Determinants of FDI attractiveness: Evidence from ASEAN-7 countries

Abstract

The article examines the impact of tax burden, institutional quality, and some macroeconomic determinants on attracting inflows of FDI in ASEAN-7. We estimate the panel data sample with pooling, a seemingly unrelated regression (SUR), and fixed-effects models. The empirical results indicate that economic growth, tax burden, quality of economic institution, and inflation are key drivers that significantly attract FDI while population growth and quality of political institution are negatively associated with FDI inflow. These findings may connect to the price transferring and demand-following hypothesis. Our results provide some important implications for policy-makers in ASEAN developing countries that are enhancing FDI inflow in the future.

Keywords:

• FDI attractiveness
• tax burden
• institutional quality
• macroeconomic determinants
• ASEAN

JEL Classification:

• E02
• E22
• H20

PUBLIC INTEREST STATEMENT

Attracting FDI inflows is a topic that always receives public attention. ASEAN countries have always tried to attract this inflow with many incentive policies through institutional quality reform and tax competition. Besides, FDI inflows also depend on other macroeconomic factors in the host country such as growth, ability to control inflation and population growth (market power). This article evaluates whether these efforts to reform institutional quality and cut down tax rate in ASEAN countries have been effective in attracting FDI inflows recently. Besides, the mentioned macroeconomic factors really contribute to attracting FDI inflow. We believe that the empirical results will provide many interesting evidences for this issue.

1. Introduction

Foreign Direct Investment (FDI) is an investment made by a foreign firm or individual to the business located in the host country. According to the United Nations Conference on Trade and Development (UNCTAD), FDI can supply capital flows and foreign currency through investments, create appropriate funds for domestic capital needs, and provide multiple benefits such as technology transfer and managerial skills. FDI thus increases competitiveness in the domestic economy and improves employment opportunities. Theoretical models and empirical evidence that support this can be found in Behrman (1960), Findlay (1978), Borensztein et al. (1998), Blomström et al. (2001), Zhang (2001), and Alfaro et al. (2004), and Dunning (2013), and Sabir et al. (2019).

Besides the internal factors of the country, such as policies for growth, as well as the advantages of geographical location, human resources, public investment, and the legal system, policy that attracts capital is also considered one of the most important factors contributing to the country’s economic growth. Previously, the literature found that not only domestic capital but also foreign direct investment plays an important role in developing countries, and developed countries are also interested in attracting this capital. Internal investment capital resources are limited in developing countries, so looking forward to foreign capital flows is really necessary and the issue is always a top priority. In particular, ASEAN countries have been gradually joining international trade organizations such as WTO, ASEAN+6 and TPP, attracting FDI has become more and more important.

In the recent two decades, ASEAN has attracted the world through its dynamic and steady growth. This region is predicted to become one of the five largest economies in the world in the near future (Nguyen, 2021a; Nguyen & Yang, 2020). However, most of the countries in ASEAN are developing countries with steady economic growth. Similar to other developing countries, these countries always have a strong demand for capital investment due to the low rate of gross domestic savings. Therefore, foreign capital inflows play a crucial role in economic development and sources of breakthrough. This analysis supports the dominant argument that there is a strong nexus between FDI and economic activities in developing countries.

A majority of research on FDI also focuses on the determinants of FDI’s attractiveness (Blonigen, 2005). As summarized in the work of Nunnenkamp (2002), empirical studies on FDI determinants in developing countries may not have the same vein. More precisely, the uncertainty regarding whether FDI in developing countries has moved from market and resource-seeking to efficiency-seeking still challenges academics (Biswas, 2002; Bokpin, 2017; Elheddad, 2018; Kumari & Sharma, 2018; Lucas, 1993; Mottaleb & Kalirajan, 2010; Tsai, 1994). Previous literature has shown that macroeconomic indicators have a significant impact on FDI inflows. Capturing these impacts in a timely manner is vital for investors’ investment. When observing the change in macroeconomic indicators, investors can preliminarily predict the change of FDI inflows into their country to inform their major decisions and implement appropriate policies and strategies to maximize the investor’s profit rate. Typically, previous empirical studies have been conducted on a large scale in many developing countries with the aim of obtaining scientific evidence for the influence of national macroeconomic variables to attract FDI inflows. Moreover, these empirical studies also demonstrate that FDI’s primary focus is business activities and often flows into economies with many comparative advantages, such as cheap labor, attractive tax incentives, and have abundant mineral resources and raw materials (Zhang, 2001).

Additionally, the tax burden and institutional quality of the host country are also factors that foreign investors consider in making their investment decisions. The tax burden directly impacts the return of investment capital flows and has an indirect impact on the competitiveness of enterprises. Bellak and Leibrecht (2009) show that the corporate income tax rate negatively impacts the profitability of FDI investment. ASEAN countries are considered attractive destinations for FDI inflows thanks to their comparative advantages in cheap labor, attractive preferential policies, rich mineral resources, and abundant raw materials. However, due to limited financial resources and considerable budget deficit pressure, it is understandable that these nations’ governments exercise high tax rates to compensate for their budget.

Currently, poor institutional quality is also a global problem, not only in each country but also in an entire region affecting many aspects of the economy. The poor institutional quality leads to corruption. Theoretically, corruption can be seen as the “grabbing hand” because it increases the risks of transaction costs and hinders FDI inflows. However, corruption occasionally acts as a “helping hand” because it helps lubricate the wheel of commerce in countries whose institutions are still heavily administrative and cumbersome. Consequently, businesses earn more profits when they only spend a small amount of money to get important information and benefits (Heckelman & Powell, 2010).

In the contemporary economy, tax competition between countries to attract FDI is becoming a global issue. Investors often compare tax burdens across countries that are similar in market size and geographic location. While the international tax competition tends to increase, tax rate cuts in these countries are considered inevitable. However, there is no concrete evidence that this tax cut will help attract FDI inflows in ASEAN countries. The reduction in tax revenue will decrease infrastructural investment, causing a decline in the provision of public goods and services and distortions in allocating government resources. Therefore, whether these countries are still ideal destinations for foreign investors or not remains unclear.

Moreover, poor institutional quality leading to corruption can be one of the main constraints hindering economic growth and development. In particular, the World Bank and Transparency International show that the corruption problem has become more complex and widespread in some ASEAN countries. Findings from multiple transnational empirical studies demonstrate that corruption and poor institutional quality tend to slow down economic development by cutting down private investment and lowering efficiency in public investment (Gupta et al., 2000; Knack & Keefer, 1995; Mauro, 1995; Tanzi & Davoodi, 1998). Most of these results are consistent with information about neoclassical profiteering activities. However, Ehrlich and Lui (1999) argue that the consequences of corruption are pretty diverse and significantly affect the economy. In various African and South American countries, corruption severely limits growth. However, rampant corruption does not appear to be slowing growth in many countries with regional differences, such as China and India.

The share of Foreign Direct Investment (FDI) in the total capital flows of ASEAN countries has increased greatly in recent years, and the proportion of FDI enterprises’ contribution to economic growth has seen rise as well. Officially, FDI inflows have been present in many ASEAN countries since 1990. Since then, these inflows have proven efficiency in bringing several positive effects to the domestic economy, and such impacts are increasing globally. The spillover effect has become widely spread in countries around the world, along with diversity in investment fields and business cooperation strategies.

While previous studies, such as Bailey (2018), Bellak and Leibrecht (2009), and Sabir et al. (2019) in relation to FDI attractiveness often focused on analyzing individual factors that can affect the above-mentioned attractiveness, this paper aims to provide new and more detailed insights regarding the determinants of FDI inflows in the ASEAN-7. This will be achieved through investigating the potential factors that can affect FDI in the context of FDI in ASEAN countries because it is increasingly attractive. More specifically, the article will deal with two significant factors: tax burden and institutional quality.

The rest of the paper is organized as follows. In section 2, we will briefly review the relevant literature concerning the FDI-growth and determinants of FDI inflows. In the subsequent section 3, we will describe the data and estimation techniques. Afterwards, a discussion on the results will be covered in section 4, and a conclusion will be drawn in section 5.

2. Literature review

2.1. Tax burden and FDI

Although the impact of taxes on FDI inflows greatly depends on each type of tax, most empirical studies suggest that countries with high tax rates will not be as attractive to FDI inflows as countries with low rates of taxation. However, Hartman () was the first on the subject to conclude that some types of FDI inflows may not be so sensitive to taxes. This implies that FDI investors, in certain sectors, are not under pressure from the tax burden of the host country.

De Mooij and Ederveen (2003), using a meta-analysis method, showed that there exists a tax elasticity for FDI of negative 3.3, meaning that on average 1% decrease in the tax rate of the receiving country will increase FDI inflows to this country by 3.3%. Meanwhile, Bellak et al. (2007) re-examined 8 previous studies on this issue. The results showed that this elasticity is only negative by 1.45.

Besides, Stöwhase (2005) analyzed the sensitivity of FDI in terms of tax rates. He concluded that this sensitivity considerably depends on the area receiving FDI inflows. Therefore, this study concludes the existence of an under-estimate or over-estimate of the tax elasticity of FDI compared to the average referred to by previous studies. The study also suggested that difficulties in data access, measurement and estimation methods can lead to biases in previous studies.

2.2. Institutional quality and FDI

Corruption is regarded as an institutional factor, and is defined by the World Bank as the abuse of public authority for personal gain. In many cases, corruption is believed to have a negative impact on FDI inflows. However, there is no strong link between corruption and FDI.

Wheeler and Mody (1992) considered the impact of corruption on FDI in the context of poor institutional quality of countries. This quality portrays itself in the form of cumbersome administrative procedures, bad bureaucracy, and a lack of transparency in the legal system. The study showed that the corruption impact on FDI is not statistically significant. In other words, corruption does not impede FDI inflows because of weak institutions in such countries. However, Wei (2000) pointed out that the study of Wheeler and Mody (1992) has some limitations and leads to bias in the research results. Wei (2000) argues that Wheeler and Mody (1992) have included the corruption variable in 12 other variables in the model. Therefore, the assessment of the corruption effects on FDI, in this case, is ambiguous. Therefore, Wei (2000) conducted data mining of 45 countries, and three different sources exploited the corruption variable. By the estimation method of the Tobit model, study results indicate that corruption harms FDI with a negative correlation. Abed and Davoodi (2002) used cross-sectional and panel data to analyze the impact of corruption levels on per capita FDI flows in transition economies. The results illustrate that countries with low levels of corruption attract more FDI inflows. Nevertheless, the corruption variable became insignificant when they included the control variable for institutional reform in the model. Therefore, this study sheds light on a significant conclusion that institutional reform is more critical than reducing corruption levels for attracting FDI inflows in various countries.

Habib and Zurawicki (2002) analyzed the impact of corruption on bilateral FDI through conducting samples from 7 funding countries, and 89 receiving ones. The article in question tested the hypothesis that if the level of corruption in the host country is higher than that of home, then the FDI inflows will be reduced. Therefore, the empirical model utilized the difference in corruption levels between the home and the host countries as the explanatory variable. It was argued that FDI inflows tend to avoid corrupted issues; as they are considered as unethical problems. Moreover, Voyer and Beamish (2004) used single data for a source country (Japan), and 59 recipient countries whose economies are emerging and developing. Using cross-unit regression, they found evidence that FDI inflows from Japan have an inverse relationship with host country corruption. Asiedu () when studying foreign direct investment in Africa, tested for the main factors determining FDI inflows into Africa. The results illustrate that both corruption and political instability have a negative impact on FDI inflows.

Mathur and Singh (2013) indicated the foreign investors’ concern about economic freedom, rather than the political freedom to make decisions for their capital flow. The paper investigated the determinants of FDI inflows into 29 emerging countries. The empirical results showed that corruption has a significant impact on choosing the investors’ destination. Particularly, FDI inflows into developing countries are highly interdependent. The high level of corruption negatively affects FDI inflows into a country.

Other studies suggested that corruption does not leave a negative impact on FDI. This assumes that corruption can act as a helping hand in some situations where other aspects of government are lacking, or where economic policy is perceived to be ineffective (Leff, 1964). In some cases, corruption is beneficial because it helps investors to avoid barriers, and to take advantage of host country incentives. Egger and Winner (2006) evaluated the relationship between corruption and FDI inflows in 73 developed, and developing countries from 1995 till 1999. The empirical results showed that corruption can stimulate FDI inflows; as it helps entrepreneurs to avoid cumbersome regulations and administrative procedures. They argued that corruption can promote efficiency by allowing entrepreneurs to correct or eliminate government failures. In addition, Lui (1985) applied the queuing model to illustrate that corruption can help businesses avoid the consequences of ineffective policies, and the results showed that bribes for officials can generate an incentive to speed up the administrative process. Bayley (1966) argued that corruption can overcome a bureaucracy by improving institutional quality, and can aid businesses to avoid public policy barriers that impede their business’ operations and thereby help them find positive and appropriate solutions.

2.3. Other macro-economic determinants

One of the first influential studies on FDI could be traced back to the work of Behrman (1960). In a survey of 72 U.S. companies that have extensive foreign operations, they discovered that FDI promotes development not only in terms of capital, but also managerial and technical skills. With a dynamic model, Findlay (1978) proved that the diffusion of technology increases the rate of technical progress in a relatively “less developed” region, then enhances the FDI attractiveness. These previous findings imply that countries with rapid development attract more FDI. However, the relationship between FDI inflow and economic growth may not be clear and could be different between developed and developing countries.

The literature agrees that the effect of FDI on growth is a win-win strategy for both foreign investors and host countries (Vukov, 2020). There are different reasons that encourage multinational companies to invest in other countries. In particular, they want to penetrate larger markets to take advantage of the low-cost production or the priorities in exploitation rights of resources. So, multinational companies are established from FDI flows, they try to take benefits of low material cost and low labor cost (Zhang, 2001). Therefore, FDI flows may be different in countries depending on the level of economic growth.

Most previous studies agree that the relationship between FDI and economic growth, in most cases, is found as a feedback loop. Usually, the economic growth creates a capital shortage so the host countries try to attract significantly more FDI resources through advantage terms and incentive policies. Ganić and Hrnjic (2019); Samir and Mefteh (2020) provide evidence that the economic environment play important role in attracting FDI. In another direction, the economic growth also increases per capita income which is an important factor of the market size. This is one of important factor attract FDI flows. Nevertheless, the effects of this relation possess frictions when both of them cannot fully absorb the advantages. Dunning (2013) documented that resources are transferred internally within multinational companies thus FDI may not increase economic growth. Furthermore, the capacity of host countries defies the efficiency of FDI (Beugelsdijk et al., 2008; Borensztein et al., 1998; Schneider, 2005). Therefore, the prior studies argue that developing countries may not attract FDI well due to low efficiency of FDI (Wijeweera et al., 2010).

Once identified the relationship between FDI and economic growth, the previous studies explore further the key drivers of this interaction. According to Nunnenkamp (2002) and Blonigen (2005), there are two main group factors that affect FDI inflows: market and/or resources (conventional factors), and efficiency (cost). Market factors include economic growth, population growth, tax incentives … while efficiency (cost) factors include inflation, tax cost … However, there is still no consensus on the classification of these two groups of factors in the literature.

Current literature on determinants of FDI inflows keeps researching on whether there is a shift between two group factors. In the recent papers, Mottaleb and Kalirajan (2010) and Kumari and Sharma (2018) found that market factors such as market size and trade openness are still dominant factors. In addition, several studies also investigate the role of institutional factors such as corruption and political stability in attracting FDI inflows (Asiedu, 2006; Bailey, 2018; Busse & Hefeker, 2007; Du et al., 2008; Janeba, 2002; Sabir et al., 2019). These studies provide some important evidence about macro-economic factors affecting FDI inflow in both developed and developing countries but the results are mixed.

With respect to efficiency research, human capital and wage cost are found significantly crucial from the view of foreign investors. Noorbakhsh et al. (2001) and Braconier et al. (2005) found evidence that less-skilled labor cost attracts more FDI, with human capital being one of its most critical drivers. Fan et al. (2018) provided evidence that the minimum wage policy in China increased the outflow of FDI. Etim et al. (2019) found that effective tax policy increases FDI flow in a country. In addition to this, the tax burden phenomenon also interests many scholars. However, the effects of the tax on FDI is heterogeneous due to the characteristics of host countries (Andersen et al., 2018; Bellak & Leibrecht, 2009; Bénassy-Quéré et al., 2005; Coelho, 2011; Hines, 1996; Hsu et al., 2019; De Mooij & Ederveen, 2003; Tian, 2018). In recent years, the concerns about price transferring and tax-base erosion have made it more complex to evaluate the effects of the tax burden. Bradley (2015) argued that price transferring increases tension between foreign investors and tax authorities. Moreover, Choi et al. (2017) concluded that the internal price transfer encourages firms to employ this technique at the expense of rival firms’ profits. As a result, the tax burden could be a minor issue for foreign investors compared with other factors.

Although previous studies found a lot of factors that affect FDI inflow and FDI attractiveness, the results are still mixed. There are very few studies investigating the determinants of FDI in ASEAN countries. Unlike previous studies, corruption is not used as a proxy for institutional quality anymore. The capacity to attract FDI inflow is now determined by the proxy variables of economic freedom (quality of economic institution) and political stability (quality of political institution). Taking into account all previous considerations, it is imperative to shed more light on the determinants of FDI inflows in developing countries over a longer time frame. It is proposed by this paper that conventional factors of host countries be seriously considered as the key drivers of FDI inflow.

3. Data and methodology

3.1. Research data

In order to explore the key determinants of FDI inflows, the ASEAN-7 (Cambodia, Indonesia, Laos, Malaysia, Philippines, Thailand and Vietnam) countries were chosen for consideration. Timor-Leste, Myanmar and Singapore were excluded from the study cohort. Timor-Leste is missing a lot of data. Myanmar’s data is not available and Singapore is classified as an outlier being a developed country. The period for study is limited to 1996–2019 as institutional indicators in these countries were generally reported or exploited only beginning in 1996.

Data for this study was extracted from several sources, among them, the World Development Indicators (WDI), The Heritage Foundation think tank (Heritage), the International Monetary Fund (IMF), the Worldwide Governance Indicators (WGI), and the Asian Development Bank (ADB). The description of variables used in this study is presented in Table 1.

Table 1. Variables description

Variables	Measurement	Source
FDI	Logarit of the annual net FDI inflows	WDI, IMF
FRE	Index of economic freedom	Heritage
GDPpc	Logarit of real GDP per capita	WDI
INF	Consumer price index	WDI
POL	Political stability	WGI
POPg	Annual population growth rate	WDI
TAX	Tax revenue as a percentage of GDP as proxy of tax burden	ADB, WDI

Table 2 contains the summary statistics of variables used on the empirical model. After missing observations and outliers were excluded, the data, as presented herein, comprises 154 country-year observations from the seven developing countries in the ASEAN region. The big differences between country-year observations are found in inflation (INF), population growth (POPg), and tax proportion in GDP. The mean of FDI inflow in ASEAN-7 is lower than that of the OPEC countries as the latter is treated as resource-less rich economies (Elheddad, 2018). The mean of GDPpc is 3.19 indicating that most ASEAN countries are developing economies to which FDI inflow is very important.

Table 2. Summary statistics

	FDI	GDPpc	INF	POPg	POL	FRE	TAX
Mean	9.297	3.192	6.943	1.477	−0.448	56.814	13.762
Standard Error	0.060	0.035	1.085	0.042	0.058	0.682	0.287
Standard Deviation	0.746	0.437	13.464	0.522	0.718	8.465	3.557
Kurtosis	0.849	−0.890	49.382	0.312	−0.995	0.207	0.095
Skewness	−1.061	0.006	6.529	0.066	−0.453	−0.688	0.246
Minimum	6.649	2.395	−1.710	0.252	−2.118	33.500	5.798
Maximum	10.400	4.049	125.272	3.021	0.552	73.800	22.401
Observations	154	154	154	154	154	154	154

3.2. Methodology

Williamson (1993) argued that the theory of FDI is based on analyses of the transaction cost. According to the theory, business organizations perform activities based on transaction costs. The theory of transaction cost is concerned with the costs of running a business in the country versus the costs in foreign markets (Williamson, 2007). Dunning (1980) developed the Ownership-Location-Culture (OLI) model to evaluate FDI inflows based on transaction theory. The model developed an assumption of multinational enterprises’ activities determined by three main factors: merits of ownership, merits of locations, and advantages of cultural acquisition.

The OLI model covers a vast range of social and economic costs (Driffield et al., 2013). Eden and Miller (2004) argued that foreign investors might face some risks of discrimination in the host country associated with institutional issues. The paper was based on the transaction cost theory and the OLI model to develop determinants on attracting FDI inflows. We addressed two critical elements of institutional quality, including economic freedom and political stability (Peng et al., 2008). Therefore, we developed an econometric model applied for panel data as follows:

(1) $FD{I}_{it}=\alpha +\beta X_{it}+{\mu }_t+{\rho }_i+{\epsilon }_{it}$(1)

Where the subscripts i,t represent country and time-period, respectively; y is the dependent variable of interest, that is, FDI inflows, X is a set of time and country varying explanatory variables, and control variables; β is the vector of coefficients to be estimated; µ_t is an unobserved time-specific effect; ${\rho }_i$ Is the unobserved country-specific effect and is the error term. We estimate Equation 1(1) $FD{I}_{it}=\alpha +\beta X_{it}+{\mu }_t+{\rho }_i+{\epsilon }_{it}$(1) with two methods: OLS for pool regression and Feasible Generalized Least Squares (FGLS) for Seemingly Unrelated Regression to figure out the parameters. Suspect with these results, we then employ two Lagrange Multiplier Tests: Breusch-Pagan and Honda for balanced panels. The test results strongly suggest that the estimation should be deployed with panel and individual effects.

Many previous studies argue that fixed-effect method is more appropriate for panel data than OLS method, we, therefore, apply the fixed-effect estimation method for equation 1(1) $FD{I}_{it}=\alpha +\beta X_{it}+{\mu }_t+{\rho }_i+{\epsilon }_{it}$(1) as a robustness test. In addition, we also apply the System GMM estimation method to treat potential endogeneity problems due to both unobservable heterogeneity deriving from missing variables and simultaneity which was applied in literature (Nguyen & Dang 2020; Nguyen 2021b).

We present in the next section our estimation approaches to explore the relationship between FDI inflows and proposed determinants.

4. Empirical results and discussion

We utilized the Pearson correlation matrix to detect the preliminary relationships between variables. Results reported in Table 3 suggest that there is evidence of correlations, but it needs more inspections. The maximum value of coefficients is 0.7 (between FDI and GDP), indicating that multicollinearity is not a concern in our models. The Pearson correlation measures are highly unreliable indicators of the relationships among many variables. Hence, we carried out the tests for our main hypotheses using a multiple regression framework.

Table 3. Pearson correlation matrix

	FDI	GDPpc	INF	POPg	POL	FRE	TAX
FDI	1.00
GDPpc	0.70***	1.00
INF	−0.29**	−0.34***	1.00
POPg	−0.38***	−0.24*	0.08	1.00
POL	0.03	0.11	−0.07	0.01	1.00
FRE	0.53***	0.68***	−0.39***	0.01	−0.1	1.00
TAX	0.49***	0.32***	−0.14	−0.38***	0.39***	0.04	1.00

Note. *p < 0.1; **p < 0.05; ***p < 0.01

To test the stationary problem of the variables, we address three different approaches of Levin-Lin-Chu, Harris-Tzavalis, Fisher. The results show that all of variables are stationary at the level.

Table 5 presents the estimation results of Equation 1(1) $FD{I}_{it}=\alpha +\beta X_{it}+{\mu }_t+{\rho }_i+{\epsilon }_{it}$(1) by applying Pool-OLS and Feasible Generalized Least Squares (FGLS) for Seemingly Unrelated Regression. Results obtained from Table 4 suggest that all the variables, with the exception of inflation, are significantly correlated to FDI inflows. However, the signs of population and variables associated with political stability are intuitively puzzling. Specifically, we found out that the coefficients on GDPpc, TAX and FRE are positive and statically significant with FDI, while the coefficients on POPg and POL are negative and statically significant with FDI. These results indicate that economic growth, tax revenue and economic freedom are positively associated with FDI inflow, but population growth and political stability are negatively associated with FDI inflow. Meanwhile, there is no evidence of any relationship between inflation and FDI inflow. Our findings are consistent with certain previous studies (Abdouli & Hammami, 2020; Dkhili & Dhiab, 2018; Sirag et al., 2018). Regarding population growth and political stability, our findings are inconsistent with previous studies (Mitra & Abedin, 2021; Rashid et al., 2017). These show a positive relationship between population growth, political stability and FDI inflow, since population explosion may harm economic growth in developing countries (Peterson, 2017). In addition, political instability adversely affects growth in lower and middle-income countries (Uddin et al., 2017). Our findings further reveal that the impact of economic factors on FDI inflow in developing countries differs from developed countries.

Table 4. Stationary tests

Variables	Levin-Lin-ChuT-statistic	Harris-TzavalisZ-statistic	Fisherchi-squared
FDI	−6.0553***	−5.2447***	37.3021 ***
GDPpc	−6.0717***	−0.0396	5.6275**
POPg	−11.2222***	3.2557	61.5799***
POL	−5.8482**	0.9033	12.0298**
TAX	−8.0642***	−1.7805**	13.9973*
INF	−8.3056***	−7.3922***	64.0335***
FRE	−4.9239*	0.9238	18.5912*

Note. *p < 0.1; **p < 0.05; ***p < 0.01

Table 5. OLS and FGLS estimations

	Pooled		SUR
	Co-eff	Std.err	Co-eff	Std.err
GDPpc	0.704***	0.132	0.826***	0.116
POPg	−0.228***	0.081	−0.212***	0.066
POL	−0.120**	0.059	−0.041**	0.047
TAX	0.069***	0.013	0.050***	0.009
INF	−0.001	0.003	−0.002	0.002
FRE	0.020***	0.007	0.012***	0.004
Constant	5.240***	0.398	5.620***	0.404
Observations	154		154
R2	0.633		0.277
Adjusted R^2	0.618		−0.010
F Statistic (df = 6; 147)	42.226***

Note. *p < 0.1; **p < 0.05; ***p < 0.01

As a robustness test, we applied the fixed-effect estimation method for equation 1(1) $FD{I}_{it}=\alpha +\beta X_{it}+{\mu }_t+{\rho }_i+{\epsilon }_{it}$(1) , and the results were reported in Table 6. Firstly, we started the base model (model 1) with three variables, before tactically adding one more variable in models 2, 3, and 4. We observe that market factors like GDP per capita significantly attract FDI, which is a confirmation that one of the crucial targets of foreign investors is the local market. Nonetheless, population growth negatively affects FDI inflow, while also inhibiting FDI attractiveness. The coefficients on TAX remain positive and significant, with FDI in regression 2 and 3. These results in Table 6 are consistent with the results in Table 5.

Table 6. Fixed-effects and GMM estimation results

	(1)	(2)	(3)	(4)	(5)
GDPpc	1.460***	1.427***	1.498***	1.426***	1.444*
	(0.130)	(0.131)	(0.135)	(0.142)	(1.023)
POPg	−0.197*	−0.214*	−0.225**	−0.201*	−2.816*
	(0.114)	(0.114)	(0.113)	(0.114)	(1.233)
POL	−0.144**	−0.129**	−0.135**	−0.124**	−3.861**
	(0.056)	(0.057)	(0.056)	(0.056)	(1.421)
TAX		0.025*	0.024*	0.020	0.378***
		(0.014)	(0.014)	(0.014)	(0.071)
INF			0.004*	0.004**	−0.031
			(0.002)	(0.002)	(0.018)
FRE				0.011	0.109***
				(0.007)	(0.010)
Observations	154	154	154	154	154
R2	0.601	0.610	0.619	0.625
Adjusted R^2	0.576	0.582	0.590	0.594
F Statistic	72.25***	55.83***	46.18***	39.24***
Hansen J (p-value)					0.521
AR(2) (p-value)					0.832

Note. Regressions 1–4 report the estimation results by applying Fixed effect method. Regression 5 reports the estimation results by applying System GMM method. *p < 0.1; **p < 0.05; ***p < 0.01

Table 7. Sub-period estimations

	(1996–2008)		(2009–2019)
GDPpc	1.698***	(0.285)	2.182***	(0.624)
POPg	0.311**	(0.137)	0.419	(0.459)
POL	0.067	(0.079)	−0.188	(0.237)
TAX	−0.007	(0.021)	0.002	(0.037)
INF	0.006**	(0.002)	0.014	(0.017)
FRE	0.009	(0.009)	0.052**	(0.022)
Observations	91		63
R2	0.367		0.403
Adjusted R^2	0.270		0.259
F Statistic	7.535*** (df = 6; 78)		5.619*** (df = 6; 50)

Note. *p < 0.1; **p < 0.05; ***p < 0.01

Considering macroeconomic factors, there is a positive correlation between inflation and FDI, but the effect is marginal. The coefficient on FRE is positive, but insignificant with FDI. These results support the hypothesis that economic growth and purchasing power of the local market are more important than other factors.With respect to institutional factors, we discovered that political stability has a negative effect on FDI at 5% level. The interesting aspect of this finding is that tax burden does not have any significant negative effect on FDI inflows. This result contrasts with many previous theoretical and empirical studies which argue that tax burdens often discourage foreign investments. On the contrary, developing countries tend to adjust tax rates to attract FDI, because of the benefits brought about by FDI enterprises. Besides, FDI inflows depend on the classification of taxes. Tanzi (2017) pointed out that the impact of tax incentive policies on FDI differs between countries and sectors. This result supports the empirical evidence of the ambiguous relationship between tax burdens and FDI (Bellak & Leibrecht, 2009; Hartman, 1984; Hsu et al., 2019; Tian, 2018). Hartman (1984) argued that for some specific sectors, especially key sectors, a number of foreign investors are not under pressure from the tax burden of the host country. Overall, the estimation results are not much different from our first results and therefore support our hypothesis.

For further investigation, we split the sample into two periods: 1996–2008 (between the two crises, 1997 and 2007), and 2009–2017 (after the crisis). The results documented in Table 7 confirm that GDP per capita is a key driver of FDI, positively significant at 1% level. Other key determinants are more pronounced during the between periods. We only find that population growth and inflation play an important role in attracting FDI for the period 1997–2007. In contrast, economic freedom only plays an important role in attracting FDI after the 2008 crisis period. This may be because foreign firms more and more consider investing to countries with a high level of economic freedom to reduce their risk. More importantly, tax burden has a minor and insignificant impact on FDI, consistent with the results obtained in the previous estimation.

The factors were found to have a different role in attracting FDI between before and after each crisis indicating that the factors affecting FDI inflow may change through the economic circle. Therefore, developing countries need to have flexible policies in each stage of the economic cycle to attract FDI effectively.

5. Conclusion

In this paper, we examine the determinants of FDI inflows in developing countries. With panel data of ASEAN-7 countries during the period 1996–2019, we found strong evidence to support the hypothesis that market factors such as economic growth, tax burden, economic freedom, and inflation are key drivers to attract FDI. Political stability and population growth are also important factors as foreign investors usually have a long-term perspective in their investment decisions. In developing countries, political stability and population growth are negatively related to FDI inflow. This study also finds that the role of these factors differs from the period. Therefore, developing countries need to have flexible policies in each stage of the economic cycle to attract FDI effectively. The poor governance, lack of transparency in financial control systems (including taxation and accounting systems), and lack of government accountability are opportunities for corruption. In fact, corruption seems to have a little negative impact on economic growth when the institutional quality is poor, and it even causes a positive effect on FDI in countries with high economic freedom. However, the structural reform aimed at controlling corruption should be a priority in long-term strategic planning. This will give confidence to the government as well as provide a healthier economy for investors. Therefore, ASEAN countries should focus on improving the quality of institutions to strengthen the business environment. Then, FDI inflows into these countries will also be more stable because it avoids the unofficial costs of corruption caused by poor institutional quality.

Regarding tax policy, there are basically two views on tax incentives: (i) Western developed countries often do not apply tax incentives because they argue that tax incentives cause policy distortions and are easy to lead to exploitation; (ii) Developing countries including most of ASEAN countries use many tax incentives because they believe that tax incentives have the effect of attracting and encouraging investment in accordance with the socio-economic development strategy. The empirical results also show that tax revenue has a positive impact on attracting FDI inflows. Thus, it is thought that ASEAN countries should choose industries for tax incentives to attract FDI inflows into that field rather than cut tax rates because tax rates are already very low. As tax rate and paid-tax amount are at the micro-level, obtaining these data are very difficult, if not saying that impossible at firm levels. Moreover, there are some other factors that may affect FDI inflows, future researches if can overcome this barrier can contribute much to the current literature on the FDI-tax nexus or investigate more determinants of FDI inflows.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research is funded by University of Economics Ho Chi Minh City, Vietnam

Notes on contributors

Van Cuong Dang

Van Cuong Dang is a senior lecturer at School of Public Finance, UEH University. He received his PhD in Public Finance from UEH University. His main research interests could be found in banking, policy and public finance.

Quang Khai Nguyen

Quang Khai Nguyen is a researcher at School of Banking, University of Economics Ho Chi Minh City. His research covers a variety of topics related to financial institutions and empirical corporate finance and banking, including capital structure, corporate governance, risk-taking behaviors, earnings management, bank stability.