What determines Chinese Foreign Direct Investments in Africa?

UNIVERSITY OF GRONINGEN

FACULTY OF ECONOMICS AND BUSINESS

MASTER THESIS INTERNATIONAL BUSINESS AND MANAGEMENT

What determines Chinese

Foreign Direct Investments

in Africa?

An Institutional Perspective

Peter Csizmadia

A

T

C

Introduction 1

Theoretical Perspectives 2

General Review 2

The Resource-Based View 5

The Institutional-based view 6

The Chinese State and the Multinational Enterprises 8

What Determines Chinese Multinational Enterprises to Invest? 10

Investments in Africa 11 Conceptual Model 13 Data 19 Methodology 21 Results 23 Robustness 25 Conclusion 26

Discussion and Limitations 27

Bibliography 28

1 | P a g e

I

China, one of the fastest growing economies in the world that might soon also become the biggest, has put a lot of pressure on strengthening its global presence lately. As a new global superpower it has started its new “Go Global” strategy that has resulted in growing foreign direct investments. Being mainly a receiver of investments in the past decades China is now close to invest more overseas than it receives.

Chinese outward foreign direct investments started to increase markedly around 2004 and since then it grew twenty-fold. According to the latest publication of Chinese FDI1 statistics, as of 2013, China is the third biggest investor worldwide with the seventh highest outward FDI stock in the world.

However, it seems that China tends to invests more in developing countries with less stable political and institutional environments (Buckley, et al., 2007) (Kaplinsky & Morris, 2009). Also Chinese investments are higher in countries with weak institutions and abundant natural resources, as proposed by Kolstad and Wiig (2012). The question arises, why is that? Why would China invest in countries with weak institutions and less political stability? I believe that China is not looking for weak institutions but as a latecomer player it invests in countries that have a similar institutional environment.

TABLE 1WHERE DID FDI GO IN THE LAST 15 YEARS (UNCTAD)?

The plethora of studies about Chinese outward investments is growing increasingly and many scholars find weak institutions to be significant determinants, but not too many researchers have

1 _{Foreign Direct Investments}

World China Developed Countries 60% 15% Developing Countries 40% 85% 50 100 150 IFDI OFDI

2 | P a g e

focused on the effect of institutions on Chinese FDI in Africa. Also, the approach of the scholars investigating Chinese investments in Africa seems to be flawed. They use a biased sample of countries including only those that have received investments, and excluding the rest. This study will correct this mistake including the countries left off the samples in the current papers.

To the best of my knowledge, there are only three studies investigating Sino-African investments and their determinants (Kolstad & Wiig, 2011) (Cheung, et al., 2011) (Carike, et al., 2012). However all these studies have left countries out from the sample, and this raises questions about the reliability of the results.

Kolstad and Wiig (2011) have measured the effect of institutions and natural resources (and their interaction) on inward FDI from China to Africa. This study attempts to address a similar research question to the aforementioned paper, but in a different manner, including also the omitted countries and with a stressed emphasis on the institutional environment. In conclusion the research question is as follows:

How do institutions affect Chinese Investments in Africa?

T

P

The structure of the theoretical review looks as follows. In the first chapter the general theory of FDI and MNE internalization is reviewed and the main hypotheses are formulated. In the second part the background of Chinese firms is explained and the main findings of the determinants of Chinese investments are detailed. Later the conceptual model and the choice of variables are explained.

G

R

3 | P a g e

The OLI paradigm2 _{is a simple, but also a “profound construct” (Dunning, 2000, p. 163). It}

claims that MNEs foreign activity and their industrial composition are determined by three simple interdependent motives. These are: Ownership Specific Advantages (or in other words,

Firm Specific Advantages, FSAs), Locational Attractions (or Country Specific Advantages,

CSAs) and Internalization Advantages (Rugman & Verbeke, 2001). Ownership advantages (O) refer to the ownership specific advantages of the firm seeking to engage in FDI. The greater the advantages are, the higher the chance that the enterprise will invest overseas. Ownership advantages mainly refers to skills such as entrepreneurial skills, production technique and trademark, all those that are unique to the firm itself, and make it capable to utilize and exploit opportunities to create future goods, services and -in this case- to engage in foreign activities (Shane & Venkataraman, 2000). Location attractions or advantages (L) are the natural or created endowments of the host country (such as natural resources, cheap labor and institutions) that can attract and/or amplify inward foreign investments. Companies are seeking benefits to gain from the countries they are to invest in, and these benefits lie in the unique setup of natural, institutional and labor endowments of the countries. MNEs hope to internalize these advantages so that they can be more competitive than other firms. Internalization (I) is a framework evaluating the alternative ways of the exploitation and creation of the core-competences, “given the locational attractions of countries and regions” (Dunning, 2000, p. 164). It is, in simple words, how much they can take advantage of the CSA and FSA for their own good. “The greater the net benefits of internalizing cross-border intermediate product markets, the more likely a firm will prefer to engage in foreign production itself, rather than license the right to do so, e.g. by a technical service or franchise agreement, to a foreign firm.” (Dunning, 2000, p. 164)

At least one of the previously mentioned interdependent variables is needed for a firm to engage in FDI activity. However, MNEs that only have competitive advantage in ownership will likely chose licensing, and those that have advantage in internalization and ownership will export according to the OLI paradigm theory. (Dunning, 1980)

The motives of MNEs can be further identified by the following drivers according to scholars of the field; Natural Resource-seeking (supply-orientation), Market-seeking (demand-orientation), Efficiency-seeking and Strategic asset-seeking. We can also distinguish between horizontal and vertical FDI. Horizontal FDI is mainly focusing on market seeking penetration (e.g.: being closer to rich customers) whereas, vertical FDI stands for obtaining natural

4 | P a g e

resources, strategic assets or lower wages and taxes. However, these are not completely efficient theories to understand FDI strategies. (Dunning, 2000)

Dunning (2000) has pointed out that different MNEs residing in different countries will not have the same FDI strategies based on the country’s economic history. Countries that have less foreign experience will less likely invest in others, and others having better experiences and historical and economic ties with other countries are more likely to invest. To give an example, in the case of African investments previous colonial ties will highly affect inward investments from the United Kingdom and France (Svedberg, 1981).

As we can see, there are many reasons for a firm to invest abroad, but, even if all the aforementioned interdependent variables (OLI) stand, investing overseas comes with a price. All companies that want to invest overseas will face the cost of doing business abroad, the liability of foreignness and the transaction costs. It means that they not only have to take into consideration their own advantages and the location attractiveness, but also the price it comes with to engage in foreign operation with uncertainty caused by being an outsider. (Eden & Miller, 2004)

There are many theories exist about FDI activities and MNE’s behavior. (Dunning, 2000). However, this thesis does not attempt to review all the theories. Nevertheless, it is important to highlight the transaction cost/internalization approach, the institutional- and the resource-based views.

According to the internalization theory, “MNEs are internalizers of pecuniary externalities due to structural market imperfections. (,,,) When natural market imperfections are high, the expansion of firms across national boundaries may be a more efficient way to internalize these non-pecuniary externalities.” Transaction cost is the cost to organize the interdependencies between the agents. (Hennart, 2009, p. 134)

The cost of doing business abroad represents all additional costs that MNEs face relative to local firms with the same operations and activities. These are activity-based costs such as investments costs, barriers, communication, insurance and liability of foreignness costs such as unfamiliarity, discrimination and relational hazards (political risks). (Chen, 2008)

5 | P a g e

T

R

-B

V

The Porter (1979) five forces model assumes that firms are homogeneous and places little emphasis on the idiosyncrasy and uniqueness of a firm’s attributes. The model assumes that they control and have access to the same set of resources and follow the same strategies. Barney (1991) has complemented the Porter model by introducing the resource-based view. Firms differ by the resources they own and these resources are the source of the competitive advantage. These resources have to be valuable, rare, inimitable and non-substitutable (VRIN). By valuable Barney (1991) refers to the added value for the customers, by rare that it cannot be developed easily, by inimitable that it is only controlled by one firm and by substitutability that there are no substitutes that are accessible by other firms.

A firm that faces growing numbers of competitors (which can be assumed in an emerging country, such as China) will find itself in vulnerable environment for its strategic position, therefore it has to seek new competitiveness, for example in the access to resources that are not available for other firms (Eisenhardt & Schoonhoven, 1996).

Barney (1991) mainly focuses on internal resources as assets of firms, but a unique accessibility or a monopoly of a firm to extract natural resources is also a “valuable” asset to firms and not all enterprises can have access to them, therefore successful resource-seeking can generate inimitable advantages for Chinese firms.

In the context of this master thesis, these resources are natural resources of African economies that will create competitive advantages for those firms that can gain rights of extraction and exploitation of them. Factor-driven economies, such as most African economies3, base their competitiveness on factor endowments such as natural resources and unskilled labor (GEM Report, 2014). In the OLI perspective, these resources are the locational attractions.

Natural resource-seeking is one of the main motives of Chinese firms for overseas investments (Kolstad & Wiig, 2011) (Kolstad & Wiig, 2012) (Cheung, et al., 2011). Therefore, the first hypothesis of the thesis is formed:

H1: Chinese investments are higher in natural resource abundant countries.

6 | P a g e

Kolstad and Wiig (2011) has proposed that Chinese investments are moderated by weak institutions in Africa. The next chapter introduces the institution-based view and explains how it affects investments.

T

I

-

The OLI paradigm was widely accepted for decades, but other models have emerged next to it, such as the institution-based, resource-based and the industry-based view (building up the

strategy tripod). Institutions, in other words, “the rules of the game”, seem to be of utmost

importance. Not only have new theories been developed but also the omnipotence of the eclectic paradigm has been questioned (Peng, et al., 2009).

2.FIGURE INSTITIUTIONS,ORGANI ZATI ONS AND STRATEGI C CHOI CES (PE NG, ET AL.,2009)

North (1990, p. 3) explains institutions as the rules of the game, they are the “humanly devised constraints that shape human interaction” and institutions shape company behaviors as they are embedded into it. Scott (2001) has established the institutional pillars, which are the tacit cognitive, normative and the better codifiable regulatory. The cognitive elements are tacit, and reflect frames, schemas and interpretations, how things are understood and interpreted. The

normative pillar stands for how things should be done in accordance with the country’s values

and norms. The regulatory pillar is how things have to be done. Institutions can provide structure to social interactions and can reduce uncertainty, however, the different institutional setup of countries can also place obstacles against international trade and investments.

7 | P a g e

The aforementioned rules of the game affects multinational firms. However, it seems that it is the instability of the rules that are detrimental for foreign direct investments and firms to engage in foreign operations. A continous stability and predictibility are of utmost importance. The change of regimes, political powers, regulations and the risk of expropriation deters investments (political stability) (Durnev, et al., 2012), but maybe not for China.. The question is “How do organizations play the new game when the new rules are not completely known?” (Peng, 2003, p. 283). It can be assumed that firms that are to invest consider the political and institutional setup of the host countries beforehand; they might look at country reports and political stability, corruption, regulation (etc.) indexes to adapt their strategies and to decide on whether to invest or not. However, institutions themselves do not have an attractive effect, but can moderate investments. Therefore, we can assume causal relations between institutions and foregin direct investments.

According to the Uppsala model (Johanson & Vahlne, 2009) of internationalization firms before engaging in foreign operations (here exporting), first gain experience in the domestic market and they tend to enter similar countries to theirs. Also, internationalization usually starts with exporting (Johanson & Vahlne, 2009). This is in perfect accordance with the findings of Yeung&Liu (2008) and Morck et al. (2008) that firms tend to invest in institutionally similar countries, as they might already have experience and competetiveness in that environment. Firms can better operate in similar settings and environment to their country of origin. Companies and investors that have experience and involvement in bribery tend to look for similalry corrupted environments. (Cheung, et al., 2011).

8 | P a g e

The next hypotheses of the thesis are articulated in connection with these theories, where the second hypothesis adresses the importance of the proximity of the institutions and the third hypothesis is the same as the one in the paper from Kolstad and Wiig (2011) measuring the effect of weak insitutions. Considering these two hypotheses we can compare whether the distance or the “badness” of institutions that matters.

H2: The smaller the institutional difference between China and the recipient the

higher the investments are.

H3: Chinese investments are higher in countries with weak institutions.

Introducing the resource-seeking into the model the fourth hypothesis is as follows (as in Kolstad and Wiig (2011)):

H4: Chinese investments are higher in naturally abundant countries with weak institutions in

Africa

T

C

S

M

E

As China has emerged as an investor country and started publishing detailed OECD compatible data, more and more scholars have started to investigate the topic. Previous articles mainly focused on Chinese inbound FDI and not on outward investments. The table below (Figure 3) shows the growing amount of Chinese OFDI from 2003 to 2012. Having the biggest foreign exchange reserve and a huge trade surplus, China is likely to invest even more in the future and the pattern we can observe on the table will not change. (Cheung & Qian X-W, 2007) Some authors (Yao & Wang, 2012) even suggest that China has already displaced the OFDI of the OECD countries.

China`s OFDI activity before 2004 was quite minute. Most of the investments were driven by political considerations and not by individual firms seeking to invest (Cheung & Qian X-W, 2007). Before 1995, private enterprises were not allowed to invest overseas. However, after

FIGURE 3.CHINE SE OUTWARD INVE STMENTS (UNCTAD)

9 | P a g e

1995, the Chinese government has started assisting firms for foreign investments (Gelb, 2010). In 2002, China`s government has pushed the “go global4_{” strategy to encourage Chinese}

companies to invest overseas, and has started to act more like a supervisor and catalyst of foreign investments, as a reaction to the China’s ongoing accession to the WTO.

After a change in policy in 2004, China has given more freedom to its companies to invest overseas and China`s strategy has transformed from purely “political oriented strategy” to a more “market oriented operation” (Cheung & Qian X-W, 2007, p. 4). The 10th economic plan of the Chinese government has listed OFDI as a key issue to promote Chinese globalization strategies. Furthermore, low-interest loans and risk insurances were introduced for enterprises engaging in foreign operations (Gelb, 2010). As a result, Chinese OFDI has started to increase substantially and has produced enough data to analyze the underlying motives of Chinese MNE`s in overseas investments. This and the fact that China has started reporting UNCTAD compatible FDI data has encouraged scholars to empirically analyze the determinants of Chinese OFDI.

Even after the “Go Global” strategy announced by the 10th _{Plan of the Chinese government,}

investment behavior is still significantly affected by government policies. The Chinese authorities are able to control and allocate OFDI movements in accordance with the will of the state through the approval system. A high percentage of the OFDI was conducted by SOEs (State-owned Enterprise) and also most of the firms are still state-owned. (Zhang & Daly, 2011) Although, “Many ‘SOEs’ function as conduits for private gain, in the sense that profits are appropriated in part by key individuals who are not formal owners of the firms. Similarly, the returns from, and decisions made in many apparently ‘private’ firms are in par a reflection of the direct decision-making power of state bodies, particularly provincial governments. ‘Private’ in China means that the state holds less than 50 per cent of the equity” (Kaplinsky & Morris, 2009, p. 552)

Furthermore, some officials also own companies using the connections they have gained by their official/political position or they have a position because of their entrepreneurial activity (Kaplinsky & Morris, 2009). As we can see, it is hard to draw an exact line between private and non-private firms and also to find out whether is it individual or governmental decision that drives firms to invest. Private and state-controlled firms are highly interconnected and enjoy the safe financial environment the government offers them in the mean of cheap loans and

10 | P a g e

funding. Also, firms that are predominantly owned by the state have to ask permission for overseas investments from the National Development and Reform Commission and the decision is purely political in most cases. (Zhang & Daly, 2011)

State controlled firms can be classified further into central owned and provincial state-owned firms. The central state-state-owned firms operate mostly under state-to-state agreements and the provincially owned are more decentralized. (Kaplinsky & Morris, 2009)

Because of the previously mentioned interconnectedness of the firms and the state, Chinese MNEs can be more long-term oriented and less risk-averse than western, purely capitalistic firms (Kaplinsky & Morris, 2009). A firm that enjoys the support of the state will not act the same as another firm that has to comply with its individual owners and shareholders’ will. Western firms are short-term oriented and more risk-averse because of “the growing emphasis on shareholder value” (Kaplinsky & Morris, 2009, p. 562). So, we can see that Chinese firms have an advantage of sustainability and risk-taking because of the unique setup of the state they are operating in.

W

D

C

M

E

I

?

After having reviewed the basic theories and the idiosyncrasy of Chinese firms, we also have to investigate what empirical studies found about Chinese MNEs.

11 | P a g e

Chang & Ma (2008) also investigated Chinese Outward Investment data between 2003 and 2006 and found out that having a common border (+), distance (-), cultural proximity (+) and

landlocked countries (-) are all significant factors. However, cultural proximity and common

borders are not the case in Africa.

Zhang and Daly (2011) also found positive significance for trade, GDP, GDP growth and

openness to FDI, and a weak significance for natural resources, which holds true generally for

investments.

In conclusion, some general patterns of the FDI literature can be observed in Chinese OFDI. That is that market size (denoted in GDP), growth (in GDP), trade and distance are of importance for Chinese firms. Next to that, it can be seen that Chinese companies are resource-seeking and tend to invest in countries that are accessible by water. That is because Chinese investments are many times preceded or followed by infrastructural investments and China tends to create market hubs by constructing ports and infrastructure for later investments and trade. These investments comes with a high level of Chinese workforce involved. Furthermore, these are the results of inter-state agreements, treaties. (Wong, 2013) (Foster, et al., 2008). And, as proposed by Kolstad and Wiig (2012), natural resource-seeking is moderated by institutional environments, but the research that proves that makes a major flaw as previously mentioned. Except for institutional variables, the findings of Chinese FDI determinants are similar to general FDI literature.

I

A

“Whilst China has a strategy for Africa, Africa lacks a strategy for China”5

Africa might be the core of the next economic boom and the source of cheap labor as Chinese labor cost is increasing (see Ceglowski, (2012)). Africa has the second highest GDP growth rate by region, and according to the CIA Factbook, 7 of the 20 fastest growing countries in the world are African (in 2013). On average, the African economy grows 5-7% annually and it has been projected by the World Bank that this pattern will not change. This is important because GDP growth attracts FDI (Chowdhurry & Mavrotas, 2006), meaning that it is likely that Africa will receive more investments in the future because of growth and access to cheap labor. Also, higher inward investments in the area might foster a drive to improve the political and

12 | P a g e

institutional environment of African countries that would further attract investments. But what determines firms to invest in Africa at the moment?

Goldstein (2004) finds that political instability is a key limitation for inward FDI for Africa. However, the region has potential. The main problems with Africa to be attractive to FDI are the political instability and the lacking infrastructure and human capital stock. (Asiedu, 2005) As China is investing in African infrastructure, it might have a positive effect on the attractiveness of African countries for investors.

Asiedu (2002) has found that openness to trade has less effect in attracting FDI to Africa than it has for other countries, but the importance of natural resources and markets is confirmed.

Naudé and Kruger (2007), controversially, find that resource and market seeking do not affect inward investments, only political stability does. These controversial findings also support to test the moderating effect of institutions on natural resources suggested by Kolstad and Wiig (2012). It is possible that Chinese investments are not triggered by either weak institutions or natural resources, but they invest more in resource abundant countries when they can take advantage of their experiences in “bad institutions” to gain rights to get hand on resources.

Data proves that international investment agreements (treaties) also help investments. (Sichei & Kinyondo, 2012). Sichei and Kinyondo (2012) further found evidence that African countries are more open to and conducive for FDI since 2000, meaning that they have recognized the importance of the inflow of foreign capital.

The natural resource seeking is further supported by Hailu (2010), who finds significance for natural resource, labor quality, trade openness, market accession and infrastructure condition.

Further determinants in the literature are financial development, corruption (+, as a helping hand), religious tension risk and oil economies. (Cattaneo & Ezeoha, 2011) (Ibrahim, et al.)

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Series1 9621 19943 14611 18164 17261 31013 35720 51364 59276 56043 47034 48021 55180 57239 0 10000 20000 30000 40000 50000 60000 70000

13 | P a g e

C

M

As it was mentioned before, the research questions is as follows:

How do institutions affect Chinese Investments in Africa?

However the measurability of these “institutions” is complicated, and can never be perfect by its nature. There are many attempts to do so, but strong criticism also exists in the literature (Gleaser, et al., 2004) (Voigt, 2007), so these variables have to handled with caution. The World Governance Indicators, the most widely cited source of institution variables, measure the Rule

of Law, Political Stability, Corruption, Accountability, and Regulatory Quality. These are

aggregate/construct indicators that combine different surveys from enterprises, experts and citizens. The data measurement is based on 32 different data sources (survey institutes, NGOs6, International organizations, firms and think-thanks…) combined together (WorldBank, 2015). However, these construct variables sometimes include similar or even the same factors and they capture different aspects of the institutional environment7. World Bank8 measures these determinants from -2.5 to +2.5, where higher is better (safer).

Arndt and Oman (2008) have expressed some concerns about the usefulness of the World Governance Indicators. According to them, the WGI indicators lack transparency, underlying theory, comparability over time, actionability and also include hidden bias. However, they also state that these kind of construct variables can never completely capture the truth.

Fortunately, the World Governance Indicators from World Bank are not the only attempt to capture the institutional environment of countries. Data is also available from The Freedomhouse, Polity IV, Political Risk Services Group (PRSG), Transparency International (CPI) and from the Fraser Institute World Economic Freedom Rating.

The Fraser Institute publishes data on manifold different aspects of institutional variables such as Business Regulations, Bureaucracy Costs, Legal System and Property Rights, Legal Enforcements of Contracts, Foreign Investment Restrictions, and the list goes on. The Transparency International coalition surveys the Corruption Perception Index (CPI) among countries.

However, the database from the PRS Group is not available free of charge, and only World Bank and Transparency International have published enough data for all the countries included

6 _{Non-Governmental Institutions}

7 _{That will cause the multicollinearity that is later explained in the research}

14 | P a g e

in the study and for the span of the analysis. Therefore, this study will use only these two databases9_.

To address the research question, the literature has been reviewed and I have established the basic theories. This thesis includes three institutional variables to address the hypotheses, these are corruption, political stability and rule of law. To measure natural resource endowments, oil rents data is chosen from the World Bank. The reason behind the selection of these variables is explained in the forthcoming chapters.

Oil

China is highly endowed with coal and minerals, but has less oil reserves. Also, oil represents a higher added-value than coal, and therefore, transporting it overseas is more profitable. Also, the findings of the scholars investigating Chinese investments find significance only for oil producing countries.

The endowment of natural resources can be measured by their exported value, their contribution to the gross domestic product or also by natural reserves that can later be extracted. Interestingly, natural resources endowment will not mean sustainable superior GDP growth, but can attract foreign investors (Sachs & Warner, 2001).

However, how natural resources scarcity should be measured is subject to debate in the literature. Scarcity can be measured by elasticity of substitution, price, rents, unit costs, and energy costs. Furthermore, some constructed indicators exist as well (such as the energy-based indicator). Scarcity can refer to exchange scarcity (price and rent) and to use scarcity (the ability of the natural resource to generate value). Use scarcity is measured “in terms of the balance between the productivity and availability of the resource base and the level of technology” such as the “general unit cost”. (Cleveland & Stern, 2001, p. 242)

I have chosen rents as rents can capture the profitability of the exploitation of natural resources. Oil rents measure the difference in the value and the production cost of oil.

Including this variable in the first hypothesis it changes as follows:

H1: Chinese investments are higher in countries with higher oil rents.

To capture whether institutions (corruption, political stability and rule of law) moderate resource-seeking, I will use a dummy variable that stands for whether the African country is

15 | P a g e

producing oil or not. Also an interaction term with a dummy variable is easier to interpret than using two continuous variables.

To test these hypotheses I do not use the institutional difference term explained in the last chapter, but the unchanged values of corruption, political stability and rule of law. The reason for this is that Chinese firms might use the bad institutional environment to gain rights for natural resource exploitation. (Kolstad & Wiig, 2012)

In the next chapters the institution variables are explained and the hypotheses are introduced. Corruption

As it has been mentioned before, Cuervo-Cazurra (2006) finds that highly corrupt countries will invest in corrupt countries. Multinational enterprises tend to invest in countries with similar institutional settings as they have developed their competence and gained experience in those environments already. They will better “play the game” in countries close to theirs (Peng & Pleggenkuhle-Miles, 2009) (Cheung, et al., 2011).

Corruption can either have a positive or negative effect on investments. The “helping hand” type of corruptions might foster investments in a country as firms can bribe governments for better possibilities (e.g. vis-à-vis concurrent firms) and to, for example, pay lower taxes. On the other hand, corruption can also be a “grabbing hand” to national firms as well, creating an unequal “playground” for agents (Egger & Winner, 2004). Kaufmann and Vicente (2005) introduces the phrase of legal and illegal corruptions, where legal corruption means that the state favors some legal entities, firms legally by differentiated taxation and tariffs, or on the other hand, disfavors. The first can be achieved by legal lobbying or with good connections and mutual interest and benefits of the agents. Illegal corruption is an undercover/underhand measure of an entity to gain benefits for exemption for a specific reason.

As proposed by Kolstad and Wiig (2012) Chinese natural resource-seeking is moderated by weak institutions. This also supports the choice of corruption as the independent variable of the test, as in highly corrupted environment (in which Chinese firms have experience) it might be easier to gain benefits and rights to exploit natural resources.

16 | P a g e

and not the exact “value” of it. However, I compare these two approaches in this thesis. Including corruption in the hypotheses, they are modified as follows:

H2a: The smaller the difference in corruption between China and the recipient the

higher the investments are.

H3a: Chinese investments are higher in countries with high corruption in Africa.

H4a: Chinese investments are higher in oil producing countries with high corruption

in Africa

Political Stability

The change of regimes, political powers, regulations and the risk of expropriation deters investments (Durnev, et al., 2012). Also, as Chinese investment decisions are strongly interrelated with politics, we can assume that there are political connections between the countries China invest in and China. However Chinese firms might be competent in rapidly changing environment and they can invest where no one else would.

Political Stability captures the likelihood that the political system/government will be destabilized (World Bank). The change of the government in a politically unstable country increases risks for foreign firms. High political risk means high cost of doing business abroad (Eden & Miller, 2004). With high political risk the hazard of expropriation and, confiscatory taxation rises as well (Schnitzer & Kesternich, 2010). Change of government does not explicitly mean that political risk rises, but causes uncertainty and uncertainty deters FDI (Erramilli & D'Souza, 1995).

Including this variable in the hypotheses, they are as follows:

H2b: The smaller the difference in political stability between China and the recipient

the higher the investments are.

H3b: Chinese investments are higher in countries with less political stability in Africa.

H4b: Chinese investments are higher in oil producing countries with less political

17 | P a g e

Rule of Law

To secure investments a high degree of rule of law is indeed of importance, however, maybe Chinese firms, having an experience in country with less “rule of law”, can better maneuver in similar countries.

Rule of Law captures the extent to which firms and individuals abide by the law and rules of society. It also includes contract enforcements, property rights, courts, police and the likelihood of violence and crime (WorldBank, 2015). Weak rule of law puts investment in risk and raise the level of uncertainty.

Again, the following hypotheses are formulated:

H2c: The smaller the difference in rule of law between China and the recipient the

higher the investments are.

H3c: Chinese investments are higher in countries with less rule of law in Africa.

H4c: Chinese investments are higher in oil producing countries with less rule of law in

Africa

Control Variables

The determinants of Chinese outward and African inward investments have been reviewed in the previous chapters. To develop the model for the control variables, I here summarized some general findings as well. Merging the general findings and the determinants of investments reviewed in the last sections, I chose the control variables.

Fortunately, there are a number of meta-analyses (Blonigen, 2005) (Zait, et al., 2014) (Chakrabarti, 2001) of FDI literature that summarize the findings of scholars in the field. These findings have to be dealt with caution, but a general pattern can be drawn for the controlling factors. “…there are no widely accepted set of explanatory variables that can be regarded as the “true” determinants of FDI.” (Zait, et al., 2014, p. 211) And there is no single answer to the question “what determines FDI.” Different countries will show different patterns depending on the receiver and “sender” of the investments as well.

Chakrabarti (2001) has reviewed the literature existing before 2001 and according to his research most of the scholars have found significance for Market Size (+), Tax (-), and Exchange

Rates (-). Although there were less papers reviewed, Openness (+) and Growth Rate (+) also

18 | P a g e

Barriers and Trade Deficit seems to be contradictory as papers find either negative and positive

relations or none. A later study from Blonigen (2005) also vindicates the importance of

Exchange Rates (-), Taxes (-) and Tariffs (-). Although Greenaway et al (2012) find that

exchange rates have small effect on the export participation but higher effect on the volume and that might be the case with FDI as well.

The Size of the Market (measured by GDP or GDP per capita) determines the level of possibilities. A “big” market with high purchasing power of its citizen indicates a good field for investments, however, the picture is not that simple, as the maturity (and entry barriers) of the particular market can be detrimental. The most tempting markets are those that are not penetrated yet, but big in size. Also, some investments are not market-seeking, but, instead resource-seeking investments. Therefore, the size of the market does determine foreign direct investments but not always significantly (there can be exceptions).

Not only the size of the market seems to be important, but also how fast it is growing. Growth can attract FDI because investing in a growing/emerging economy could mean establishing early presence and strengthen positions before the market becomes mature. There is a relation between investments and growth (Guillaumont, et al., 1999), but the literature on the direction of causality is ambiguous. However, some authors (Chowdhurry & Mavrotas, 2006) conclude that it is GDP growth that attracts FDI and not vice versa. I believe that this is a mutual causality, and that these two views are mutually exclusive. Therefore GDP growth indeed attracts FDI, but FDI indirectly or directly helps growth. Furthermore, the effect of FDI on growth is more significant for LDCs than DCs (Blonigen & Wang, 2004). The relation between growth and FDI is amplified by the fact that a high level of IFDI in a country also attracts other investments (Cheng & Kwan, 2000). So, if growth attracts FDI, then the attracted FDI will generate even more IFDI.

As we have learned from the gravity model of trade, physical distance matters for cross-national operations. But distance not only has an effect on trade, but on FDI as well. (Gao, 2009)

Other existing determinants in the literature are ample, Zait et al (2014) have listed more than 30 of them in their literature review.

19 | P a g e

countries of Africa and exchange rates and taxes were also not found significant, specifically for Chinese investments according to the findings. Instead of distance, a variable for Landlocked countries is included to control for the accessibility of the countries. Next to the basic control variables, I included Treaties that are specific for Chinese investments. Wage and infrastructure data is not available for the span (time) of the study unfortunately and the way trade openness is measured causes the problem of multicollinearity. Therefore the conceptual model is as follows:

D

China has started publishing OECD compatible FDI data since 2003 in the Statistical Bulletin of China’s Outward Foreign Direct Investment by the Ministry of Commerce (MOFCOM). Before that, MOFTEC10 has published approved outward foreign direct investment data. With the introduction of OECD compatible data there was a change of methodology in the data gathering process as well. Not only the change in the methodology would make the results biased, but also there was a significant change in Chinese foreign investment policy with the introduction of the Go Global strategy. (Cheung, et al., 2011) This study works with the OECD compatible investment data spanning from 2002 to 2013 of 39 countries, and including the rest

10 _{Ministry of Foreign Trade and Economic Co-Operation} Control Variables

20 | P a g e

11 where there were no investments11_{. Trade is collected from the National Bureau of Statistics}

of China and GDP, GDP growth, Oil rents are collected from the World Bank database. The institutional variables are collected from Transparency International and the World Bank Governance Indicators.

The following table summarizes the descriptive statistics of the variables.

Variable Mean Std. Dev. Min Max Observations

FDI overall 34.72704 230.6285 -814.91 4807.86 N = 500 between 77.48736 0 500.083 n = 50 within 217.4707 -1280.266 4342.504 T = 10 GDP overall 29019.5 62465.17 409 462979.2 N = 500 between 57544.75 748.1632 300472.9 n = 50 within 25499.49 -123420.9 271902.5 T = 10 Trade overall 182033 489198.7 181 5999428 N = 500 between 382313.2 1242.65 2060476 n = 50 overall 309496.6 -1491508 4120986 T = 10 GDP Growth between 5.090512 7.397286 -62.07651 104.4845 N = 500 within 2.359512 -1.346654 10.64284 n = 50 overall 7.018045 -66.23706 100.3239 T = 10 Inflation overall 61.50863 1077.671 -8.97474 24059 N = 500 between 379.9993 1.686127 2694.523 n = 50 within 1009.743 -2633.014 21425.99 T = 10 Political Stability overall -.4768325 .8985419 -2.660021 3.827751 N = 500 between .8271345 -2.261666 1.000315 n = 50 within .3681921 -2.325299 4.011985 T = 10 Corruption overall 2.931254 1.011097 -1.376426 42160 N = 500 between .9330967 1.774046 5.837788 n = 50 within .4090856 -.2732398 6.366344 T = 10 overall -.650238 .6398975 -1.841827 2.392344 N = 500

Rule of Law between .5858647 -1.626792 .946252 n = 50

within .2691135 -1.55307 2.681101 T = 10

As there are extreme values and different measurements in the dataset, FDI, GDP, trade, inflation and GDP growth are all in natural logs (shifted when needed). Also for the residuals to have normal distribution, this transformation is needed. It is also the same method used in the FDI literature.

As there are many negative observations for FDI, normal natural log transformation would not work, therefore the below transformation is used (area sinus hyperbolicus) (Busse & Hefeker, 2007). This transformation does not change the feature of the data (see appendix) but is a good tool to include negative results as well. Many studies simply delete negative FDI and replace them with zeroes. Other studies take average values of years (this method not only vanishes

21 | P a g e

zero values in most cases but also smooths the highly alternating investment data) however there are not enough observations to do so (Busse & Hefeker, 2007).

𝑌 = ln⁡(𝑥 + √(𝑥2_{+ 1))}

The following table summarizes the correlation between the variables. Trade and GDP show a little bit of a higher correlation between the independent variables. However, a little bit of a higher correlation for Trade and GDP is no surprise. As we need to control for market size, the only variable that could be omitted is trade, however, this neither would change the outcome of the tests or the adjusted R2_{. Therefore, we can assume that leaving them both in the model will}

not make the test biased. Corruption, political stability and rule of law seem to correlate, therefore I will not use them in the same regression (different regression will be run for each).

M

The rule of thumb for multiple regressions suggests to use ideally 20 observations per predictor or following Tabachnick and Fidell (2007) N(*T) should be equal to 50 + 8(k). Having 500 observations allow 10 explanatory variables in the model maximum. None of the models use more than 8 predictors, therefore the model is appropriate.

To deal with panel data there are multiple models available, such as the OLS pooled effect, the random, and the fixed effect model. The pooled model ignores the difference between the dimensions, therefore it is an inappropriate approach for the analyses, as we can suspect that Chinese firms would choose countries to invest in for various reasons that cannot be captured by statistics (and by available data). The pooled model assumes that the effect of variables is the same for every country. The fixed effect model however allows one to observe individual

FDI Trade GDP GG Infl Trea Landl Oil Cor Rule Pol

22 | P a g e

heterogeneity and uses different intercepts for the dimensions based on the observed differences embodied by the predictors.

“…the crucial distinction between fixed and random effects is whether the unobserved individual effect embodies elements that are correlated with the regressors in the model, not whether these effects are stochastic or not” (Greene, 2011, p. 183)

The fixed effect model can control for the omitted time-invariant variables in the model. As the within variation for FDI is higher than the between variation, we can assume that there is a reason to use the fixed effect model. Moreover, the fixed effect model with FDI is the general approach in the literature for foreign direct investments.

There are statistical approaches to choose between these models fortunately. Both the Bresuch Pagan Lagrange Multiplier and the Wald test reject the null hypothesis on the 99% significant level, therefore the Pooled OLS model cannot be used (as proposed above) but both fixed and random models seem applicable. The Breusch Pagan test shows that there are random effects present. Therefore, to further choose between the models, the Hausman selection test was used, that also rejected the null hypothesis indicating to select the fixed effect model. Because of presence of heteroskedasticity (modified Wald test), robust standard errors are used. The fixed effect equation is as follows, where 𝜂 is the time-invariant fixed effect estimate, u is the error term,⁡𝛼⁡the intercept. (Clark & Linzer, 2012)

𝑌_𝑖𝑡 = 𝛼 + 𝛽₁𝑋_𝑖𝑡+ 𝜂 + 𝑢_𝑖𝑡

To check whether time-fixed effect is needed in the model, I used the “testparm” command in Stata to see whether the time dummies are all equal to 0 or not. The results showed that there is no need to include a time dummy in the model. Also, including the time dummy does not change the outcome results12_.

There were four main hypotheses proposed in the hypotheses development chapter of this thesis. To test these hypotheses, the following three models were developed.

Dependent Variable Model Variables

Foreign Direct Investment Value

Model 1 Institutional Distance + Controls Model 2 Institution + Controls Model 3 Interaction Term + Controls

23 | P a g e

In equation format these models are the following:

1) 𝑙𝑛𝐹𝐷𝐼_𝑖𝑡 = 𝛼 + 𝛽₁𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑎𝑙_𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒_𝑖𝑡 + 𝛾₁𝑙𝑛𝑀𝑎𝑟𝑘𝑒𝑡𝑆𝑖𝑧𝑒_𝑖𝑡−1+ 𝛾2𝑙𝑛𝑇𝑟𝑎𝑑𝑒𝑖𝑡−1+ 𝛾3𝑙𝑛𝐼𝑛𝑓𝑖𝑡−1+ 𝛾4𝑙𝑛𝐺𝐷𝑃𝐺𝑟𝑜𝑤𝑡ℎ𝑖𝑡−1+ 𝛾5𝑂𝑖𝑙𝑅𝑒𝑛𝑡𝑠𝑖𝑡−1+ 𝛾₆𝑇𝑟𝑒𝑎𝑡𝑦_𝑖𝑡+ 𝛾₆𝐿𝑎𝑛𝑑𝑙𝑜𝑐𝑘𝑒𝑑_𝑖 + 𝑣_𝑖𝑡 2) 𝑙𝑛𝐹𝐷𝐼_𝑖𝑡 = 𝛼 + 𝛽₁𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑠_𝑖𝑡+ 𝛾₁𝑙𝑛𝑀𝑎𝑟𝑘𝑒𝑡𝑆𝑖𝑧𝑒_𝑖𝑡−1+ 𝛾₂𝑙𝑛𝑇𝑟𝑎𝑑𝑒_𝑖𝑡−1+ 𝛾3𝑙𝑛𝐼𝑛𝑓𝑖𝑡−1+ 𝛾4𝑙𝑛𝐺𝐷𝑃𝐺𝑟𝑜𝑤𝑡ℎ𝑖𝑡−1+ 𝛾5𝑂𝑖𝑙𝑅𝑒𝑛𝑡𝑠𝑖𝑡−1+ 𝛾6𝑇𝑟𝑒𝑎𝑡𝑦𝑖𝑡+ 𝛾₆𝐿𝑎𝑛𝑑𝑙𝑜𝑐𝑘𝑒𝑑_𝑖+ 𝑣_𝑖𝑡 3) 𝑙𝑛𝐹𝐷𝐼_𝑖𝑡 = 𝛼 + 𝛽₁𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛_𝑖𝑡+ 𝛽₂(𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛_𝑖𝑡∗ 𝑂𝑖𝑙𝑃𝑟𝑜𝑑𝑢𝑐𝑖𝑛𝑔_𝑖𝑡−1) + 𝛾1𝑙𝑛𝑀𝑎𝑟𝑘𝑒𝑡𝑆𝑖𝑧𝑒𝑖𝑡−1+ 𝛾2𝑙𝑛𝑇𝑟𝑎𝑑𝑒𝑖𝑡−1+ 𝛾3𝑙𝑛𝐼𝑛𝑓𝑖𝑡−1+⁡𝛾4𝑂𝑖𝑙𝑃𝑟𝑜𝑑𝑢𝑐𝑖𝑛𝑔𝑖𝑡−1+ +𝛾₄𝑙𝑛𝐺𝐷𝑃𝐺𝑟𝑜𝑤𝑡ℎ_𝑖𝑡−1+ 𝛾₅𝑂𝑖𝑙𝑅𝑒𝑛𝑡𝑠_𝑖𝑡−1+ 𝛾₆𝑇𝑟𝑒𝑎𝑡𝑦_𝑖𝑡+ 𝛾₆𝐿𝑎𝑛𝑑𝑙𝑜𝑐𝑘𝑒𝑑_𝑡+ 𝑣_𝑖𝑡 Where 𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑎𝑙_𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 is measured as follows:

𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑎𝑙_𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒_𝑖𝑡 = 𝑎𝑏𝑠(𝐶ℎ. 𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑠_𝑖𝑡 − 𝐴𝑓𝑟. 𝐼𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑠_𝑖𝑡)

Smaller values refer to less institutional distance between the host and the recipient countries. This measurement emphasizes countries with closer institution variables.

The first two equations are to test whether institutions matter for Chinese investments and the third to test the moderating effect of institutions on Natural Resources as proposed by Kolstad and Wiig (2011). Except for the institutional variables all variables are lagged with one period, which is one year in this case. GDP, trade and inflation are most likely to affect the succeeding years’ investments and the use of lags is also grounded by literature13_{. On the other hand}

institutions will affect the investments made in the same year.

R

To test the hypotheses of the thesis, the following regression tests were conducted14,15. As both corruption and political stability are not significant, H2a and H2b are not supported, however,

rule of law is significant on the 95% level and therefore, h2c is supported. Chinese investments

are higher in countries with the same level of rule of law.

13 _{However, the model without lags do not change the results significantly.}

14 _{As a representation I have also reported the model with the flawed methodology used in the papers from Kolstad}

and Wiig (2011), Cheung et al. (2011) and Carike et al. (2012), where they have only used those countries that received FDI in the span of the experiment in the appendix

15 _{Furthermore I have also enclosed results with including time-fixed effects, but the change caused by including}

24 | P a g e

FDI Pol_dif Cor_dif Rule_dif

Treaty 0.193 0.193 0.066 (0.477) (0.483) (0.434) lnGDP 1.321 1.463 1.399 (0.368)**** (0.390)**** (0.369)**** lnInflation -0.147 -0.152 -0.135 (0.112)* (0.114)* (0.102)* lnGDPG 0.010 0.005 0.013 (0.015) (0.012) (0.015) lnTrade 0.298 0.253 0.243 (0.160)** (0.163)* (0.159)* Oilrent -0.028 -0.027 -0.032 (0.009)**** (0.009)**** (0.010)**** Inst_dif 0.041 0.008 1.147 (0.297) (0.006)* (0.538)*** _cons -30.952 -34.267 -33.018 (7.206)**** (7.773)**** (7.289)**** R2 0.20 0.21 0.21 N 500 500 500 F 11.64 12.08 13.19 * p<0.2; ** p<0.1; *** p<0.05; **** p<0.01

Standard errors in parentheses

The second table addresses the third hypotheses of the thesis, that Chinese investments are higher in countries with weaker institutions. As higher values are better, none of the hypotheses are supported (H3a, H3b, and H3c). However, we can see an opposite effect of rule of law as it

was proposed by this thesis. Chinese investments are higher when rule of law is less efficient, but elaborating on the second hypotheses, it is more likely that Chinese investments are simply looking for same institutional setups in which they have gained experience already, as the significance is higher when taking the difference in the values.

PolStab Corruption RuleOfLaw

Treaty 0.226 0.259 0.206 (0.491) (0.475) (0.503) lnGDP 1.296 1.263 1.301 (0.373)**** (0.360)**** (0.368)**** lnInflation -0.142 -0.157 -0.125 (0.111) (0.118)* (0.098) lnGDPG 0.009 0.010 0.008 (0.015) (0.015) (0.015) lnTrade 0.305 0.286 0.299 (0.160)** (0.160)** (0.157)** Oilrent -0.029 -0.027 -0.028 (0.008)**** (0.008)**** (0.008)**** Institutions 0.122 0.313 0.278 (0.174) (0.298) (0.158)** _cons -30.389 -30.421 -30.346 (7.314)**** (7.093)**** (7.240)**** R2 0.20 0.21 0.20 N 500 500 500 F 12.95 19.13 11.90 * p<0.2; ** p<0.1; *** p<0.05; **** p<0.01

25 | P a g e

Interestingly, oil rents is significant in all of the regressions but with a negative 𝛽 value, it is robust even if we change the oil rent variable to the oil producing dummy as well. Therefore, the first (H1) hypothesis is also not supported; China is not resource-seeking in Africa.

The next table introduces the interaction term of the institutional variables and the oil producing dummy. The results show that political stability has no moderating effect whatsoever, therefore H4b is not supported. Chinese investments in oil producing countries are indeed higher when

there is less corruption (recall that higher values means less corruption). However, H4b is

supported, Chinese firms tend to invest more in oil abundant countries with less stable rule of law.

FDI Corruption PolStab RuleOfLaw

Treaty 0.186 0.037 0.135 (0.489) (0.466) (0.486) lnGDP 1.080 1.209 1.299 (0.349)**** (0.366)**** (0.327)**** lnInflation -0.005 -0.007 -0.125 (0.008) (0.008) (0.093)* GDPGrowth 0.003 0.002 0.008 (0.011) (0.013) (0.015) lnTrade 0.423 0.378 0.340 (0.166)*** (0.172)*** (0.160)*** Corruption 0.083 (0.201) Polstab 0.106 (0.177) RuleOfLaw 0.321 (0.146)*** i.Oilproducing -10.692 -6.603 -8.204 (4.177)*** (2.441)**** (2.302)**** Interatction Term 1.624 -0.074 -1.612 (0.945)** (0.385) (0.939)** _cons -24.985 -27.002 -28.275 (6.756)**** (7.062)**** (6.314)**** R2 0.26 0.24 0.25 N 500 500 500 F 17.81 13.37 12.75 * p<0.2; ** p<0.1; *** p<0.05; **** p<0.01

Standard errors in parentheses

R

26 | P a g e

When introducing the interaction term, the WGI corruption variable was significant on the 80% level but none of the others were found to be significant at all.

When changing the natural resource indicator to oil exports or to natural resource rents I have found no significance at all again, that further supports the use of oil rents and the dummy variable.

Including a time dummy in the model to control for time-fixed effects, or rerunning the regressions without lagged variables, the results are still the same.

I was also interested whether wars have any effect on Chinese investments and used the Polity IV database to check, however the variable was also not found to be significant.

Some studies only investigates Sub-Saharan countries and excludes North-Africans, differentiating between Arab and non-Arab countries in Africa (and so does the statistical bureau and the Ministry of Foreign Affairs (FMPRC, 2015)). Therefore, I have also included a dummy to test the significance of this distinction, and found no evidence to distinguish between them.

C

27 | P a g e

D

L

This study only investigates Chinese investments in Africa, but a comparative study of other investors (e.g.: United Kingdoms, France, USA) could reveal more details about whether China seeks another investment strategy or not. Also, a review of this study several years later could increase the robustness of the findings. Especially, with the new more assertive foreign policy following the inauguration of president Xi Jinping in 2013, and new Chinese-initiated multinational trade initiatives like the 21st Century Maritime Silk Road, Silk Road Economic Belt, AIIB and New Development Bank. This study serves as a good asset for later studies to compare. (Raby, 2015)

I also have to note, that the reliability of these construct variables of the instructions is debated. Also, as was mentioned before, China (also the United States) tends to censor some of their investments. Access to the real database would also further show some other insight of Chinese investments in the continent.

As investments occur to shape institutions as well, later studies about how the Chinese investments affected institutions could widen this study.

The Aiddata.org publishes disaggregate investments statistics about Chinese firms in Africa, including the type of the investments, amounts and sector. However this database is far from being complete, when comparing it with the UNCTAD database we can see that there are many investments missing from the aiddata database. If these two databases becomes more comparable at a later stage, it would be interesting to compare them and see the real motives of Chinese firms.

28 | P a g e

region, but France have historical colonial connections with the region and that alters the picture.

B

Alden, C., Large, D. & Soares de Oliveira, R., 2008. China Returns to Africa: Anatomy of an Expansive Engagement, Working Paper. Real Instituto Elcano.

Anyanwu, J. C., 2011. Determinanst of Foreign Direct ivnestment Inflows to Africa, 1980-2007. African Development Bank Group.

Arndt, C. & Oman, C., 2008. The Politcs of Governance Ratings. Maastricht Graduate School

of Governance, Working Paper.

Asiedu, E., 2000. Foreign Direct Investment in Africa: The Role of Natural Resources, Market Size, Government Policy, Institutions and Political Instability. University of Kansas.

Asiedu, E., 2002. On the Determinants of Foreign Direct Ivnestment to Developing Counries: Is Africa Different?. Social Science Research Network.

Asiedu, E., 2005. Foreign Direct Investment in Africa: The Role of Natural Resources, Market Siza, Government Policy, Institutions and Political Instability, Working Paper. Social Science

Research Network.

Barney , J., 1991. Firm Resources and Sustained Competetive Advantage. Journal of

Management, 17(99), pp. 99-120.

Bernur, R. & Ersoy, A., 2009. Analyses of FDI Determinants of Mergers and Acquisitions in Banking. Journal of Finance and Financial Services, pp. 105-123.

Bezuidenhout, H. & Naudé, W., 2008. Foreign Direct Investment and Trade in Southern African Development Community. United Nations University.

Blonigen, B. A., 2005. A Review of the Empirical Literature on FDI Determinants. Atlantic

Economic Journal, pp. 383-403.

Blonigen, B. A. & Wang, M., 2004. Inappropriate Pooling of Wealthy and Poor Countries in Empirical FDI Studies.

29 | P a g e

Buckley, P. J. et al., 2007. The Determinants of Chinese Outward Foreign Direct Investment.

Journal of International Business Studies, Volume 38, pp. 499-518.

Busse, M., Erdogan, C. & Mühlen, H., 2012. China's Impact on Africa - The Role of Trade and FDI.

Busse, M. & Hefeker, C., 2007. Political Risk, Institutions and Foreign Direct Investment.

European Journal of Political Economy, pp. 397-415.

Bülent, D., 2013. The Effect of Institutional Variables on FDI Inflows: Evidence From Upper-Middle Income Countries. African Journal of Social Sciences, 3(2), pp. 113-125.

Carike, C., Elsabe, L. & Henri, B., 2012. Chinese Foreign Direct Investment in Africa: Making Sense of a New Economic Reality. African Journal of Business Management, pp. 11583-11597. Cattaneo, N. & Ezeoha, A. E., 2011. FDI Flows to Sub-Saharan Africa: The Impact of Finance, Institution and Natural Resource Endowment. CSAE Conference 2011: Economic Development

in Africa.

Ceglowski, J. & Golub, S. S., 2012. Does China Still Have a Labour Cost Advantage. Global

Economy Journal.

Chakrabarti, A., 2001. The Determinants of Foreign Direct Investment: Sensitivity Analyses of Cross-Country Regression. Kyklos, pp. 89-114.

Cheng, L. K. & Kwan, Y. K., 2000. What are the Determinants of the Location of Foreign Direct Investment? The Chinese Experience. Journal of International Economics, 51(2), pp. 379-400.

Cheng, L. K. & Ma, Z., 2008. China`s Outward Foreign Direct Investment.

Chen, R. R., 2008. The Cost of Doing Business Abroad in Emerging Markets and The Role of MNC Parent Companies. Multinational Business Review, 16(3), pp. 23-40.

Cheung, Y.-W. & Qian X-W, 2007. The Empirics of China`s Outward Direct Investment.

University of California.

Cheung, Y. & Qian, X., 2009. Empirics of China`s Outward Direct Investment. Pacific

Economic Review.

30 | P a g e

China, N. B. o. S. o., 2015. National Bureau of Statistics of China. [Online]

Available at: http://www.stats.gov.cn/english

[Accessed 10 05 2015].

Chowdhurry, A. & Mavrotas, G., 2006. FDI and Growth: What Causes What?. The World

Economy.

Clark, T. S. & Linzer, D. A., 2012. Should I Use Fixed or Random Effects?. Emory University. Cleveland, C. J. & Stern, D. I., 2001. Natural Resource Scarcity Indicators: an Ecological Economy Sythesis. im The Economics of Nature and the Nature of Economics (Edited by Cutler

J. Cleveland, David I. Stern, Robert Costanza), pp. 238-261.

Cuervo-Cazurra, A., 2006. Who Cares About Corruption?. Journal of International Business

Studies, Volume 37, pp. 807-822.

Deng, P., 2009. Why do Chinese Firms Tend to Acquire Strategic Assets in International Expansion. Journal of World Business.

Dunning, J. H., 1980. Toward an Eclecti Theory of International Production: Some Empirical Tests. Journal of International Business Studies, 11(1), pp. 9-31.

Dunning, J. H., 2000. The Eclectic Paradigm as an Envelope for Economic and Business Theories of MNE Activity. Intarnational Business Review, 9(2), pp. 163-190.

Dunning, J. H. & Lundan, S. M., 2008. Multinational Enteprises and the Global Economy.

Edward Elgar Publishing.

Dupasquier, C. & Osakwe, P. N., 2005. Foreign Direct Investment in Africa: Performance, Challenges and Responsibilities. African Trade Policy Center.

Durnev, A., Enikopolov, R., Petrova, M. & Santarosa, V., 2012. Politics, instabiity and International Investment Flows. CEFIR/NES Working Paper Series, Working Paper No 190.

Eden, L. & Miller, S. R., 2004. Distance Matters, Liability of Foreignness, Institutional Distance and Ownership Strategy, The Theory of the Multinational Firm. Advances in

International Management.

31 | P a g e

Egger, P. & Winner, H., 2004. How Corruption Influences FDI: A Panel Data Study, Unpublished Working Paper. University of Innsbruck.

Eisenhardt, K. M. & Schoonhoven, C. B., 1996. Resource-based View of Strategic Alliance Formation: Strategic and Socail Effects in Etrepreneurial Firms. Organization Science, 7(2), pp. 136-150.

Elsabe, L., Henri, B. & Carike, C., 2012. Chinese Foreign Direct Investment in Africa: Making Sense of a New Economic Reality.

Erramilli, M. K. & D'Souza, D. E., 1995. Uncertainty and Foreign Direct Investment: The Role of Moderators.

Fiodendji, K. D., 2013. Do Institutions Quality Affect FDI Inflows in Sub Saharan African Countries?.

FMPRC, 2015. Mnistry of Foreign Affairs of the People's Republic of China. [Online]

Available at: http://www.fmprc.gov.cn/

[Accessed 10 06 2015].

Foster, V., Butterfield, W., Chen, C. & Pushak, N., 2008. Worldbank.org. [Online]

Available at:

http://siteresources.worldbank.org/INTAFRICA/Resources/BB_Final_Exec_summary_Englis h_July08_Wo-Embg.pdf

[Accessed 20 05 2015].

FreedomHouse, 2015. FreedomHouse. [Online]

Available at: https://freedomhouse.org/

[Accessed 14 05 2015].

Gao, S., 2009. The Predictive Capacity of the Gravity Model of Trade on Foreign Direct Investment. Uppsala University.

Gastanaga, V., Nugent, J. & Pashamova, B., 1998. Host Country Reforms and FDI Inflows: How Much Difference Do They Make, World Development.

Gelb, S., 2010. Foreign Direct Ivnestment Links Between South Africa & China. African

32 | P a g e

GEM Report, 2014. The Global Competetiveness Report. World Economic Forum.

Gleaser, E., Porta La, R., Lopez-de-Silanes, F. & Shleifer, A., 2004. Do Institutions Cause Growth?. Journal of Economic Growth, 9(3), pp. 271-303.

Globerman, S. & Shapiro, D., 2003. Governance Infrastructure and US Foreign Investment.

Journal of International Business.

Greenaway, D., Kneller, R. & Zhang, X., 2012. The Effect of Exchange Rates on Firms Exports and the Role of FDI. Review of World Economy.

Greene, W. H., 2011. Econometric Analysis. 7 ed. s.l.:Prentice Hall.

Guillaumont, P., Guillaumont, S. & Brun J, F., 1999. How Instability Lowers African Growth.

Journal of African Studies.

Hailu, Z. A., 2010. Demand Side Factors Affecting the Inflow of Foreign Direct Investment to African Countries: Does Capital Market MAtter?. Itnernational Journal of Business and

Management, 5(5), pp. 104-116.

Henisz, W., 2000. The Institutional Environment for Multinational Investment. Journal of Law,

Economics and Organization.

Hennart, J.-F., 2009. Theories of The Multinational Enerprise. Editor: Alan M. Rugman ed. s.l.:The Oxford Handbook on International Business.

Ibrahim, G., Elhiraika, A., Hamdok, A. & Kedir, A., n.d. Revisiting the Determinants of Foreign Direct Investment in Africa: The Role of Institutions and Policy Reforms. UN

Economic Comission for Africa.

James, Q., 2013. Telegraph. [Online]

Available at:

http://www.telegraph.co.uk/finance/newsbysector/mediatechnologyandtelecoms/telecoms/978 3296/Vodafone-faces-1.6bn-tax-bill-from-India.html

[Accessed 16 05 2013].

Johanson, J. & Vahlne, J.-E., 2009. The Uppsala Internationalization Process Model Revisited: From Liability of Foreignness to Liability of Outsidership. Journal of International Business

33 | P a g e

Kang, Y. & Jiang, F., 2012. FDI Location Choice of Chinese Multinational in East and Southeast Asia: Traditional Economic Factors and Insitutional Perspective. Journal of World

Business, 47(1), pp. 45-53.

Kaplinsky, R. & Morris, M., 2009. Chinese FDI in Sub-Saharan Africa: Engaging in Large Dragons. European Journal of Development Research.

Kaufmann, D., Kraay, A. & Mastruzzi, M., 2010. The Worldwide Governance Indicators: Methodology and Analytical Issues. World Bank Policy Research Working Paper No. 5430.

Kaufmann, D. & Vicente, P. C., 2005. Legal Corruption. Available at SSRN.

Kellor, B., Hauser, W. & Griffin, A., 2009. The Relationship Between Political Risk, National Culture and Foreign Direct Investment as a Market Entry Strategy: Perspectives from U.S. Firms, Innovative Marketing.

Kersan-Skabic, I., 2013. Institutional Devleopment as a Determinant of FDI Attractiveness in South-east Europe. Juraj Dobrila University, pp. 215-235.

Khachoo, A. Q. & Khan, I. M., 2012. Determinants of FDI Inflows to Developing Countries: A Panel Data Analyses.

Klomp, J. & Haan, J. D., 2009. Political Institutions and Economic Volatility. European

Journal of Political Economy.

Kolstad , I. & Wiig, A., 2011. Better The Devil You Know?. Journal of African Business, pp. 31-50.

Kolstad, I. & Wiig, A., 2012. What Determines Chines Outward FDI?. Journal of World

Business, Volume 47, pp. 26-34.

Kostova, T., 1997. Country Institutional Profiles: Concept and Measurement. Academy of

Management Proceedings.

Kostova, T., 1999. Transnational Transfer of Strategic Organizational Practices: A Contextual Perspective. The Academy of Management Review, 24(2), pp. 308-324.

Lorenz, A. & Thielke, T., 2007. The Age of The Dragon: China's Conqust of Asia.