Determinants of Vietnam’s export to EU market Thanh-Long Trana* and Minh-Vuong Vob

Article history: Received April 1 2020 Received in revised format April 1 2020 Accepted April 27 2020 Available online April 27 2020 The paper uses the gravity model and panel data to examine the determinants of Vietnam's export to the EU market during the 2007-2017 period. The article uses the Hausman-Taylor estimator for the optimal combination of Fixed-Effects (FE) and Random-Effects (RE) methods. The estimation results indicate that the determinants of Vietnam's export to the EU market include: aggregate population (+2,512); aggregate GDP per capita (+1,256); Vietnam's index of infrastructure (+0,824) and aggregate trade openness (+0,377). Based on these results, the paper suggests policies to boost Vietnam's export to the EU. y the authors; licensee Growing Science, Canada b 20 © 20


Introduction
In the context of international economic integration, Vietnam is among countries with very high export growth. According to the data from UN Comtrade, Vietnam recorded very high export turnovers in the 2007-2017 period with 4.5 times increase in export volumes, from 48 billion USD in 2007 to 215 billion USD in 2017(COMTRADE, 2017. The European Union (EU) is the largest economy in the world with a GDP of 18.82 trillion USD and a population of 512 million in 2017 and is also the second-largest import market of Vietnam. In 2017, Vietnam's export turnover to the EU market reached 38.31 billion USD, placing it second to only our export turnover to the US (41.5 billion USD) and above China (35 billion USD) (COMTRADE, 2017). A special characteristic in the two-way trade relationship between Vietnam and the EU is the high complementarity and low competitiveness, which favours Vietnam's exportation. However, Vietnam's export maintains a modest position in the EU. According to UN Comtrade's data, Vietnam's exports only accounted for 1.99% of the EU's import turnover in 2017 (COMTRADE, 2017). Therefore, the paper raises the following research question: Which are determinants of Vietnam's export to the EU market? In search of an answer to the question above, the paper uses the gravity model to evaluate the determinants of exports or trade flows of a country. In particular, relevant domestic studies include Tho (2013), Mai (2017) with a focus on Vietnam's exports, Duong (2016) on trade flows between Vietnam and the EU. Through prior studies from Vietnamese authors, there has not been any research that thoroughly examines the determinants of Vietnam's export to the EU market. Therefore, it is necessary to estimate the determinants of Vietnam-EU export market growth. The paper is structured as follows. Section 2 includes theoretical framework and proposed research model. Section 3 is about data and methods. Section 4 presents empirical results. Section 5 concludes. 422

Theoretical framework
Export is the exchange of goods and services from one country to another through market relations to make use of the country's advantages and thus allow the country to benefit from them. Export activities can take various forms, mainly in the following five: direct export, indirect export, re-export, on-site export and export processing. In terms of economic development of a country, export plays four main roles: First, export serves as the primary capital source for import to augment the industrialization process of the country; Secondly, export contributes to economic restructuring and endorses production; Thirdly, export improves the employment situation as well as living circumstances; Lastly, export is fundamental to expand and promote foreign economic relations. The gravity model of international trade is used by many economic researchers worldwide to analyse trade activities between two or more countries. The model's application was first reported by (Tinbergen, 1962) and (Pöyhönen, 1963) in the following form: where is the constant; is the value of trade flow from country i to country j; and are respectively the gross domestic product (GDP) of country i and j and finally, is the geographical distance between country i and j. The value of export flow from country i to country j is positively correlated with the GDP of each country and negatively correlated with the geographical distance between the two countries. From the initial research model, researchers have also included other factors into the model, such as per capita income, tariffs (Linnemann, 1966), border effects (Anderson, 1979), commodity prices, real exchange rates (Bergstrand, 1985), expenditure capabilities, taste preferences (Bergstrand, 1989), while other researchers helped to improve the model, such as Helpman and Krugman (1985), Anderson (1979), and Song (2011). Based on all studies mentioned above, the current generalized gravity model includes three main groups of factors impacting export or trade flows, namely supply-side factors (regarding exporting country i) demand-side factors (regarding importing country j), as well as facilitating and obstructing factors. The generalized gravity model has the following linear form: where, = 1,2,3, . . . , is the value of the independent variable , is the number of independent quantitative variables; = 1,2,3, . . . , is the value of dummy variable , which takes the value of 1 under certain circumstances and equals 0 otherwise; m is the number of dummy variables. This paper uses the generalized gravity model (2) to propose a research model for examining the determinants of Vietnam's export to the EU market. The focal point of the research model is that the author uses the variable combination method to better identify the impact of such factors on export and uses delayed variables to eliminate the impact of the dependent variable (export turnover) on independent variables (including GDP) (Nguyen, 2010;Tesfaye, 2014;Duong, 2016). The model can be expressed in the natural logarithmic form as follows: where: : exporting country (Vietnam), ( = 1,2, … 28): importing countries (28 EU countries), = 2007, 2008, … , 2017. 1 + : Logarithm of Vietnam's export turnover to importing country j in year t. As there was no record of Vietnam's export turnover to some importing countries j (which did not indicate the lack of export activities), the author uses 1 + for accuracy.

Data and methods
The article uses the panel data of Vietnam and 28 EU countries from the 2007-2017 period. The author applies the Hausman-Taylor estimator (Hausman & Taylor, 1981) for the optimal combination of Fixed-effects (FE) and Random-effects (RE) for panel data. This paper uses the statistical software Stata 13.0. The author uses panel data of Vietnam and 28 EU countries,

Empirical results
As aforementioned, the author first applies pooled OLS and RE estimators and conducts tests on estimation results. Next, FEM is employed and the Hausman Test is operated to decide the best method between REM and FEM. As a result, FEM is selected. Consequently, through the Breusch-Pagan Lagrange Multiplier Test (LM), REM is proved to be the optimal method. Limitations of FEM are examined, where heteroscedasticity is detected (  (2015), and Do Thi Hoa Nha and Nguyen Thi Thanh Ha (2019).

Table 2
The results of estimation Secondly, at a 1% significance level, aggregate GDP per capita is positively correlated with Vietnam's export turnover to the EU market. If the coefficient increases by 1%, export turnover increases by 1.2561%. This result is consistent with the research hypothesis and the studies by Trần Nhuận Kiên and Ngô Thị Mỹ (2015), and Do Thi Hoa Nha and Nguyen Thi Thanh Ha (2019). Thirdly, at a 5% significance level, aggregate trade openness is positively correlated with Vietnam's export turnover to the EU market. If the coefficient increases by 1%, export turnover increases by 0.3773%. This result is consistent with the research hypothesis and the study by Abidin andSahlan (2013), Trần Nhuận Kiên andNgô Thị Mỹ (2015), and Do Thi Hoa Nha and Nguyen Thi Thanh Ha (2019). Fourthly, at a 10% significance level, Vietnam's index of infrastructure is positively correlated with Vietnam's export turnover to the EU market. If the coefficient increases by 1%, export turnover increases by 0.8243%. This result is consistent with the research hypothesis and the studies by Martínez-Zarzoso and Nowak-Lehmann (2003).

Conclusion and implication
Based on the above research results, this paper has proposed the following policy implications to boost Vietnam's export turnover to the EU market. First, aggregate GDP per capita should be taken into consideration. It is vital for Vietnam to maintain high growth rate to create stable conditions for the export growth to the EU market: persistently aiming to maintain macroeconomic stability, regulating inflation, preserving the economy's major balances, improving the business investment environment, endorsing economic growth; promoting institutional improvement, improving the appraisal of legal documents to ensure the uniformity and feasibility of the legal system; advocating economic restructuring in line with the innovation of growth model, enhancing growth quality, labour productivity and the competitiveness of the economy to make more substantial changes in every sector and industry, with regard to the implementation of the national strategy for the Industry 4.0. Regarding EU countries, export to markets with high GDP per capita should be promoted. Secondly, the aggregate population should be taken into consideration. Vietnam's Government should make use of abundant labour supply in both quantity and quality, focus on the specialization of the labour force and attract talents. For EU countries, exports to highly populated countries should be promoted. Thirdly, the infrastructure of Vietnam should be taken into consideration. The Government should take measures to modernize the economic infrastructure (including hard and soft infrastructure). Fourthly, the aggregate trade openness should be taken into consideration. The Government should endorse negotiations as well as bilateral and multilateral free trade agreements, strengthen trade promotion activities to boost Vietnam's exports to other markets and in turn attract investment in terms of capital, modern technology, and equipment from these countries. This study has estimated the impact level of determinants of Vietnam's export turnover to the EU market. However, the study is subject to the following limitations: the study has not quantitatively analysed specific key export products to the EU, nor has it utilized SWOT Analysis. Therefore, it is possible to do some research on Vietnam's key export products to the EU market in details by applying SWOT Analysis.