,2 % respectively in 1986 to 26.7% and 35.6% in 2018. Figure 2.4 . Proportion of GDP in the sectors 10 Figure 2.7. Labor Structure in the sectors Despite a positive shift, our country's economic structure is still backward compared to some regional countries. In addition to the economic restructuring of the industry according to the proportion of GDP and the proportion of labor, the economic restructuring in the international trade structure, the economic structure has also taken place in a positive direction. The economic restructuring of exports takes place in the direction of developing value-added products with high technology and gray matter content, gradually creating branded products to meet the requirements of the world market. In the structure of economic components, the shifting trend is to reduce the proportion in the state economic sectors, the collective economy, the individual economy and increase the proportion of the private sector, the sectors with foreign capital. Link the process of the economic restructuring to economic development, during the period of consideration, increasing GDP per capita is related to the reduction of the proportion of GDP and the proportion of labor in agriculture. Along with that is the increase of the proportion of GDP and the proportion of labor in both industrial and service sectors. For the agricultural sector, at a low level of development, the proportion of added value is significantly lower than the proportion of labor. For the service sector, both the proportion of GDP and the proportion of labor are blocked by a value that is not zero even at the lowest level of development. The increase in GDP is also lower than that of labor. For the industrial sector, the movement of labor to the industrial sector seems to be slower than the service sector as average income increases, but the level of increase in GDP is much faster than services. It is especially observed that the proportion of labor as well as GDP in the industrial sector appears to be dispersed as per capita income increases. This figure shows that the increase in per capita income increases the disparities in the proportion of industry among provinces. 11 Chapter 4 MODELS TO ANALYZE IMPACT OF FDI ON ECONOMIC RESTRUCTURING IN VIETNAM 4.1. Models and data The static model: = + + + (4.1) The dynamic model: = , + + + + (4.2) Other explanatory variables are summarized in Table 4.1 and the descriptive statistics of the variables in the model are presented in Table 4.2. Data used in the models is data of 60 provinces and cities nationwide for the period 2000-2017 and provided from the General Statistics Office of Viet Nam. 4.2. The impact of foreign direct investment on economic restructuring from non-spatial models The thesis in turn estimates the static and dynamic models with different methods to overcome existing problems. In the static model is the problem of the variance of random error varies, the model is estimated by FGLS. The dynamic model has the endogenous problem of the lagged dependent variable, estimation methods include IV-GMM, D-GMM and S-GMM. Through tests and analysis, choosing between models, the dynamic model estimated from S-GMM is the most suitable for analysis. Table 4.5 shows the results of estimation of dynamic models by S-GMM to support the analysis of economic implication. The main conclusions drawn from the results of estimation of dynamic models include: + FDI has the effect of promoting the process of restructuring to take place faster, but with the condition of human capital being met. This positive effect occurs in both short run and long run. The period from 2004 onwards it shows more clearly. However, lag 1 of this variable has a negative impact on economic restructuring possibly stemming from the overwhelming effects of FDI. This negative impact is only present in the short run and has no impact in the long run. 12 13 + Trade openness: Despite having a positive coefficient, it is not statistically significant, while population size has the effect on reducing the level of economic restructuring. + Human capital, changes in income, and private capital are factors that contribute significantly to accelerating the process of restructuring, in which human capital and income have a strong influence. Urbanization has left consequences of inhibiting clearly economic restructuring. The above effects are present in both short run and long run. + The variables belonging to the group of deficit with input factors are not reaching the necessary statistical significance. Among them, factors related to the industrial sector all have positive coefficients, showing the positive influence of this sector on the economic restructuring. 4.3. The impact of foreign direct investment on the imbalance of the structure of the economic 4.3.1. Evaluate the level of imbalance of the structure Using the index to assess the structural imbalance (d) according to Ando & Nassar (2017) shows that the agricultural sector always has a surplus of labor, industry and services are the two sectors that lack labor, in which the industrial sector demonstrates greater severity. For the whole economy, the imbalance in existing tends to decrease over time. 4.3.2. The model analyzes the impact of factors on the level of structural imbalance The estimation results of selected models for analyzing the impact of factors on the general structural imbalance as well as each sector are presented in Table 4.8. Some of the main conclusions drawn from the estimation results include: + Despite being recognized as having positive contributions to the process of economic restructuring, but with the main investment in some industries, FDI creates a significant imbalance in the economic structure. The effects on this imbalance of FDI are maintained in both short run and long run. If we consider only each sector, FDI reduces the imbalance in the industrial sector, with the other two sectors, the coefficient of FDI is positive but only shows a clear impact on the agricultural sector, this effect is still quite weak in the service sector. The positive sign of FDI partly reflects the imbalance in the structure of agriculture and services caused by FDI. This result further clarifies the above comments. 14 15 + Among the explanatory variables included in the model, only private capital contributes to reduce the general imbalance, but it has not been really strong. The remaining explanatory variables including human capital (vcn), change in income (tdtn), population size (ds), the level of urbanization (dth) and trade openness (xnk) all have positive impact on the index of general imbalance, in which vcn, dth and xnk are statistically significant effects. 4.4. Spatial spillover in economic restructuring 4.4.1. Test for spatial correlation Spatial correlation was examined through Moran's test with the I-Moran index. The results of testing the spatial correlation in economic restructuring (table 4.11) show that there an existence in a spatial correlation among localities except in 2000. Table 4.11. Result of testing Moran's I Năm Moran’s I Năm Moran’s I I z p-value I z p-value 2000 0,014 1,211 0,113 2009 0,079 3,720 0,000 2001 0,038 2,146 0,016 2010 0,039 2,181 0,015 2002 0,123 5,473 0,000 2011 0,037 2,105 0,018 2003 0,134 5,903 0,000 2012 0,038 2,112 0,017 2004 0,176 7,470 0,000 2013 0,036 2,042 0,021 2005 0,145 6,306 0,000 2014 0,041 2,250 0,012 2006 0,109 4,927 0,000 2015 0,040 2,191 0,014 2007 0,116 5,152 0,000 2016 0,030 1,803 0,036 2008 0,097 4,408 0,000 2017 0,041 2,252 0,012 4.4.2. Economic restructuring from spatial econometrics models To assess the spatial spillover effects of the factors on the economic restructuring, the thesis in turn estimates static and dynamic spatial panel models including SAR, SEM and SDM. Based on analysis, testing and model selection, the dynamic SDM is the most suitable model for this impact assessment. Estimation results of the dynamic SDM model are presented in Table 4.13. And table 4.14 shows the results of decomposition according to the direct and indirect impacts from the dynamic SDM model. Some of the main results drawn from estimation results of spatial models are as follows: 16 Table 4.13. Summary of estimation results of the dynamic spatial models (1) fdiDSDM (2) fdiDSDMtt (3) xnkDSDM Main L.cdcc 0.992*** 0.997*** 0.984*** L.Wcdcc -0.324*** -0.349*** -0.366*** fdi 0.0155** fdivcn 0.00382* vcn 0.172 0.137 ds 0.784 0.818 0.370 tdtn 0.0648*** 0.0660*** 0.0646*** vtn 0.0192*** 0.0198*** 0.0159** dth -0.0319** -0.0325** -0.0296** xnk 0.00157** tdldcn 1.042* 0.852 tdvcn 0.193 0.173 tdlddv -0.211 -0.174 tdvdv -0.235 tdldnn -0.351 tdvnn -0.932*** Wx fdi -0.0398* fdivcn -0.0202*** vcn 0.147 0.146 ds 11.33 11.56* 12.79** tdtn -0.101*** -0.0972*** -0.0931*** vtn 0.0527 0.0460 0.0633* dth -0.294* -0.328* -0.262** xnk -0.0035 Spatial rho 0.324*** 0.357*** 0.342*** Variance sigma2_e 2.427*** 2.442*** 2.408*** Log Likelihood -1872.7 -1876.4 -1869.1 AIC 3855.5 3850.9 3848.1 BIC 4126.5 4092.3 4119.1 17 Table 4.14 The results of decomposition of direct and indirect impacts (1) fdiDSDM (2) fdiDSDMtt (3) xnkDSDM SR_Direct fdi 0.0157** fdivcn 0.00365* ds 1.026 1.115 0.617 vcn 0.177 0.142 tdtn 0.0637*** 0.0639*** 0.0637*** vtn 0.0199*** 0.0209*** 0.0172** dth -0.0369** -0.0390** -0.0348** xnk 0.00159** SR_Indirect fdi -0.0510 fdivcn -0.0302** ds 16.34 18.92* 19.11** vcn 0.325 0.313 tdtn -0.115** -0.115** -0.111** vtn 0.0841* 0.0803 0.104** dth -0.425* -0.547** -0.396** xnk -0.00410 SR_Total fdi -0.0353 fdivcn -0.0266** ds 17.37* 20.03* 19.73** vcn 0.502 0.455 tdtn -0.0511 -0.0508 -0.0475 vtn 0.104* 0.101* 0.121** dth -0.461* -0.586** -0.430** xnk -0.00250 18 + Economic restructuring in a province taking place faster will also be a factor promoting the process of economic restructuring in neighboring localities at the same time. However, in the latter period, these spatial influences change in the opposite direction. + FDI has opposite effects in direct and indirect effects. FDI is a resource that accelerates the process of economic restructuring in the receiving investment locality, which is consistent with the results analyzed from the dynamic non- spatial panel model. However, FDI restrains the level of economic restructuring in neighboring localities. The estimated coefficient of FDI is 0.0155, while the direct effect is 0.0157, so the feedback effect is only 0.0002. + Some other factors that also have direct and indirect effects are the changes in income (tdtn), private capital (vtn) and the level of urbanization (dth). Like FDI, changes in income have a strong impact on economic restructuring in the short term in the local area, but restraining neighboring localities. The changes in income has only a positive effect in each locality, not a positive spatial spillover effect to neighboring localities. The consequences of over-urbanization not only negatively affect the local area but also affect the neighboring localities. Import and export only show a direct positive impact on economic restructuring on the local area, but not a spillover effect to neighboring localities. Likewise, the local's human capital does not show indirect effects on the economic restructuring of the neighboring locality. The most special of the above factors is population size (ds), which has only a positive spatial spillover effect on the restructuring of neighboring localities, but almost no direct impact on dependent variable on the locality + The spatial spillover effects analyzed above are short-term impacts. The indirect impacts coming from neighboring localities on the level of local restructuring will not last long in the long-term. 19 Chapter 5 CONCLUSION AND RECOMMENDATION 5.1. Conclusion The thesis has provided evidence for the impact of FDI on the process of economic restructuring in Vietnam. The obtained conclusions include: Firstly, the allocation of FDI is consistent with the transformation of the economy as well as within the industrial sector. For the whole economy, the overwhelming participation of FDI in industry has led to the shift in the proportion of GDP as well as in labor towards increasing the proportion of the non- agricultural sector and reducing the proportion of the agricultural sector. In the industrial sector, FDI is mainly in the manufacturing and processing industry where many labor-intensive manufacturing industries are concentrated, which makes the proportion of production value of the manufacturing and processing sector accounting for a large proportion but still tends to increase. Secondly, the presence of FDI drives the process of restructuring to take place faster, but it was only really evident in the period 2004 and after that FDI has been allocated to other sectors than industry. This effect of FDI appears both in the short term as well as is accumulated in the long term, but the level of influence is still quite limited compared to some other factors. However, in the short term, FDI still has a negative impact on economic restructuring when considering the one-period lag of this variable. This negative effect can be caused by the temporarily overwhelming effect and not in the long term. Thirdly, because of the overwhelming participation of FDI in the industrial sector, it has created a structural imbalance for th