INVESTIGATING THE DETERMINANTS OF MACROECONOMIC FACTORS ON OUTWARD FDI OF VIETNAM: THE GRAVITY MODEL APPROACH

Objective: The main purpose of this paper is to investigate the motivations of Vietnam’s outward FDI. Theoretical framework: We extend the gravity model proposed by by Tinbergen (1962) and developed by Ryan W. Tang et. al (2022); Correa da Cunha et. al (2022), Hui-Ching Hsieh et. al (2019) to evaluate the influence of macroeconomic factors of Vietnam and the host countries on the volume of Vietnam’s OFDI flow. Method: This paper intends to analyze the motivations of outward FDI pattern of Vietnam using the gravity theory and panel data of 15 main OFDI host economies during a 2007-2021 period. Results and discussion: The regression results confirm that the size of the economy, social index, common borderline and level of economic integration have positive influence on OFDI flows of Vietnam. Meanwhile, geographical distance has negative effect on OFDI flows Implications: As practical policy recommendations, it is suggested to implement measures aimed at enhancing trade relations and to introduce a new strategy regarding the policy for Foreign Direct Investment (FDI) in neighboring regions. Originality: To the best of author’s knowledge, there has not been any in-depth academic study focusing on the Vietnam’s outward FDI. In addition, robustness checks have been conducted to ensure the validation of empirical findings.


INTRODUCTION
In the context of integration, in addition to attract more inward FDI flows for the development of the domestic economy, Vietnamese businesses are also increasingly interested in outward foreign direct investment (OFDI) to expand market share, enhancing competitiveness.With a series of new-generation Free Trade Agreements such as CPTPP, EVFTA, RCEP, etc along with deep integration into the ASEAN Economic Community (AEC), Vietnamese firms have gained a wide range of opportunities to promote OFDI activities.By the end of 2022, Vietnam has had 1611 OFDI investment projects with a total registered investment capital of over 21.75 billion USD.Vietnam's OFDI mostly focuses on the mining industry, meet expectation with the accumulated loss is up to 1.34 billion USD by the end of 2021.In comparison, losses incurred in OFDI projects of state-owned enterprises increased by 42% compared to 2020.With $293.3 million in losses from 8 projects in 2021, telecommunications are the field with the largest loss of projects.Moreover, the average capital scale of projects is still small and tends to decrease.To improve competitiveness, the top priority is to raise capital for OFDI projects.However, the reality is that while the number of projects is going up, the registered capital has decreased sharply after reaching its peak in 2010.This leads to the average capital size of each project tend to gradually decrease over time.In addition, the disbursed capital ratio is low, reaching less than 60% in the 2007-2021 period, also lead to a low amount of actual capital for projects.As a result, it will be difficult for Vietnamese enterprises to invest in modern technology and capture foreign market share with a small capital.These significant limits require a thorough investigation of the variables influencing Vietnam's capital flow from both home and host countries.To that end, this study is divided into five parts.In addition to the research introduction, part 2 introduces an overview of the research, part 3 addresses the research methodology.Research results are mentioned in section 4 and some policy implications are in section 5.

LITERATURE REVIEW
Research on factors affecting outward FDI have received the attention of many scholars worldwide.These studies can be divided into 3 main groups: a group of studies using the Eclectic Theory, a group of studies using the Investment Development Path model (IDP), and a group of studies using the Gravity Model.Regarding the IDP model, the main issues considered are how push factors from the home country affect that country's OFDI capital flows.With the Eclectic Theory, issues affecting OFDI capital flows are considered from both perspectives: push factors from the home country and pull factors from the host countries.In recent time, the Gravity model is an increasingly popular model used to evaluate factors   This study aims to bridge this gap using the gravity theory.Additionally, it investigates the impact of social and economic indexes on Vietnam's OFDI, aiming to propose practical policies to foster sustainable OFDI and development objectives for the country.

RESEARCH METHODS
To evaluate the factors affecting OFDI in Vietnam, the Gravity Model proposed by Tinbergen (1962) and developed by many other researchers in the review was used.
Mathematically, the gravity model of OFDI from country i to country j can be written as an equation as follows: Where: OFDI represents the flow of FDI from country i to country j, while GDP is Gross Domestic Product and Dis is the geographical distance between countries i and j.To determine the sign and magnitude of the coefficients of the explanatory variables, the fully adjusted least squares method (FMOLS) was used.The FMOLS estimator is considered as a suitable panel data estimator due to its ability to overcome common problems of endogenous bias and series correlation.To select a suitable regression model, the study conducted LLC, ADF and PP-Fisher tests to check the model's defects.Besides, to ensure the reliability of the results, the study also conducted robustness checks of the model.The results of the ADF test and PP test show that there is no pseudo-regression relationship between the variables in the model.After conducting tests to select the model, the study uses FMOLS estimates to determine the relationship between variables in the model.The results of the FMOLS test are as follows: The regression model results show that combined GDP as a proxy for economic size has a significant and positive impact on Vietnam's OFDI capital flows.A 1% increase in joint GDP would increase OFDI from Vietnam to key destinations by 0.12%.The geographical distance coefficient between Vietnam and partner countries shows a negative relationship, meaning neighbor countries will have an advantage in attracting FDI from Vietnam.For the economic index, the coefficient is positive but not statistically significant.Therefore, it can be affirmed that variables such as inflation rate, unemployment rate, and bilateral exchange rate as a single economic index do not play an important role in promoting /decreasing Vietnam's OFDI.For social indicator (including poverty rate, proportion of people of working age, and urbanization growth), the impact of social indicator is positive and statistically significant.When the social index increases by 1%, the OFDI from Vietnam to the main destinations can increase by 0.34%.

To
In addition, the coefficients of the dummy variable are all positive and statistically significant.
The regression results show that the degree of globalization (represented by the WTO variable) is less effective than proximity (represented by the Border variable) in promoting Vietnam's OFDI flows.

CONCLUSION
The results of FMOLS model show that joint GDP, social index, and shared borders have a positive influence on Vietnam's OFDI to major destinations.In addition, the level of economic integration and openness (represented by the WTO accession variable) also has a positive influence on Vietnam's OFDI.Meanwhile, the distance factor hurts Vietnam's OFDI capital flows.In reality, Vietnam's OFDI tend to prioritize neighbor countries such as Laos, Cambodia, and Myanmar, which are three countries in the top 5 countries receiving the largest FDI capital from Vietnam.Meanwhile, variables such as the inflation rate, unemployment rate, and bilateral exchange rate represented by the economic index do not play an important role in promoting Vietnam's OFDI capital flows.This is explained by the fact that OFDI of Vietnamese enterprises during the research period were mainly carried out by state-owned enterprises, with priorities to achieve political goals.Therefore, economic factors have not played an important role in investment decisions.To promote Vietnam's OFDI capital flows in the upcoming time, it is necessary to prioritize investments in countries with a high proportion of working-age workers (countries with a young population structure) and a high level of urbanization.Vietnam also needs more policies to encourage "nearby FDI" to promote Vietnam's OFDI to neighboring countries such as Laos, Cambodia, and countries in the ASEAN region.In addition, Vietnam needs to strengthen integration and take advantage of the opportunities that FTAs bring to promote Vietnam's OFDI to partner countries.furthermore, the Vietnamese authorities also need to closely monitor the investment efficiency of projects, especially projects implemented by state-owned enterprises.
Investigating the Determinants of Macroeconomic Factors on Outward FDI of Vietnam: The Gravity Model Approach ___________________________________________________________________________ Rev. Gest.Soc.Ambient.| Miami | v.18.n.8 | p.1-13 | e06134 | 2024.4 affecting OFDI capital flows of countries.This model was proposed by Tinbergen (1962) and developed by many scholars (such as Shun-Chiao Chang (2014), and Ryan W. Tang (2022)).The Gravity Model originates from physics, first developed in the field of economics to evaluate factors affecting bilateral trade between two countries.The Gravity Model of bilateral trade assumes that the volume of trade between two countries will be directly correlated to the size of their economies and, in reverse, proportional to the geographical distance between the two countries.In recent years, a few studies(Goh et al. (2013); Correa da Cunha (2022), Bashir A.J.
of people of working age, and combined GDP have a positive influence on countries' OFDI capital flows.Yonghui Han et al. (2022) study the impact of sister-city relationships between China and partner countries on China's OFDI, with confirmation that partnerships between sister cities and BRI (with participation in the Belt and Road Initiative), and combined GDP both promote China's OFDI.Correa da Cunha et al. (2022) study how host country factors influence OFDI flows in Latin American and Caribbean countries.The study uses the entropy weighting method and gravity model with the main variables namely institutional, infrastructure, technology, economic openness, combined GDP, and the proportion of people of working age.The authors assert that there is a positive relationship between macroeconomic performance, formal institutions, infrastructure, technology, and OFDI intensity.Strong formal institutions, along with the quality of infrastructure and technology, have a positive impact on the relationship between macroeconomic performance and OFDI intensity.Based on the aforementioned studies, it's evident that there's a gap in the literature regarding Vietnam's outward FDI trends.The examination of how various factors, from Vietnam and host contries, influence Vietnam's outward FDI has not been adequately explored.
To construct the indices (economic index, social index), the authors use the Principle Components Analysis technique (PCA).This is a commonly used technique to reduce dimensionality data (reducing the number of variables in the model) by creating new variables while still retaining the basic characteristics of the original component variables.The results of using the PCA technique are presented in Table2as follows: To evaluate the factors affecting the OFDI of Vietnam, the authors collect the data from top 15 OFDI host countries in the period 2007-2021.These 15 countries accounted for 93.9% of total OFDI capital of Vietnam during the 2007-2021 period.Therefore, the research results are valuable in determining the factors affecting the OFDI of Vietnam into the main host countries.
select the appropriate regression model, the study uses Levin, Li & Chu (LLC), ADF-Fisher, and Philips-Perron-Fisher tests for all variables (except dummy variables and distance variables).The results of these tests are shown in the following table: The data in square brackets are p-values Note 2: LOFDI, LGDP, LECOIN, LSOIN are logarithmic values of OFDI, GDP, economic index and social index.Source: Calculated on Stata The results of the tests show that all the variables in the model are nonstationary, but the first difference between the variables are stationary.The study also used a cointegration test to detect spurious regression relationships.The results of the cointegration test are presented in the following table: Stoian and Alex Mohr (2016)r (2016)used data from 29 emerging economies over 17 years(1995 -2011)to evaluate factors affecting OFDI in emerging economies.Key variables include protectionism, level of corruption, competitive advantage, company ownership advantage, urbanization rate, and gaps in regulations and laws of the home country.
Tang et al. (2022)gawa (2022)s no visible impact.Tian, W. and M. Yu (2020) studied China's OFDI in the period 2003-2018.The authors affirmed that China's OFDI tend to focus on large markets and countries with abundant natural resources but weak institutions.Chen.J.E et al. (2019) studied the factors that determine the investment location of Malaysia's OFDI with panel data from 34 receiving countries in the period 2000-2018, with confirmation that the important factors that have a positive influence on OFDI are the market size of the host countries and geographical distance.Meanwhile, Sheng Ma et al. (2020) used panel data for the period 2005-2018 to evaluate the location advantage that China chose to carry out OFDI, especially in ASEAN countries.The authors confirmed that China's OFDI often focuses on ASEAN countries with large potential markets and low tax rates, favorable business environments, rapid urbanization rates, and low labor costs (without the goal of natural resource seeking).The authors confirm the positive relationship between regulatory gaps in the home country, ownership advantages, and OFDI for emerging economies.Jiang, G. and X. Zhang (2016) used panel data from 32 developing countries in the period of 2000 -2014 to evaluate factors affecting OFDI in developing countries, including main variables: Level of economic development; trade openness; political risks; technological innovation; monetary power.The authors confirmed that political risk hurts OFDI while investment in science and technology has a positive impact.Shun-Chiao Chang (2014) used the gravity model to evaluate the factors affecting China's OFDI.Key variables include China OFDI, GDP, population, distance, relativeexchange rate, unemployment rate, poverty rate, patents, exports fuel, ore and metal exports, and dummy variables (border, language).The author confirms that GDP, common language, and fuel exports have a positive impact on promoting OFDI, but the distance variable has no impact on China's OFDI.Li, S. and H. Nakagawa (2022)studied factors affecting China's OFDI capital flows with panel data in the period 2003 -2020, by using a gravity model with the following variables: OFDI, exports, imports, GDP per capita, average exchange rate, inflation, GDP growth, IFDI/GDP ratio, and the ratio of ore and metal exports to commodity exports.environmentandlevel of integration have a positive influence on OFDI of China into BRI countries.Ryan W.Tang et al. (2022)study the effects of institutional distance, combined GDP, borders, bilateral exchange rates, WTO accession, and unemployment rate on OFDI capital flows for developing countries.The authors confirm that institutional distance, the proportion is the OFDI of country i investing in country j in year t; GDP is the annual gross domestic product of countries; Ecoin and Soin respectively represent the economic index and social index.The variable DIS is the distance of the partner country to Vietnam.Dummy variables BORDER and WTO are used to determine whether two countries share a common border or have joined the WTO.The data for the study were taken from 2007 to 2021.The authors chose the research period starting from 2007 because this is the year Vietnam joined the WTO -an important milestone for Vietnam in integration and economic opening.In addition, the authors convert all variables in the model to logarithmic form to minimize model defects such as variable variance.

Table 2
Principle Components Analysis results

Table 3
Top 15 host countries of Vietnam's OFDI Unit: Million USD, %

Table 5
Cointegration test results

Table 6
Regression model results