An empirical study of the real effective exchange rate and foreign direct investment in Vietnam

Foreign direct investment (FDI) inflows to Vietnam have increased significantly in recent years. Theoretically, capital inflows will put pressure on the overvaluation of local currencies in countries, despite different exchange rate mechanisms. So, the problem facing the Vietnamese government is the need to examine the relationship between the exchange rate and FDI in order to develop effective policies. This study examined the relationship between the exchange rate and FDI in Vietnam in the period of 2005–2019 using the VAR (vector autoregression) model based on quarterly frequency data. The new points of this study are: (i) using the real effective exchange rate (REER) of the Vietnamese currency with 143 major trading partners of Vietnam; and (ii) adding two control variables into the VAR model to examine the relationship between the exchange rate and FDI in Vietnam – a case study for developing countries. The findings show that, firstly, there is a positive causal relationship between FDI and Vietnam’s real effective exchange rate. Secondly, trade openness has a positive impact on FDI and REER in Vietnam. Thirdly, economic growth has an impact on REER, but no statistically significant impact on FDI was found. The findings can provide useful information to help policymakers plan and make decisions on future policies and support further research studies.


INTRODUCTION
According to Barrell and Pain (1997), Borensztein et al. (1998), FDI is a capital source that plays an important role in promoting productivity growth in both developed and developing countries. In particular, for developing countries, FDI is seen as a valuable additional source of domestic savings. On the other hand, when FDI enterprises invest in a country, they also help transfer technology, create jobs, and train workers for knowledge and skills. These benefits reinforce the importance of FDI.
Over the past decades, many studies around the world have been conducted to assess the relationship of exchange rates and FDI inflows. Overall, existing studies provide evidence that a relationship exists between these two variables (Froot & Stein, 1991;Blonigen, 1997;Takagi & Shi, 2011;Boateng et al., 2015;Djulius, 2017). However, studies are still controversial about the direction of the impact between FDI and the exchange rate. On the other hand, the studies show that the characteristics of the relationship between these two variables may differ due to the characteristics of the research data, the research model and the exchange rate regime in each country.
In the process of global integration, foreign direct investment (FDI) is becoming more and more important for Vietnam, a developing country. According to the General Statistics Office of Vietnam (2018), the foreign direct investment sector accounts for approximately 20% of gross domestic product (GDP), with the contribution increasing over the years. The FDI sectors help to improve Vietnam's export capacity, becoming a bridge for Vietnam to quickly approach and cooperate with many countries, step by step raising the position and strength of Vietnam in the globalization context. Exports from the FDI sector averaged over 66% of Vietnam's total merchandise exports in the 2011-2016 period (General Statistics Office of Vietnam, 2018).
For many years, Vietnam has been known as a developing country with a relatively high growth rate and a stable exchange rate, increasing trade openness. These features have contributed to consolidating Vietnam's image as an attractive destination for foreign investors. As a result, FDI inflows to Vietnam have increased significantly in recent years. Theoretically, capital inflows will put pressure on the overvaluation of local currencies in countries, despite different exchange rate mechanisms. In a flexible exchange rate regime, an increase in the real value of the local currency occurs through an increase in the nominal value of the local currency. Meanwhile, in a fixed exchange rate regime, an increase in the real value of the local currency is mainly driven by higher prices for non-commercial goods.
Vietnam is a country that is pursuing a mixed exchange rate mechanism. Therefore, the appreciation of VND depends on both processes: an increase in the nominal value of the local currency and an increase in the price of non-commercial goods. The issue to be concerned about is how FDI inflows to Vietnam are related to the exchange rate, from which appropriate policies can be proposed to improve the efficiency of FDI inflows. However, studies on the relationship between the exchange rate and FDI in Vietnam are still limited in number and focus on one-way impacts of the exchange rate on FDI (Pham & Nguyen, 2013). Therefore, this study attempts to contribute to the current research gap. This paper provides the empirical evidence on the two-way relationship between the exchange rate and FDI with the following key points: (1) using the real effective exchange rate variable in Vietnam (REER) with a large number of trading partners (including 143 partner countries); and (2) considering the impact of trade openness and economic growth as control variables in the VAR model of the relationship between REER and FDI in Vietnam.

Theoretical approaches
According to Cushman (1985), for market-oriented FDI (mainly using input resources from the home country to produce and sell products in the host country), the depreciation of the host country's currency will cause profits to be transferred back to the home country become low, discouraging investors and decreasing FDI. However, if cost-driven FDI mainly uses the input resources of the host country to produce and then sell products to the home country or export to a third country, the depreciation of the recipient country's currency will cause production costs of FDI enterprises reduce, creating an advantage for exports, thereby promoting FDI. Dunning (1988) introduces the OLI (Ownership, Location, Internalization) model that states that the motive for offshore investment of multinational companies is that they possess certain advantages in terms of ownership, and location and localization enable them to compete with domestic enterprises in the host countries' market. Dunning's OLI model (1988) emphasizes the direct role of real exchange rates for FDI arguing that one driving force for FDI is that multinational companies want to move to lower cost locations (including costs related to exchange rates). Dunning (1988) argued that investment decisions of companies that want to own income-generating assets abroad tend to be influenced by financial and exchange rate variables.
The theory of market imperfections states that markets with imperfections make business inefficient. Meanwhile, businesses will implement direct investment abroad to promote business activities and overcome those imperfect factors. Two theories that analyze the effects of exchange rates on imperfect markets have a great influence on the theoretical approach to the relationship between the exchange rate and FDI later: Froot and Stein (1991) and Blonigen (1997). Froot and Stein (1991) argue that the relationship between exchange rates and FDI is due to market imperfections. A change in the exchange rate would increase the financial value of companies holding denominated assets in a currency that would increase in value versus firms holding denominated assets in currency that would decrease in value. When the other conditions are constant, the value of the recipient country's currency is inversely proportional to the FDI in that country. Blonigen (1997) further emphasized that exchange rates have the greatest impact on FDI acquisitions of high technology industries (industries in which special assets of enterprises are of significant importance).
However, using a theoretical approach based on expanded production, Campa (1993) suggests that exchange rates may influence future profit expectations. Therefore, the depreciation of the recipient country's currency may weaken FDI inflows into that country. Itagaki (1981) makes another argument based on the purchasing power parity (PPP) theory of Gustav Cassel (1920). According to Itagaki (1981), if PPP theory really exists, foreign companies will not have any risks related to exchange rates because the exchange rate changes will be offset by changes in the relative price among countries. However, the conclusion by Itagaki (1981) remains controversial as to whether PPP holds even in the long run, let alone at all points in time and for relative price levels in all industries. In general, the relationship between exchange rates and FDI is a problem that has been of great interest in many countries. Research results on this issue are quite diverse, indicating many research gaps. First, most studies agree that there is a relationship between exchange rates and FDI, while some studies have not found statistically significant evidence. Secondly, studies on the relationship between exchange rates and FDI have been carried out in developed countries, while in emerging markets, they are limited in number and only encompass the recent decades. Thirdly, the characteristics of the research country and the research period significantly affect the research results. Vietnam is a developing country, trying to reform policies (including the exchange rate regime) to attract FDI inflows, but research in this field is limited in number and in the absence of quantitative research. Therefore, this study has been conducted to fill the gap in research on the relationship between exchange rates and FDI in Vietnam, a case study for emerging countries.

METHODOLOGY
Based on previous studies, such as Mo (1988) Table 1).

Unit root tests
The study used the Dickey-Fuller test (Dickey & Fuller, 1979) to test the stationarity of the data series. According to hypothesis H0, the data series are not stationary. Stability test results show that REER variables are not stationary in the original data series. Therefore, the study conducted the first-order differential calculation for all variables, and then tested the sta-tionarity of all first-order differential series. The stationarity test results confirmed that all series were stationary in the first-order differential. Thus, it was decided to use the first-order differential data series to conduct the study.

Determining the optimal lag
Based on the results of testing the optimal lag, the criteria LR and AIC showed that the model has an optimal lag of 3 and a lag of 4. However, to explain more clearly the relationship between the variables, the study identified the estimation model with a lag of 4.

Cointegration test
The results of the cointegration test (see Table 3) showed that there was no cointegration in the long term in the data series; therefore, the VAR model is used in the study to examine the relationship between the real effective exchange rate and foreign direct investment in Vietnam, 2005:Q4 -2019:Q3 (Engle & Granger, 1987).

VAR coefficients
The results of the stability test of the model and the autocorrelation test showed that the VAR model with four lags is stable and appropriate. Testing results of the VAR model show that the real effective exchange rate and FDI have a two-way relationship. This result is consistent with previous studies such as Kosteletou and Liargovas (2000), Shrikhande (2002), Lee (2015), and Tsaurai (2015). At the same time, the research results show that the effective FDI and exchange rate variables were most affected by these factors in the past.

The effect of the real effective exchange rate on FDI inflows
The research results show that DREER (-4) has a positive effect on FDI inflows to Vietnam. This means that VND depreciation increases FDI inflow to Vietnam. The decrease in the value of VND increased the wealth of foreign investors,  Note: * , ** and *** indicate significance at the 10%, 5% and 1% levels, respectively. making investment costs relatively lower, thus encouraging investment. This effect takes place at the 4-quarter lag (which is a relatively long period of time), which is also consistent with the nature of FDI inflows, which are fixed, decisions related to FDI can only change in the long term. This result is in line with Froot and Stein (1991), Blonigen Lartey (2011) argues that an increase in FDI inflows leads to an increase in the value of the host country's currency only in countries with financial openness. According to Lartey (2011), in economies with financial openness, there is a trade-off between resource shifting effect and spending effect after FDI inflows increase. Thus, this result may reflect the fundamental characteristics of Vietnam, with insufficient financial openness.
The chart of FDI and real exchange rates in Vietnam in recent years contributes to illustrating the results of the study. Figure 3 shows the upward trend between the effective exchange rate and FDI in Vietnam. In fact, Vietnamese currency depreciation reduces production costs and asset prices for foreign investors, thus attracting FDI in Vietnam. Moreover, the depreciation  In addition, the trade openness factor has a positive impact on the effective exchange rate and FDI. Specifically, the trade openness of DOPEN(-2), DOPEN(-3) and DOPEN(-4) has a positive impact on FDI. Recently, implementing the policy of international integration, Vietnam's economy has a relatively high trade openness and has increased relatively quickly, accompanied by a significant increase in FDI inflows. However, the realistic analysis of the characteristics of the openness of Vietnam's economy still points out many weaknesses. High trade openness and rapid increase are largely due to the large contribution of the FDI sector. Import and export of Vietnam reached the state of trade surplus due to the relative decrease in import. Therefore, in the coming time, the Government should have policies to improve trade openness towards increasing the advantages of trade openness, promoting the internal strength of Vietnam's economy, creating a good foundation for the stability of the economy, continue to increase FDI attraction.
The study also found that economic growth has a small but statistically significant impact on the real effective exchange rate. Economic growth at a 1-period lag has a positive impact on REER. Economic growth at a 2-period lag was found to have a negative effect on the real effective exchange rate. Meanwhile, the study could not reveal a statistically significant impact of economic growth on FDI in Vietnam. Vietnam's economic growth in recent years has remained stable at a high level in recent years (average GDP growth of 6.5% -7%/ year). In the coming time, the Government should continue to stabilize the economy, strive to maintain a high and sustainable growth rate in order to have an impact on FDI attraction.
On the other hand, the study found that past FDI at one-and two-period lags had a negative impact on current FDI. Therefore, the government should control to prevent hot growth of the economy, avoid attracting too much FDI leading to inefficient use allocation. At the same time, the government should select and have policies to promote FDI attraction in areas that can create good premises to attract FDI investors in the future, such as attracting FDI in the field of infrastructure development, especially transport infrastructure for the whole country; FDI attraction focuses on high-tech industries to access advanced technology and strengthen the human resource training of Vietnam.
Test results also show that REER in the past at a 4-period lag has a positive impact on the current REER. According to Combes, Kinda, and Plane (2012), a more flexible exchange rate regime can undermine the appreciation of the real effective exchange rate (derived from the pressure of inflows). Recently, the Central Bank of Vietnam has had effective exchange rate management solutions that need to continue to be promoted: allowing the exchange rate to be more flexible, closely monitoring the movements of the domestic and international foreign currency market (especially the large markets for Vietnam such as the US and China) in order to flexibly manage the exchange rate in the direction of loosening the exchange rate band, allowing the market to adjust itself according to the supply and demand of foreign currencies.

CONCLUSION
The study used the vector autoregression (VAR) model to test the relationship between exchange rates and FDI in Vietnam for the period 2005-2019. The research found an important empirical evidence on the relationship between the exchange rate and FDI in Vietnam with three key findings. First, there Source: Authors' analysis results. is a positive causal relationship between FDI and Vietnam's real effective exchange rate. Second, trade openness has a positive impact on FDI and REER in Vietnam. Third, economic growth has an impact on REER but did not have a statistically significant impact on FDI.
According to the research results, an increase in REER (Vietnamese currency depreciates) will attract more FDI inflows. FDI and REER have been most affected by these factors in the past. Therefore, maintaining a stable exchange rate in the direction of continuous real depreciation will benefit FDI attraction.
At the same time, the research results show that trade openness has a positive impact on FDI and REER. Therefore, it is necessary to pursue policies aimed at improving the quality of trade openness, continuing to increase the number of trade openness, creating a good foundation for exchange rate stability and increasing the attraction of foreign investment.