The location determinants of Chinese FDI in African
countries: empirical analysis of institutional and financial
factors.
University of Amsterdam
Faculty Economics and business
MSc Business Administration---International Management First Supervisor: Dr. Vittoria G. Scalera
Name: Fang Xia
Student number: 1186526 Date: 27-01-2017
Statement of originality
This document is written by Student Fang Xia who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Despite the fact that China’s outward FDI to Africa enjoys a steady increase in recent years, the distribution or location of these investment is uneven among African countries. This paper will examine whether factors that determine FDI location choices of Chinese MNEs into sub-regional level African countries differently. Using a data sample of 18 low-income and 28 high-income African countries covering 2003 to 2013, we find that the FDI location considerations for Chinese investors differ based on income level of African countries. Our results suggesting that better infrastructure has a significantly positive effect on FDI to low and high income level African countries, while no significant relationship between political instability and Chinese FDI. In addition, We also find that factors high level of economic freedom and openness to trade will attract more Chinese FDI into Low-income African countries, while the decrease in corruption and trade openness may promote Chinese investors to do business in high-income African countries. In general, these results imply that Chinses MNEs can not weight all factors equally, Although they are making investment in the same region.
TABLE OF CONTENTS
1. Introduction... 1
2. Foreign Direct Investment to African : Stylized facts...3
3. Literature Review...9
3.1 Natural resources ...11
3.2 Political risks in host country... 11
3.3 Level of corruption in the host country...14
3.4 Infrastructure development in host country...15
3.5 Economic freedom in host country... 16
3.6 Trade openness in host country...16
4. Hypotheses development... 17
4.1 Political environment... 17
4.1.1 Political instability... 18
4.1.2 Corruption...18
4.2 Infrastructure development ...20
4.3 Trade openness of host country ...21
4.4 Economic freedom ... 22
5. Data and methods...23
5.1 Variable measurement...23
5.2 Estimation method...26
6. Results...27
6.1 Descriptions of the data ...27
6.2 Correlations analysis... 28
6.3 Regression method... 30
6.4 Empirical results...32
7. Conclusion and Limitation...33
7.1 Academic relevance ... 35
7.2 Mangerial implication... 36
7.3 Policy implication...37
References... 39
LIST OF FIGURES
Figure 1: China’s FDI outflows, 1990-2009...2
Figure 2: FDI inflows in Africa, 2000-2009...6
Figure 3: Africa's Real GDP Ggrowth rates, 2000-2014...7
Figure 4: Chinese FDI outflows to Africa, 1999-2013...8
Figure 5: GDP per capita (current USD)...5
LIST OF TABLES
Table 1: China's OFDI by regions, 2005-2011...4Table 2: Distribution of inward FDI flows, by regions (Share in world total %)...8
Table 3: Data and sources...25
Table 4: Descriptive statistics for the sample...28
Table 5: Differences between high and low income level African countries (mean of independent variables)...28
Table 6: Correlation matrix ...29
Table 7: Collinearity Statistics...29
1. Introduction
It is widely acknowledged that FDI from developing countries has increased dramatically recently. According to World invest report in 2009, Outward FDI from developing economies has increased shapely, accounting for almost 16% of global outward FDI, up 13% from 2007 (UNCTAD, 2009), particularly to China. Actually, from 1979 to 2000, the main purpose of China’s so-called “Bringing in” strategy is to attract foreign investment, which increased China's inward FDI rapidly. Nevertheless, the size of outward FDI before 2000 was relatively small. Cai (1999), using UNCTAD’s FDI statistics, reported that the country’s annual FDI outflow began to increase from nearly 0 in 1979, when China’s economic policy of opening to the outside world was implemented and promoted, to US$628 million in 1985, and to US$913 million in 1991, before shooting up to US$4 billion in 1992. However, with the development of general global economy, China’s economy saw a massive growth especially after entering to the World Trade Organization in 2001 and it is quickly emerging as the world’s fourth importer and third biggest exporter in 2003, respectively (WTO 2004). Moreover, the followed implementation of strong economic reform policies carried out by government promotes the expansion of Chinese investment activities with other countries around the world. For example, “Going out” global strategy formally proposed and brought out by the Chinese government in 2000 encourage Chinese local enterprises to participant into the process of economic globalization and accelerate the pace of foreign direct investment in recent years. According to data from 2002 to 2012 (Figure.1), outward FDI from China surged from USD 2.7 billion to USD 87.8 billion, a compound annual growth rate of 41.6%, making China the world’s third-largest outward investor (World Investment Report. 2006). Therefore, after a 20-year-reform and opening-up, promoted by these incentives, China’s foreign direct investment not only has grown remarkably and significantly but also has expanded year by year. Accounting for the statistics data from World Investment Report 2005, up to the end of 2004, the stock of China’s inward FDI is USD 245.467 billion, whereas the stock of China’s outward FDI is only USD 38.825 billion. The ratio of the stock of China’s outward FDI and inward FDI is only 0.16:1, which shows huge sluggishness between inward and outward FDI (Su 2006). In 2014, China’s outward FDI reached USD 116 billion, while in the same year China’s inward FDI is USD
119.56 billion. It was the first time that inward and outward FDI were close to balance.
Figure 1: China’s FDI outflows, 1990-2009
Source: UNCTAD, FDI/TNC database
Differing from previous papers which study a large general range of determinants of FDI, this paper contributes to the literature in three ways. First, this paper focuses on the Chinese FDI location choice in African countries, given the differences between MNEs from emerging country like China and the firms from developed countries which are largely investigated in prior papers. Second, Africa as recipient countries is scarcely studied in existing literature of FDI determinants. because of the obviously different situation and structures, the lessons or theories drawn from other regions can not be adapted to African countries. Furthermore, the differences not only lie in regions (African and non-African countries) but also exist in a state of development of different African countries (low and high income level African countries). More specifically, some factors may not equally affect the FDI in different income level African countries. Hence, this thesis only analyzes the determinants of location decisions in African countries based on the different income level, which gives insight on the considerations when investors undertake FDI in different African countries. Third, in contrast to other studies that access many factors affecting FDI and generate inconsistent results, this study only considers some aspects of recipient countries, which indicate significant differences existing in African countries.
in low and high-income level African countries, respectively. The analysis performs regression for a cross-section of 46 African countries for 2003-2013, covering both low and income level and the results are represented as follows. First, high-income level African countries on average attract more Chinese FDI than low-high-income level African countries. Subsequently, Improving the development of infrastructure in both high and low income level recipient countries may obviously result in an increase in Chinese FDI to that country. Then, a higher level of openness to trade promotes Chinese FDI to low-income African countries, but negatively inhibit FDI investment made by Chinese investors to high-income African countries. Furthermore, economic freedom is a significant determinant of Chinese FDI to low-income African countries but has no significant impact on Chinese FDI inflows to high-income African countries. However, less perceived corruption in high-income African countries is positively related to the Chinese FDI. In contrast, the negative relationship between corruption level in low-income African countries and Chinese FDI to that country is not significant. In summary, these results suggest that MNEs from China should take different factors into account when they invest in Africa. The aim of the thesis is to build up a conceptual framework discussing the locational determinants of China's FDI to Africa and the thesis is structured as follows: the second section presents an overview of FDI inflows into African in recent years. The third section explains the theoretical background focusing on the institutional perspective along with the traditional economic perspective. Given the inconsistent economic development in the African continent, I want to analyze the determinants of China's FDI to Africa by dividing into two groups (low and lower middle-income African countries; upper middle and high-income African countries) according to their income level. Thus I build a series of hypothesis regarding the factors affecting Chinese FDI into different income level countries in the fourth section. Next, the fifth section shows model, variable selections, data definition, and methodology. Hypothesis testing, statistical analysis, and empirical results are presented in the sixth section and the last section finally offers discussion, conclusion, and recommendation for further analysis.
In this open policy background, China has become the leading country in the developing world. More Chinese enterprises are encouraged to invest abroad. However, China’s FDI capital is concentrating on a few countries and limited sectors. Averaging the amount in recent years, Chinese firms seem to expand into countries that do not fit the standard profile of host countries comparing with the FDI from developed countries. (Ramasamy, Yeung, & Laforet, 2012). Ramasamy, Yeung, and Laforet (2012), after analyzing the volume of investment from China during 1990s-2000s, noticed that there are some radical changes in the geographical distribution of Chinese FDI overseas. They found that FDI from China in the early 1990s mainly crowded and accumulated in Canada, the U.S., and Australia, namely developed economies, and accounted for nearly 40%. But by the end of 2000, the Chinese FDI proportion in these regions had declined to only 10%. On the contrary, since 2005, almost half of China’s FDI was made in developing countries, especially countries in Asia, and showed a growing trend. By 2008, the tendency of China’s outward FDI in countries located in the African continent is continuing to increase, accounting for about 10% in 2008 (table 1). Overall, until recently, China’s direct overseas investment mainly focus on the countries in Asia, Latin America, and Europe. After a systematic analysis of the size and composition of China's overseas FDI during 2003-2012, Cheng and Ma (2012) expected an upward trend in China’s outward FDI and also mentioned the characteristics possessed by Chinese investors, such as state-owned business. Therefore, the research on the location choice is greatly influential for China's investors when making FDI decisions during their business expansion process.
Table 1: China's OFDI by regions, 2005-2011
Year Destination of China's FDI flow ( USD 10 000 )
Asia Latin
America
Europe Africa North America
Oceania
2005 437464 646616 50502 39168 32084 20283
(35.6%) (52.6%) (4.2%) (3.3%) (2.6%) (1.7%)
(43.5%) (48.0%) (3.4%) (2.9%) (1.5%) (0.7%) 2007 1659315 490241 154043 157431 112571 77008 (62.6%) (18.5%) (5.8%) (5.9%) (4.3%) (2.9%) 2008 4354750 367725 87579 549055 36421 195187 (77.9%) (6.6%) (1.6%) (9.8%) (0.6%) (3.5%) 2009 4040759 732790 335272 143887 152193 247998 (71.5%) (12.9%) (5.9%) (2.6%) (2.7%) (4.4%) 2010 4489046 1053827 676019 211199 262144 188896 (65.2%) (15.3%) (9.8%) (3.0%) (3.8%) (2.7%) 2011 4549445 1193582 825108 317314 248132 331823 (60.9%) (15.9%) (11.0%) (4.3%) (3.3%) (4.4%) Source: China Statistics Yearbook, various years
Africa, a continent with the most of developing and less developing countries, due to the improvement of investment environment and the progress of regional economies, should not be ignored and have begun to gradually integrate into the trend of development among the world and take part in the global economic activities. As shown in figure 2, in Africa, there has been a decade of significant growth in FDI inflows despite a decline of FDI inflows to 59% in 2009, rising to a peak of $72 billion, almost twice their 2005 level (UNCTAD, 2009). Although there was a slight decrease since 2009. WIR expected that FDI inflows to Africa could return to a growth tendency in 2016, increasing by an average of 6 per cent to USD 55–60 billion (UNCTAD, 2016). At the same time, due to the coordination and assistant of United Nation and modern international society, the political area in most of the African countries have been improved into a comparatively stable environment, which provides the basic guarantee of the economic activities and lay a strong foundation for investment activities from all countries. For example, multiple international organizations such as the IMF and the World Bank have great influential on FDI by promoting good institutional environment (Villamil and Asiedu, 2003; Hakura and Nsouli, 2003 and Frankel, 2003).
Based on the given data published by World Development Report in 2016 (figure 3), as shown in figure 3, Africa is one of the world’s fastest growing regions, with average GDP growth rates exceeding 4 percent per year since 2000. The average economic speed in Africa area is 5.7% during the period 2003-2010, particularly in 2007 Africa enjoyed its robust and significant economic growth of 6.3 percent. Although with slight fluctuation in following year and with a decrease at 3.7 percent in 2014, Africa will still remain among the fastest-growing regions of the world, much higher than the rest areas in the world as an emerging market (except for the developing Asia since 2007), acting as an attracting choice for foreign investors.
Figure 2:FDI inflows in Africa, 2000-2009
Figure 3: Africa's Real GDP Growth rates, 2000-2014
Source: UNCTAD, 2016 (accessed December 2015); World Bank, 2016
Africa owns 15.73 billion tons of oil, accounting for 8.6% of total reserves in the world, and 13.86 trillion cubic meter gas, accounting for 7.9% of total reserves in the world, due to rich natural resources and sustainable economic growth, Africa has become a popular choice for foreign investors to operate business abroad. Over the past years, African countries have paid attention to improving its investment climate. For example, In 2007, some African countries introduced and engaged in a number of FDI-related policy and institutional reforms at both national and regional levels to positively attract and drive the FDI flows in Africa, making the investment environment more favorable to MNEs (UNCTAD, 2008). Many MNEs, particularly those from developed countries already operating in the region, undertook the largest number of investment activities, which largely focus on the investment in gas and oil industries (table 2).
Meanwhile, due to the growing international of emerging economies, MNEs from fast-growing developing countries have started to significantly increase investment in African countries recently. WIR (2010) reported that, although investors from developed countries still hosted the major part of FDI flows to African countries, the proportion of FDI in Africa from developing countries increased from the average of 18% during 1995-1999 to 21% for the period 2000-2008 (UNCTAD, 2010). Especially, China has become one of the most major foreign players in some African countries after a close look at the given data from WIR (2010). Largely because the
Chinese government has always attempted to build up and consolidate friendly relationship and cooperation with African countries for more than half century, providing the strong political foundation for those Chinese enterprises who tend to do business abroad. Moreover, “China-Africa economic and trade operation has accessed to a comprehensive, the rapid and stable development stage since the establishment of China-Africa Cooperation Forum in 2000” reported by Chinese authorities. As a result, according to the latest research report from Barclay Bank, more than 2000 Chinese companies locating in 45 African countries or area, thus China is gradually one of the most prominent players in FDI in many African countries. Up to 2007, China has invested a total amount of more than 6.6 billion U.S. Dollars, ranging from agriculture to energy industry. From the table, In 2008, Chinese FDI in Africa surged to $54.91 billion, up 132% from 2007 (UNCTAD, 2010).
Table 2: Distribution of inward FDI flows, by regions (Share in world total %)
Source: UNCTAD, 2010
Figure 4: Chinese FDI outflows to Africa, 1999-2013
Source: UNCTAD, FDI/TNC database
Despite the rapid increase in Chinese foreign investment in Africa, a large number of Chinese companies invest in Africa, but their operating conditions are extremely varied, namely, there are many high-yield successful investors in Africa, at the same time, many other investors failed or are at a loss. On the one hand, the possible
Home region
1995-1999 2000-2008
Developed countries 79 72.1
problem is the complex political environment, which put the foreign investors doing business abroad at a disadvantage. These risks have seriously damaged the economic interests of Chinese investment and the stability of the overseas operations in Africa. On the other hand, the huge differences in economic performance between the developing countries and least developing countries in African continent have a significant impact on the inward FDI from foreign investors. After the analysis of the distribution of FDI flows by geographical location, the ratio of FDI inflows to sub-regions of Africa was not distributed evenly since the beginning of investment activities (UNCTAD, 2002-2015). In average, Egypt, South Africa, and Nigeria absorbed the largest percentage of FDI flows to the region and acted as the main largest recipient, despite a steady growth in FDI inflows to African LDCs during the period 2000-2008, Particularly, from 2002-2006, these countries accounted for nearly 55% of region’s inflows (World Bank, 2004b). Analyzing the locational determinants of Chinese FDI in Africa may give some suggestions regarding the difficulties they faced and help Chinese companies perform well in their internationalization process in Africa.
Figure 5: GDP per capita (current USD)
Source: World Bank National Accounts data
3. Literature Review
Although there are some papers focusing on the FDI location choice of Chinese multinationals, these studies mainly analyze the factors affect the FDI choices of Chinese MNEs in the geographically close or culturally similar countries, especially in
neighboring Asian countries, such as East and Southeast Asia (Kang & Jiang, 2011). Additionally, other extant literature investigate the different motivation of Chinese firms investing in advanced economies, mostly for the strategic-seeking purpose (Helpman, Melitz, & Yeaple, 2004; Duanmu & Guney, 2009; Duanmu, 2011) However, little attention was paid to the Chinese investment in Africa, and there is a lack of comprehensive analysis. As shown in the previous introduction (figure 2), because of the rich resource and sustainable economic growth, Africa has attracted a great amount of foreign investment, especially the FDI from China. It is well-known that China exerts a dominant influence in Africa, where is neither geographically or culturally close to China nor have advanced technology. Africa enjoys a lot of specific characteristics.
The traditional economic model originally proposed by Dunning (1998, 1993) has become one of the popular theoretical approaches for the researchers to study the motivation of international activities of MNEs, which give a comprehensive study associated with the ownership, location, and the internalization. Although the eclectic paradigm about the foreign investment activities was considered to mainly originate from developed economy multinationals, the results from some extant studies on the empirical evidence confirm that the conventional FDI theories are also applicable to analyzing the determinants of FDI from emerging countries (Kang & Jiang, 2011). From the conventional theory, the foreign firms’ FDI activities are motivated by four major forces, namely resource-seeking, market-seeking, efficiency-seeking and strategic-seeking (Dunning, 1998, 1993). Studies by Buckley et al. (2007, 2008) on the determinants of Chinese outward FDI show that these four primary factors are also positively associated with the FDI locational choice of Chinese MNEs.
Realizing that some extant literature taking little consideration on the normative dimension and part of regulative dimension of institutional environment, and that the lack of an agreed conclusion on the relationship between the political considerations and the preference of location decision, which mean that further investigating about cognitive aspect and the extending research on regulative institutional factors should be taken into account. Therefore, the goal of this paper is to investigate the locational determinants of Chinese firm's FDI, with particular emphasis on the factors which either are rarely discussed (cognitive institutional dimension) or yield mixed results in
extant papers. In this section, I will discuss and provide some insight on key factors that have been generally used in the previous FDI literature by different scholars as shown below.
3.1Natural resources
Acquiring and gaining access to the natural resource is one of the strategic considerations when MNEs are assessing the location choice (Dunning, 1993). Research undertaken by Kang and Jiang (2001) found that the availability of richness natural resources endorsement in eastern and southeastern Asian countries positively promotes the Chinese FDI flows to that particular location, suggesting that natural resources are an important determinant of Chinese FDI flows. Also, Jakob and Cheung (2011), testing the approved FDI data from China Statistic Yearbook in various years, agreed with this statement and pointed out that seeking natural resources is one major driver behind China’s outward investment in Africa especially significant in FDI data from 2003-2007, and additionally affirmed that China’s driver for African natural resources is likely to be a recent but lasting phenomenon because of the implementation of the “ going-out global ” policy. Apart from the natural resource seeking motivation, attaining large host country markets and attempting to avoid the export restrictions are found to be important in attracting FDI and FDI location choices, which is known as market-seeking motivation. It is well accepted that a promising domestic market in host country indicates more potential consumption. Moreover, studies find that there is a positive relationship between the choice of Chinese firm's FDI location and the market size which is measured by using urban population size and market growth in the host economy (Kang & Jiang, 2011). For example, the choice of a Chinese firm's FDI location is positively related to the market size of the host country, namely the larger the market size as measured by host country GDP and GDP growth, the more likely an attractive location the host country is.
3.2 Political risks in host country
In the case of China’s FDI in Africa, applying the conventional theories of traditional economic factors alone, however, can just explain partially how Chinese MNE's choice their FDI location in Africa because of the lack of the considerations about institutional
context in both the home and the host country. Drivers relevant to the institutional forces may also shape the locational determinants of Chinese overseas FDI. Distinctive institutional factors are likely to change the operating cost of doing business overseas or reduce (increase) uncertainty related to outward investments, furthermore attracting (deterring) the location choice of MNEs (Henisz & Swaminathan, 2008; Grosse & Trevino, 2005). Thus, institutions are broadly viewed at least as important as the traditional economic factors in the choice of MNE’s investment location. To lend support to this perspective, the study by Campos, Nauro, and Kinoshita (2003) showed that one of the main determinants of FDI is the institutional factors after the examining the data for 25 emerging economies from 1990 to 1998. Similarly, in a comparative study of outward FDI determinants in China and India, Zheng (2009) concluded that better institutions are considered to be increasingly influential in location decisions, especially for the multinationals from China. Scholars, such as Wight, Fliatotchev, Hoskisson, and Peng (2005), even agreed that, in some case, the institutional theory is the most useful approach to investigating FDI behavior from emerging economies and institutional factors will act as more significant drivers than traditional economic factors in determining FDI location choice.
In the past few years, the institutional approach is the most using theory to study how institutional variables affect MNEs' FDI behavior, particularly its role in firm’s location decision or business strategy choices. Following North’s (1990) definition, “institutions are humanly established the rules of the game, incorporating both formal institutions, such as laws, and informal institutions, traditions, and norms, which structure interactions and thus profoundly affect the social stability and orders.” Building on the comprehensive perspective, Scott (2001) recognized regulative, normative and cognitive channels in his study as three primary aspects of the institutional environment in both providing a foundation for legitimacy and influencing the firm's business behavior. In the case of China, the significant impact of the Chinese government has attracted attention and been recognized as the most striking feature of the FDI by Chinese MNEs. Recent studies have suggested that the institutions from home country provided strong support to the foreign expansion of Chinese MNEs, encouraging the outward FDI behavior of these firms. More specifically, Buckley et al. (2007) explained in the thesis that institutional forces in China and in host countries
are likely to alter the FDI location choice of Chinese MNEs.
From the normative institutional perspective, culture forms the basic foundation for the informal institutions, emphasizing the impact of the social values and obligatory norms. Therefore, cultural distance is recognized as the measurement when researchers study the extent to which normative institutional forces may influence the Chinese firm's FDI. For example, Kang and Jiang (2011), testing the panel data of Chinese outward FDI to some eastern and southeastern countries during thirteen years, found that Chinese MNEs are more likely to locate their FDI operations in countries where cultural differences between China are relatively smaller. Also, the study by Bhardwaj, Diets, & Beamish (2007); and Quer, Claver, & Rienda (2011) examined the influence of host country culture on the location choices of foreign investing firms and find the same negative relationship between cultural distance and the choice of a firm’s FDI location as Kang and Jiang (2011) concluded previously.
Apart from the above normative dimension, the regulative institutions are also found to be other influential factors that investing firms may evaluate when they decide whether or not expand their business in a particular foreign location. With regard to regulative institutional perspective, a growing amount of literature has focused on the political and legal regime dimension, together with the economic regime. Political considerations, such as political risks or government policies, were examined in the study of determinants of foreign investment. Political risks mean the instability and uncertainty of the political system which would pose a threat to the market stability and economic conditions. Therefore, the paper developed by Pak and Perk (2004), measuring the data from IMD and World Economic Forum, confirmed a negative association of political risks with the FDI location choices, suggesting that multinational firms are less likely to invest in countries that suffer high political risk, given that the companies will be more reluctant to locate their operations in an uncertain and unguaranteed business environment. After carrying out researches base on conventional theory, in consequence, a large number of scholars reached an agreement and pointed out that the low political risk of the host country is conducive to attracting investment flows from MNEs (Brouthers, 2002; Duanmu, 2011; Duanmu & Guney, 2009; Schneider & Frey, 1985).
conclusion. Based on the empirical results of the research’s data, Kandiero and Chitiga (2006) estimated political instability variables representing political risks and surprisingly found that the results were statistically insignificant in influencing the decisions made by MNEs about their FDI location choice. Thus, what the role of political risk plays in the deciding location choice for multinationals is not obvious. For example, pointed out by some different scholars, a low level of political instability is likely to be responsible for promoting the investment flows (Kariuki, 2015). In addition, some papers even provided supportive evidence that Chinese MNEs have a positive attitude and tendency to expand in countries with higher level of risks, probably because Chinese MNEs were able to receive the strong political or financial support from the Chinese government during their expansion process.
3.3 Level of corruption in the host country
Another factors that affect FDI is corruption in the host country, which is defined as the misuse of the power by government staff. Theoretically, the previous surveys by scholars largely came to an conclusion that a high level of corruption within the political system deters foreign investment (Kandiero and Chitiga, 2006; Wei, 2000; Busse et al., 1996). Furthermore, Kandiero and Chitiga (2006) provided evidence that the effect of corruption is statistically significant, with an increase in the corruption index by 1 percent level will increase FDI to GDP by 60 per cent. Because perceived high level of corruption represents the unhealth or immature institutional environment, which would sap investors’ confidence in the system of government capacity to deal with FDI and directly discourage the investment decision behavior. In general, the level of corruption is viewed highly related to a attractiveness of institutional investment environment in host countries (Asiedu and Villamil, 2000). On the other hand, Left (1964), in contrast, concluded that corruption may have some positive effects. Since building and maintaining human relations with government officials might contribute to reduce the uncertainty of artificial factors to some extent, from a social perspective, corruption can foster investment. Additionally, assessing short term and long term influence of corruption of a sample of 73 developed and developing economies, Egger and Winner (2005) confirmed a clear positive relationship between corruption and foreign investment, the higher level of corruption, the more attractive to
the foreign investors.
3.4 Infrastructure development in host country
Furthermore, Government policies and political goals are relating variables incorporated in political considerations. On the one hand, Incentives such as good and reliable infrastructure provided by host country government can attract the potential foreign investors. government policy changes, which lead to the high level of trade barriers for foreign investors, or more restrictions on foreign business behavior, may negatively associate with the likelihood of being invested in (Baniak. Cukrowski, & Herczynski, 2003). On the other hand, the political embeddedness, such as FDI project, in a home country may act as motivations for the domestic investors to do business overseas and invest more in other nations. Lunding (2006) indicated that the Chinese authorities support the expansion of overseas investment in Africa countries by the enforcement of Chinese foreign aid programs, such as providing much-need transport and communications infrastructure. Moreover, some researchers also considered the home country political goals attached to MNEs (especially the state-owned enterprises) may impact the location decisions (Ramasamy, Yeung, & Laforet, 2012; Voss, Buckley, & Cross, 2009; Morck et al., 2008). A large proportion of the Chinese outward FDI has been allocated directly to the countries with China has close political relationships (Quer, Claver & Rienda, 2011). Furthermore, the political goals attached to the multinationals are likely to play a role in its location decisions. Apart from the political goals, for the host countries, excellent policies aiming at improving current infrastructure and construction provides location advantages on reducing business costs associated with un.pleasant investment environment. Specifically, the study proposed by Asiedu (2015) confirmed that African countries which are commonly considered have a small market size or in a shortage of natural resources can obtain FDI by improving their investment framework (for example, policy environment and condition of infrastructure). In consequence, existing papers acknowledged the benefit of political considerations changes, infrastructure development policy, but indicates mix results and failed to come up with a consistent conclusion.
3.5 Economic freedom in host country
Although the economic regime is also incorporated in the regulative institutional dimension, there is little discussion on the effect of this factors on FDI location decisions. Economic freedom is regarded as positively related to the national economic growth, thus possibly promoting the foreign direct investment and helping transfer technology and skills. Several studies have aimed at whether the high level of economic freedom would result in the increased FDI in host countries. However, using the regression model to test a sample of developing country on the role of economic freedom, Kapuria-Foreman (2007) pointed out in the paper that FDI is less responsive to changes in levels of economic freedom, which is the contrary to the expected results.
3.6 Trade openness in host country
A Paper by Kang and Jiang (2011) on the FDI location choice of Chinese MNEs highlighted the cognitive dimension of institutions, which constitute the nature of reality and the frames of the social environment (Scott, 2001). Bilateral trade between two countries refers to the degree of the openness to trade which may present the influence of FDI. For example, Osakwe and Dupasquier (2006) emphasized the need for more trade and investment relations between Africa and Asia. Analyzing panel data about 38 developing countries, Kandiero and Chitiga (2006) concluded that FDI has a positive correlation with the degree of trade openness. This is also supported by Lunding (2006) that constituting and strengthening the bilateral trade relationships encourage the domestic firms to locate FDI operations in countries with openness to trade. Also, according to a statistical analysis for 122 developing countries during the period, 1970 to 2000, Büthe and Milner (2008) affirmed that the country would receive a higher amount of FDI after becoming a member of some trade organizations. However, Asiedu (2002) then suggested that the impact of trade openness can not be generalized, since the distinctive types of foreign investment, such as market-seeking and nonmarket-seeking FDI, and the country samples also matter in impacting investors’ behavior.
4. Hypothesis development
In this section, I organize hypothesis of this thesis with the respect of cognitive aspect and the extending aspects of regulative institution, since I realized that some extant literature taking little consideration on the normative dimension and part of regulative dimension of institutional environment and that the lack of an agreed conclusion on the relationship between the political considerations and the preference of location decision.
This section presents the hypothesis regarding the determinants of Chinese FDI into African countries. First, in order to determine the factors that influence Chinese FDI to African countries, it is necessary to divide African countries into two subgroups according to the income level which is indicated by Word Bank. The huge differences in economic conditions and growth among low-income countries and high-income countries impact the investor's behavior. For example, Market-seeking may be more major objectives of Chinese FDI to high-income host countries compared to low-income countries because of the higher purchasing power in high-low-income countries. Hence, domestic demand factors in low-income countries are less likely to be significantly relevant to Chinese FDI.
As described above, In order to control for the market size, I decide to select and collect data on low-income African countries and high income African countries based on the Word Bank Atlas Method 2016, low and lower middle income countries with a gross national income per capita under $4035, and high and upper middle income countries with a GNI per capita more than $4036. With control of the income level, then I pose hypotheses below.
4.1 Political environment
Political risks reflect the political climate and consist of different dimensions. The two commonly accepted and considered are political instability and corruption of the government officials in a country, with political instability to indicate the possibility of the change of current government and corruption to assess the degree of misuse of political power among public officials in the current government system.
4.1.1 Political instability
Many studies indicated that political instability is related to institutional uncertainty and thus may create an uncertain economic environment for the foreign investors, further reducing economic growth and investment opportunities in that country (Asiedu, 2015; Zheng, 2009; Dupasquier & Osakwe, 2006). Besides, Loree and Guisinger (1995) addressed that less political risk is expected to increase more foreign investments in the host country because of a lower risk premium, reducing the uncertainty of potential investors when foreign investors committing their funds. Also, better economic and institutional environment may represent a lower level of information asymmetries, which can increase investment in the host country (Busse & Hefeker 2007).
In contrast, some studies regarding the impact of political instability on the FDI location choice seem to support the negative relationship, there is empirical analysis finding a positive relationship, claiming that Chinese MNEs prefer to invest in riskier countries in order to catch up with MNEs from developed countries (Voss, Buckley et al., 2008). Also, Kang and Jiang (2011) tended to report that Chinese firms are likely to locate their FDI activities in countries with high levels of risk due to the adaption of experience gained from a similar political environment in their home country.
Since China has experienced decades of reforms in economic and institution in attempt to increase China's economy toward market efficiency, many overseas activities conducted by MNES are in fact led and funded by state government, or at least get strong support from domestic institutions, which result in Chinese FDI to be less risk averse (Quer, Claver, & Rienda, 2011). Chinese firms might be more likely to be attracted by the unexpected and unexploited opportunities presented by high-risk countries. For example, “ Angola, an oil-exporting country, has experienced a large in FDI despite its perceived unstable political environment ” reported by Asiedu (2000; 2002). As a result, I propose that:
HYPOTHESIS 1A: In low-income level African countries, Political instability is positively associated with the location decision of Chinese FDI.
HYPOTHESIS 1B: In high-income level African countries, political instability is negatively associated with the location decision of Chinese FDI.
Another aspect of the political risks which should be taken into account is the level of corruption among the government system in recipient countries, which directly reflects the quality and transparency of the host country’s institutional environment. Accordingly, corruption can be viewed as either a grabbing hand, to deter FDI, or a helping hand, to stimulate inflows in host countries. (Bardhan, 1997; Leff, 1964). Previous scholars, after performing empirical analysis of a cross-country dataset, have mainly reached an agreement that an increase in the corruption level negatively affects the long-term inflows of foreign direct investment to that particular country, because corruption in the host country can create operational inefficiencies and increase the operating costs of foreign companies (Wei, 2000; Habib & Zurawicki, 2002). In the short term, firms need to pay bribes or special payments in order to enter the markets or get access to some resources, which result in distortion of the cost structure and reduction of profits generated through investments, thus discouraging investor to invest in that country. Moreover, Lambsdorff (2003) and Rose-Ackermann (1999) also provided support to high corruption in government may imply the weak governance, reducing the productivity of public inputs, thus MNEs have to bear the additional risks, which decreases a host country's attractiveness for foreign investors (Boycko, Diets & Vishny, 1995).
On the other hand, some MNEs might regard these additional payments as speed money, obtaining the legal permission for their investment projects (Lui, 1985). Analyzing four different types of bribes, Glass and Wu (2002) made a significant but contrasting contribution to the general equilibrium effects of corruption on FDI and concluded that corruption may foster foreign investment in that country. Following the “helping hand” conclusions demonstrated by Glass and Wu (2002), Egger and Winner (2015), using data set of both developed and less developed countries, also confirmed the clear positive long and short run effects of corruption on a host country’s locational attractiveness. Corruption can be beneficial in circumventing regulatory and administrative controls. In contrast, the study by Ackay (2001) suggested that corruption has become less and less important for FDI decisions in general speaking, especially for the FDI to less developed countries, because corruption help foreign investors to gain prominence as the contacts. Moreover, a country with high level of corruption may not provide equal and even market access to all foreign investors which would lead to imperfect and inefficient competitive market.
However, The factors that determine FDI to African countries are different from the factors that attract FDI in other developing regions (Batra, Kaufmann & Stone, 2003). Nigeria, which is perceived the high level of corruption, seems to attract large absolute FDI inflows. Chinese firms prefer to invest in a country that is perceived to be corrupt because of the experience of operating business in the similar environment and of dealing with the corruptive government officials. Thus, I propose that:
Hypothesis 2A: In low-income level African countries, corruption is positively related to the location of Chinese FDI.
Hypothesis 2B: In high-income level African countries, corruption is negatively related to the location of Chinese FDI.
4.2 Infrastructure development
The availability of quality infrastructure in the recipient countries, such as telephones, roads, and highways, helps to increase the efficiency of production, reduce the operating and transaction costs of setting up a local business faced by foreign investors and thereby attract more FDI in that country. Loree and Guisinger (1995) showed that the access to either the transport or communications infrastructures in host countries positively promote foreign investment. Accordingly, Morrisset (200) and Asiedu (2002) also got the same conclusion that adequate well-developed infrastructures increase locational advantages in that country. since with good quality of physical infrastructure in the host country, it will become more effective to communicate with their potential clients and get feedback more quickly about their business activities from consumers in the host country, thus lowering the information costs, making their operating efficient, and performing better. Thus, foreign investors are more likely to locate investment in these countries (Bevan, Estrin & Meyer, 2004). Analyzing the relationship between the progress of business environment and FDI into 29 African countries, Morisset (2000) reported that some African countries tend to attract more FDI than other African countries with more natural resources and large domestic markets by reaching a better level of business operating environment.
to be a significant variable for FDI in Africa after Onyeiwu and Shrestha (2004) analyzed a dataset for 29 African countries. They pointed out that the effects of other determinants, such as market size, openness and the access to natural resources, may more overweight the beneficial effect of infrastructure. Besides, the finding from Asiedu (2002) showed the level of infrastructure has little impact on FDI flows into SSA countries. One of two plausible reasons is that FDI to SSA countries is driven by natural resource, mainly in extractive industries. For example, Nigeria, on the largest recipients of FDI, is widely acknowledged to receive a lot of foreign investments in recent years (e.g., in 2006 Nigeria ranked first among all African countries in receiving FDI inflows) despite its poor infrastructure. Another explanation provided by Asiedu (2002) is that the lack of access to some basic infrastructures, such as telephones and internet, is not very related to natural resource-based FDI.
Currently, China keeps providing awarding aid in much-needed infrastructures, such as roads and telecommunications, in order to positively affect Chinese firms' location decisions in African countries. Due to the contradictory findings, I, therefore, propose that:
HYPOTHESIS 3: Infrastructure development in both high-income level and low-income level African countries is positively associated with the Chinese FDI inflows.
4.3 Trade openness of the host country
Openness to trade of the host country has also been considered significant in influencing the FDI flows into a country. Trade openness is often related to the policy of the government, reflecting the flexibility of foreign investors to move capital in and out of a host country (Chakrabarti, 2001). Several studies support the notion that economies with a greater degree of trade openness may stimulate inward inflows into that country made by foreign investors, compared to the countries with few liberal policies, because of the decreased transaction costs (Morisset, 2000; Edwards, 1990). Greater trade openness helps some African countries to catch up with the rest of developing countries in FDI attractiveness (Kandiero & Chitiga, 2006). However, Trade openness of the economy also refers to the ease with which multinationals can export their manufactured goods and primary products in commodities and services
(Kandiero & Chitiga, 2006). There is a well-accepted consensus that lower purchasing power is a characteristic of low-income countries. As discussed in many previous papers, high-income level countries indicate higher GNI per capita, and thus stronger purchasing power in the host country, which represent a huge local market demand (Chakrabarti, 2001; Anyanwu, 1998; Kang & Jiang, 2011). Besides, it is also important to take into consideration that the different types of FDI in African countries, such as the market-seeking investment and non-market seeking FDI (Asiedu, 2002). In high-income countries, FDI is driven by the large host market demand because the output of the foreign investors is produced and sold in the local market. For this purpose, Multinationals operate businesses in high-income level countries may be driven by the relatively high degree of openness to trade because of the difficulties for competitors to export to the host domestic market. In contrast, the main purpose of non-market seeking FDI is not the market demand in host countries but the flexibility of firms to export their products. Hence, foreign investors invested in low-income countries may be attracted by the lower level of trade openness (restrictive policies or capital controls) (Busse & Hefeker, 2007; Asiedu, 2002). For the particular purpose of studying the impact of openness in different income level countries on Chinese FDI. My paper follows the research Asiedu (2002). Therefore, I hypothesize that:
HYPOTHESIS 4A: There is a negative relationship between Chinese FDI to high-income level African countries and openness to trade.
HYPOTHESIS 4B: There is a positive relationship between Chinese FDI to low-income African countries and trade openness.
4.4 Economic freedom
Economic freedom captures the overall domestic investment climate in recipient countries. Economic freedom is defined as "the absence of state intervention, government coercion or constraint on the production, distribution, or consumption of goods and services beyond the extent necessary for citizens to protect and maintain liberty itself" (Heritage Foundation/Wall Street Journal .P.50). Thus, Economic freedom generally presents the degree of which a country is turning to a free market. Prior empirical research about the effect of economic freedom in host countries is a
general consensus that economic freedom has beneficial effects on FDI inflows into the host nation, thus increasing economic growth. (Onyeiwu & Shrestha,2004; Bengoa & Sanchez-Robles, 2002; Quazi & Rahim, 2004). The study by Quazi and Rahim (2004) suggested that the increasing level of economic freedom favors investments by contributing to the access of the host country. Following the previous studies, I propose that:
HYPOTHESIS 5: The higher level of economic freedom in both low and high income level African countries positively affect the Chinses FDI.
5. Data and methods
5.1 Variable measurement
Dependent variable: I employ the FDI stock from China in each African countries as the dependent variable. The reason why I do not use FDI inflows to these countries is because the stock variable can measure the location distribution of FDI more accurately than inflows variable according to the paper developed by Filippaios, Papanastassiou and Pearce (2003). Data were taken from annual Statistical Bulletin of China’s Outward Foreign Direct Investment (Volumes, 2004-2014).
Independent variables: World Bank defines political instability variable as "reflecting the risk of the current government being destabilized or overthrown. Following Asiedu (2006) and Busse & Hefeker (2007), I use political stability ranking index to measure political risks of the host country, which can be collected from International Country Risk Guide published by PRS group.
In accordance with the index selected by Egger and Winner (2005), I obtain data which is annually reported by Transparency International (TI, 2004-2014) to test the effect of corruption. The corruption perceptions index ranges from 0 to 100, where 0 is the highest level of corruption and 100 corresponds no corruption among public officials. Corruption perception index is a composite index, including and integrating information from other institutions, such as the World Bank and the International Country Risk Guide, but it mainly focuses on the pure corruption. Hence, we use CPI
to proxy the degree of corruption.
With respect to the progress in infrastructure, as is discussed in Asiedu (2002), I apply the number of telephone main lines per 1000 population, which are derived from the World Bank, to measure infrastructure development. Following previous literature (Asiedu, 2002; Kandiero & Chitiga, 2006; Kariuki, 2015), I adopt the widely used the radio of imports and exports as a share of a country's GDP to proxy openness to trade. For the measurement of economic freedom, I employ the integrated Economic Freedom index of these factors to measure a country's overall level of economic freedom, which can broadly show the institutional environment of the economy. The index of Economic Freedom is available for a large number of sample countries according to a scoring system by Heritage/WSJ. Besides, this index is annually published by Heritage Foundation and Wall Street Journal officially. The Economic Freedom index is constructed by analyzing four broad categories ranging from rule of law, limited government, regulatory efficiency to open market, including series independent factors. The index is ranked between o and 100, with a score of 100 signifying the most freedom, while the score of 0 indicating the least freedom.
Control variables: Based on prior work, some determinants other than independent variables discussed above have the effect on the FDI inflows, which are measured as control variables in this paper, consisting natural resource availability and GDP growth rate.
Natural Resource Availability: First, according to the previous theoretical literature on determinants of FDI. The availability of local natural resources, especially the raw material and energy sources in African countries, has been well accepted to attract more FDI from China. According to the study, Kandiero and Chitiga (2006) used the share of fuel, ores and metal exports in total exports to measure the host country's natural resource availability. Besides, Buckley et al. (2007) employed the ratio of ore and metal exports to merchandise exports in the host country as a proxy for the host country's natural resource endowment. Based on these studies using different variables as measurement for the host country's availability of natural resource, I would use the share of minerals and oil exports in total exports as a measurement of natural resources, mainly because having ready access to this energy resources may increase
the per capita availability of natural resource in China, which is important for production process. This data is accessible in World Bank.
GDP Growth Rate: Apart from the natural resource availability, the second control variable that has impact on FDI location decision in the host country and included in this model is GDP growth, which indicates the potential market demand. The economic growth of African countries provides a potential incentive and profitable opportunities for foreign investment activities (Buckley et al. 2007). According to Carkovic and Levine (2005), high GDP growth rate may show a high level of return on investment in the host country and, thus, get more investment. Additionally, nearly every study on FDI has proven the positive relationship between GDP growth rate and FDI. Therefore, I include the real growth rate of GDP to control the market growth and potential. This control variable is expected to be positively associated with the FDI location choices in both low-income and high-income African countries. Similarly, the series data of GDP growth during 2003-2013 can be taken from World Bank.
Overall, I list the measurement of all variables and the source of data as follows in table 3.
Table 3: Data and source
Variables Data Source Measurement-Index
Dependent variable FDI inflows
Annual Statistical Bulletin of China’s Outward Foreign
Direct Investment
FDI inflows
Independent variable Political risk environment
International Country Risk Guide (PRS).
Political instability rating Corruption rating Independent variable
Infrastructure development
World Bank
The number of telephone per 1000 capita Independent variable
Economic freedom
Heritage Foundation and
Independent variable Trade openness
World Bank
World Development report
Exports + Imports/ GDP
Control variable Natural resource
availability
World Bank Ores exports +metal exports/ total exports
Control variable GDP growth rate
World Bank Real GDP growth rate
5.2 Estimation method
Having presented the theoretical structure and the statement of hypothesis, I assume a linear relationship, thus I develop the following regression model to estimate.
FDIit=
α+β1(POLIINS)it+β2(CORRUPTION)it+β3(INFRASTRUCTURE)it+β4(OPENNESS)i
t+β5(ECONOMIC)it+γ1（NATURAL RESOURCE）it+ γ2(GDPGRW RATE)it+ε
Where，
FDIit= Chinese FDI stock in country i in year t.
POLITICAL INSTABAILITYit= government instability in country i in year t CORRUPTIONit= corruption index in country i in year t
INFRASTRUCTUREit= phones per 1000 population in country i in year t OPENNESSit= imports+exports/GDP in country i in year t
ECONOMICit= economic freedom index of country i in year t
NATURAL RESOURCEit= metal and oil exports as a percentage of total exports in country i in year t
GDPGRW RATEit= GDP growth rate of country i in year t
The sample for this model consists of 46 African countries over the period 2003-2013, including 18 high-income level African countries and 28 low-income level African countries, which is based on the Word Bank Atlas Method 2016. The observation is 517. The data on China’s FDI stock in each African countries are collected from Annual Statistical Bullet ion of China’s Outward Foreign Direct Investment (Volumes
2004-2014). The data source for most of the variables, such as trade openness, availability of infrastructure, GDP growth rate and natural resources availability is the Word Bank. Additionally, Transparency International (TI, 2003-2013) provide rating index of perceived corruption existing in public officials, and data for political instability is proxied by the Political risk rating which drawn from International Country Risk Guide published by PRS group. Data on economic freedom are obtained from the Index of Economic Freedom published by Heritage Foundation and Wall Street Journal.
I perform Ordinary Least Squares method to estimate the equation, mainly because our sample is a cross-sectional data set, and pooled OLS model is efficient to avoid interference of the time nature of panel data.
6. Results
6.1 Descriptions of the data
Table 4 reports the summary statistics of the data that are included in the model, while Table 5 compares the variables for low-income level and high-income level African countries. According to table 4, the average FDI stock from China to high-income level African countries is higher than the FDI stock in low-income level African countries. Additionally, the average number of mobile phone subscribers per 100 people in high-income level African countries is more than double that of low-income level African countries. However, based on Table 4, the average score of economic freedom index shows little differences between these two income level countries.
Table 4: Descriptive statistics for the sample
Table 5: Differences between high and low income level African countries (mean of independent variables)
Variables
High income level African countries
Low income level African countries
FDI stock 323.5807107 170.7009494
Trade openness 102.6043367 74.7178947
Mobile phone subscribers, per
100 people 90.9227778 39.2652117
Economic freedom, overall
index 60.7925530 54.3387620 Political stability 0.1584343 -0.6072414 Corruption perceptions -Transparency International 45.0312500 29.4485294 GDP growth 5.1964772 4.8318173 Natural resource 48.0400409 35.4537814
6.2 Correlations analysis
First, I begin my analysis by doing the correlation analysis in order to check how independent variables are related to each other. If the independent variables are strongly correlated, Multicollinearity may occur, thus making the effect of explanatory
Variables Minimum Maximum Mean DeviationStd.
FDI stock 0.0100004775.070000229.40916
2 544.083566 Trade openness 22.130000224.71000085.649380 35.826742 Mobile phone subscribers, per
100 people 0.670000194.51000059.519069 45.568410
Economic freedom, overall index 21.000000 77.00000056.789899 8.484584 Political stability -2.690000 1.540000 -0.314004 0.926902 Corruption perceptions
-Transparency International 11.000000 77.00000035.896552 13.484095
GDP growth -62.075920104.486790 4.973114 7.056858
variables on the dependent variables less accurate. Furthermore, VIF (variation inflation factor) also is needed to test whether the multicollinearity exists between independent variables. The results are reported in table 6. The variation inflation factor is reported when using the regression analysis, and the results are presented in table 7. According to Craney and Sueles (2002), if VIF>5 or VIF>10 are presented, there is an existence of multicollinearity in the model. VIF and tolerance in our model are provided in table 7 below.
Table 6: correlation matrix
Variables 1 2 3 4 5 6 7 8 FDI stock 1 Trade openness -.144** 1 . Mobile phone subscribers, per 100 people .261** .370** 1 Economic freedom index .018 .069 .349** 1 Political stability -.112* .457** .226** .394** 1 Corruption perceptions -.038 .471** .536** .682** .625** 1 GDP growth -.052 -.024 -.061 .008 .068 .025 1 Natural .222** -.131** .109* -.157** -.210** -.153** .113* 1
Variables Tolerance VIF Natural resources availability
(control) 1.176 .850
GDP growth rate (control) 1.122 .891
Political stability 1.619 .618
Corruption perceptions index 3.608 .277
Infrastructure 1.877 .533
Trade openness 1.696 .589
Economic freedom index 2.265 .441
6.3 Regression method
Since I split the African countries into two different level groups in the hypothesis, I, therefore, perform four regression equation employed ordinary least square (OLS) method, mainly because the pooled OLS provide a way to ignore specific nature of time and efficient in examining cross-section data. First, I include independent variables associated with the hypotheses in model 1 in order to test the overall effects of independent variables on African countries. Model 2 and model 3 is estimated regarding the two different group separately. Table 6 below reports the results of four models.
Table 8: Results of OLS estimation.
Std. Error are in parentheses; *p<0.05,**p<0.01,***p<0.001
Variable model 1 (full) model 2 (low) model 3 (high)
Natural resources availability (control) .174**
(0.814)
.481*** (0.722)
-.095 (2.005)
GDP growth rate (control) -.056
(7.672) .072 (6.922) -.092 (14.538) Political stability -.030 (41.639) (36.464).014 (87.019).072
Corruption perceptions index -.107
(3.886) (3.798)-.101 -.162 * (8.023) Infrastructure .383*** (0.792) .357 *** (0.744) .375 *** (0.000) Trade openness -.215* (1.120) (0.958).028** -.323***(2.484)
Economic freedom index .066
(0.365) (5.781).153** (12.342).019
Adjusted R2 .162 .332 .163
6.4 Empirical results
According to the results from Column (3) of table 8, the full sample estimation results give little support to the hypotheses. Among the five independent variables, only the infrastructure development variable and openness to trade show significant influence on the FDI location choice in African countries. Results of infrastructure development provide support to hypothesis 3, indicating that Chinese firms are more likely to locate their business in a African countries with high level of infrastructure development because of the lower operating costs and efficiency.
The next two Columns report the estimated coefficients of variables for low and high income level countries. First, Column (3) are the regression results in low income African countries. Three of five independent variables remain significant, indicating that these variables are moderately to highly significant with Chinese FDI in low income level African countries. Furthermore, the coefficients for all the three variables are positive. More specifically, Chinese investors are moderately sensitive to changes in trade openness based on the F-test. A 1% increase in the openness to trade in low income level African countries will lead to a moderate increase of Chinese FDI inflows into these host countries by 0.028 % (p=0.041). Thus, Hypothesis 4B is confirmed. In term of economic freedom, The coefficient of economic freedom is found to be highly significant in model 2, suggesting that a 1% increase in economic freedom leads to a 0.153% increase in Chinese FDI in low income level African countries. Therefore, Hypothesis 5 is supported. Hypothesis 1A and Hypothesis 2A which are associated with domestic political risks and investment climate are not statistically confirmed. Political instability is not statistically significantly (p=0.838) in attracting Chinese FDI into these countries. However, the explanatory variable - corruption perception index-has a negative coefficient despite of its insignificance, rejecting Hypothesis 2A.
In Column (4), regression results for high income level African countries are presented, which are sharply distinguished from those got from full sample of African countries testing and also from the low-income African countries. First, Based on the coefficient of political stability, unlike the insignificant negative relation between FDI inflows and political instability in low income level African countries, in high income
level countries, political instability is positively associated to the FDI inflows from Chinese firms despite its statistically insignificant. Another measurement of political risks is the corruption perception index. this empirical finding suggests that the level of corruption is moderately influencing the FDI from China into high income level African countries, supporting the Hypothesis 2B. Specifically, a 1% decrease in corruption index slightly leads to a 0.162% increase in Chinese FDI into these countries (p=0.109). Hypothesis 4A regarding to the effect of trade openness in high income countries received statistics support, with the coefficient -0.323 and a marginal significant level at 5% (p=0.002). This result indicates that Chinese MNES are more likely to locate in high-income African countries with a lower degree of trade openness. The positive relationship between economic freedom and FDI inflows is partly supported because of the insignificant P value.
As discussed above, the factors that influencing FDI location choice made by Chinese firms are different among sub-regional countries. In general, independent variables like infrastructure development significantly favor FDI into both high and low income level countries, but will attract more FDI to high income level countries than low income level countries.
The results also reflect the impact of control variables. GDP growth rate has a limited role in attracting Chinese FDI to both low and high income level African countries. There are two plausible reasons, one is because African countries are relatively small, in term of population and income, compared to other developing countries such as India and Brazil. Despite the rapid development and high GDP growth rate, the absolute value of the GDP per capita is low, thus the high GDP growth rate in African countries provide little help to attract Chinese FDI. Another possible is that Africa is perceived as overly less developing in general, so investors may regard the rapid GDP growth rate in this region as the bubble or unstable stage. As a consequence, GDP growth rate in African countries exerts little effect on Chinese FDI.
7. Conclusion and Limitation
African countries, using cross-section regression model to examine a sample of 29 low -income and 19 high-income African countries respectively. Although there are many papers focus on the FDI determinants, as pointed out earlier, most of them concentrate on the factors related to regional or global level. based on existing literature which give some insights on the location determinants of FDI, I contribute to FDI literature by emphasizing on Chinese FDI into sub-regional countries and aiming at the factors regarding political climate for investment, institutional environment for investment. Thus the summarized results are presented as follows: (i) the empirical results reject both H1 and H2 which related to the political climate in recipient countries. First, the results suggest that political instability is not significantly affecting Chinese FDI into both low and high income level African countries. One plausible reason is that most of Chinese FDI in Africa is supported and funded by the government, namely, state-own business. More precisely, State-owned Chinese MNEs may have some political goals during their FDI process, thus being less risk averse and not being largely profit-pursing investment activities. (ii) The second hypothesis partial gets supported which is regarding another aspect of political risks known as corruption. More specifically, on the one hand, Corruption existing in high-income African host countries significantly hinders FDI decisions made by Chinese MNEs. This may be because the relatively low operating costs when investors deal with the relationship with officials. One the other hand, our H2A is rejected, suggesting that the empirical relationship between Chinese FDI into low-income African countries and corruption level in that country is unclear, which indicate that corruption has little influence on FDI in those countries. The reason may be that low-income African countries are mainly characterized and considered low wages so that the increased operational cost caused by corruption in these African countries could not affect the FDI decisions made by Chinese MNEs. (iii) The fifth hypothesis regarding the effect of economic freedom obtains a partially empirical support, confirming that economic freedom is a robust and positive determinant of Chinese FDI into low-income African countries. However, the positive impact of economic freedom on Chinese FDI into high-income African countries is not statistically supported. One possible explanation provided by Kapuria-Foreman is that the measurements of economic freedom are highly aggregated in these high-income countries. This is an interesting finding and needs to be studied more in future papers. Apart from economic freedom, trade openness also presents the
