The influence of environmental, social and governance on Chinese company's resources related investments in Africa.

Amsterdam Business School

Master Thesis

The Influence of Environmental, Social and

Governance on Chinese Company’s Resources

Related Investments in Africa

Xiheng Shen

10667938

Supervisor: Prof. Tomislav Ladika

September 15, 2014

Table of Contents:

Abstract ... 3

1. Introduction ... 4

2. Literature Review and Hypotheses ... 8

2.1 The institutional background of Africa ... 8

2.2 Political Stability and Absence of Violence and FDI ... 10

2.3 Government Effectiveness and FDI ... 12

2.4 Control of Corruption and FDI ... 13

2.5 Health and FDI ... 15

2.6 Security Risk and FDI ... 17

3. Methodology ... 18

3.1 Dataset ... 19

3.2 Data Collection and Measures ... 20

3.2.1 Dependent Variable ... 20 3.2.2 Independent Variables ... 21 3.2.3 Control Variable ... 23 3.3 Method of Analysis ... 25

4. Results ... 26

5. Robustness Checks ... 32

5.2 Additional Analysis – Variation of Chinese FDI under Major Events ... 34

6. Discussions ... 39

6.1 Conclusions ... 39

6.2 Limitations and Direction for Future Research ... 39

Abstract

Under the background of Chinese increasing need of natural resources from Africa, this

thesis examines how host country’s ESG issues, including five factors: Political

stability and absence of violence, government effectiveness, control of corruption,

improved sanitation facilities and country security risk, affect Chinese resources related

investment in Africa. Due to the data availability, I use net FDI instead of firm-level

Chinese resources related investments as the dependent variable. It is assumed that

political stability and absence of violence, government effectiveness, control of

corruption and improved sanitation facilities are positively related to net FDI. Country

security risk is negatively related to net FDI. In this study, I employ the panel date of

20 African countries from 2003 to 2012 to test the hypothesis. Considering the

population, GDP growth rate and population as the control variable, my results show

that political stability and absence of violence (PV) and country security risk have

negative relation with FDI. Control of corruption and improved sanitation facilities

have positive relation with FDI, while Government effectiveness is not directly related

to FDI.

1. Introduction

According to the statistical data from IMF (2013), the average annual growth rate of

GDP in African area is roughly about 5.33%. As potential market growth in Africa has

been shown as outstandingly strong, Africa is estimated to be the hottest investment location in the next decade. This has large to do with the African continent’s huge

amount of mineral resources. As a matter of fact, Africa has been ranked by the US

Geological society as the largest or second-largest reserve of industrial diamonds,

manganese, bauxite, phosphate rock, cobalt, zirconium and platinum group metals.

During 2011, 6.5% of the world’s mineral exports is directly contributed from mining 20% of the world’s land area - the African continent.

Figure A: Chinese FDI to the Africa

Units: 100MM US dollars

In the recent years, Chinese foreign investments mainly focus on the two areas in the

0 50 100 150 200 250 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Flow Stock

world --- Southeast Asia and Africa (Figure A above shows both Chinese FDI flow and

stock to the Africa from 2003 to 2012), and nearly all the resources related investments

concentrated on Africa. From year of 2000, the volume of trading between Africa and

China first exceeded the amount of 10 billion US dollars and has been growing at a

high speed ever since. In 2009, China become the top 1 trading partner of Africa from

the third place in the previous year. Deborah (2013) states that from 2010 to 2012, the

annual growth rate of China-Africa trade reached to 19.3% with an absolute amount of

198.5 billion US dollars reported in 2012(Figure 1 shows China-Africa Trade Volume

from 2000 to 2012). Currently there are roughly 2000 companies from China running

business in Africa (Chinese Ministry of Commerce).

On the other hand, however, Africa is also believed to be a place with high risk of

continuous war chaos, political instability and corruption. Especially sub-Saharan

African countries have notoriously reputation of political conflicts and unstable

environment in the world. Such country risks lead to huge loss of Chinese firms’

investments in Africa for the last several years. As an illustration, in year 2011 a joint

investment of Iron project in Madagascar by two Chinese companies, which are Wuhan

Iron and Steel Corporation and Hony Capital, failed due to the political crisis of the

host country. Earlier that year, Wuhan Iron and Steel Corporation had invested in

Madagascar for about 100 million US dollars to get the rights of prospecting in iron ore.

Due to the political crisis in Madagascar, the iron project did not conduct successfully.

And share price of Wuhan Iron and Steel failed from 4.3 RMB per share in May 2011

Also on year 2011, 16th _{February Libya outbroke large scale of military chaos and later}

the chaos transformed into internal strife. More than 50 large investments projects from

China were forced to suspended, which led to the loss of about 18.8 billion US dollars.

Chinese thirteen state owned companies such as China State Construction Engineering

Corporation, Metallurgical Corporation Of China Ltd., China Railway Construction

Corporation ltd, China GEZHOUBA Group Corporation and so on, which focus on the

business of infrastructural construction, mining and telecommunication, had given up

all their projects within Libya. Most of the loss were deducted from the revenue of the companies’ annual income statement and hence reduce their net income subsequently.

And more than 33 thousand people from Chinese companies were asked to evacuate

from Libya. Therefore, for Chinese company it is extremely important to know how to

prepare for the political risk prevention. And examining the impact of different ESG

related factors on FDI will help the future Chinese companies to be more cautious.

In this study, I will research on the topic of: The Influence of Environmental, Social

and Governance on Chinese company’s resources related Investments in Africa

Certain amounts of literatures have already examined how was foreign companies

investments affected by political risks, corruption, health and security risk,

correspondingly. Other than the impact of political risks on investments that mentioned

earlier, corruption is also found to have significantly negative effect on firm’s value and

hence investments (Stefan Z 2014). Elizabeth and James (2008) also claims that

corruption has a negative and significant impact on investments for firms. And

changes the ownership structure into joint ventures (Javorcik and Wei 2009).

In 2008, Elizabeth and James find that Transition countries are more significantly

negatively affected by corruption, while in Sub-Saharan African and Latin America,

corruption has no obvious effect on firms’ investment. Using sample data covering

2,752 companies from 53 developing countries, they also state that corruption is the

most essential determinant of investments for transition countries. Later in 2009,

Javorcik and Wei indicate that corruption makes the government less transparent and

thus it is considered as a tax on foreign firms. They also mention that corruption reduces

the effectiveness of protection for foreign firms’ intangible assets and lower the value

of foreign companies’ cooperation with a local business partner. Moreover, Stefan Z

also claim that UK firms running business in higher corruption level area of the world

show negative irregular returns on passage of the UK Bribery Act 2010. He points out

that company operating exclusively in the least corrupt area has a 6.2% increase in

value compares to the company operating exclusively in the most corrupt area. Thus

companies would not be willing to invest lots of capital in more corrupt area due to

decreasing in firm value.

Therefore, this paper will consider not only the political and corruption risks but also

factors of health and security risk to make the analysis more comprehensive. Then the

thesis is going to find out the relation between all these determinants and the net FDI

by running multiple linear regression. The data sample will cover 20 specific African

2. Literature Review and Hypotheses

In the below section the literature review of the thesis will be conducted with, firstly,

the institutional background of Africa will be studied. Secondly, ESG issues will be

explored in two areas: For the part of governance, it includes Political stability and

absence of violence, government effectiveness and control of corruption. For the part

of society and environment, it includes improved sanitation facilities and country

security risk. The explanation will focus on how does each factor affect Chinese companies’ foreign investment in African countries.

2.1 The institutional background of Africa

Political stability and control of corruption are two important aspects when I study on

Africa political development. On the basis of World Bank research report focusing on measuring 212 countries’ government performance from 1996 to 2006, African

countries had made the most impressive progress in the area of control of corruption.

The measurement criteria of the report mainly contained freedom of the press, political

stability, law-based administration and control of corruption. Researcher in World Bank

claims that decreasing frequency of government corruption activities contributes to the

effective implementation of international aid and countries’ sustainable development.

The report had specifically spoken highly of Kenya, Niger and Sierra Leone, saying

those countries improved significantly in the realm of democratic accountability and

Another survey from the Political Risk Services in 2012 shows the level of government

control in corruption. In the data, the survey measures the level from 0 to 1 and lower

score represents higher level of corruption. I may see from the data that other than

countries of Botswana, Namibia, Ghana, Liberia, Senegal, SA, Tunisia and Zambia, the

rest of African countries were ranking in the bottom 50% of all countries. This survey

revealed that though African countries had made big progress in corruption

management, their level in general are still far behind other countries around the world.

The same survey also provided the result of ranking in the aspect of political stability

and absence of violence. The measure standards in this one remained the same as the

previous item - control of corruption. There are four African countries including

Namibia, Mozambique, Botswana and Zambia ranked in the top 30 countries around

the world, while the rest African countries still largely ranked in the bottom 50% of all

countries.

Except for the improvement in the area of political stability and control of corruption,

African countries has made preferential policies and rule of law in order to attract

foreign direct investment.

Ping Guo (2004) states that more than 37 African countries are member countries of

multilateral investment guarantee agency, which was established by world bank group;

More than 42 African countries are convention signed countries of international centre

for settlement of investment disputes; Over 26 African countries have signed

Convention of the Recognition and Enforcement of Foreign Arbitration Awards; More

protection of investments with foreign countries; Over 40 Africa countries are member

countries of world trade organization. Ping Guo also claims that most African countries

implement preferential policies towards foreign investment including: First, law is

enacted to ensure that government will not nationalize the foreign investment asset, and

fair and timely compensation will be provided to company if such nationalization

happened for the reason of national security and public interest. Second, tax incentives

are provided to import goods related to foreign investment and to corporate business

tax and income tax. The last but not least, some foreign investment projects will receive

subsidies for capital expenditure and other expenses.

On the other hand, Guimei Yao (2006) mentions that some African countries have

limitation towards FDI, such as the minimum requirements for project investment will

be at least 0.5 million U.S. dollars in Ethiopia; The license system is required in the

import and export business in Gabon; Compensation package has minimum

requirements when foreign companies recruited local employees and so on.

Based on Estrin and Bevan (2004), host country policies including recognition of

country risk, framework of legal system, government efficiency and so on are the most

related determinants of foreign direct investment. Thus, the following part of this

section will comprise the review of literature of governance, which consists of Political

stability and absence of violence, government effectiveness and control of corruption.

2.2 Political Stability and Absence of Violence and FDI

unstable politics, policies and foreign-exchange regime. And the result of this

uncertainty may affect the target operating result of international project or company

covers the area of revenue, cost, profit, market share, continuing business running and

etc. He also claims that political risk can be seen as economic variation caused by

political power, and thus variation would bring significant loss to the investment of

multinational enterprise (MNE). Thus MNE faces political risk covering several risks

that related to politics, policies and corruptions. In this part, I will review political

stability and absence of violence.

MIGA-Vale Columbia Center Political Risk Survey (2009) shows that no significant

difference had been found in the sequence of political risk between large companies

and small and medium enterprises (SME). All the investors consider political risk as a

serious limitation in the medium level; large firms may viewed it somewhat more so.

Nathan M. Jensen (2007) states that though high levels of political risk diminish company’s investment into several emerging markets, MNEs have found multiple

strategies to face the risks. Using a sample firm-level data on all American foreign

subsidiaries, he also explains that under high levels of political risk, MNEs choose to

invest in more liquid assets which can be easily transferred if the political condition

goes wrong and to maximize political influence by taking active cooperation with local

governments or participating in funding election of politics.

Iris and Monika (2010) finds that the ownership share of MNEs have in the host country

tends to go down under higher level of political risk, while whether leverage ratio

that investors prefer to reduce their debt exposure to avoid costly dead weight losses,

since political violence would increase the risk of default.

Further, MIGA (2011) from World Bank Group indicates in its survey that political risks

have led to huge losses to numerous foreign enterprises having business in certain

countries, and they have come up with three ways to mitigate the risks. Firstly, some

MNEs will use outside consultants or internal professionals to assess the level of risks

from time to time. Secondly, MNEs may prefer to implement non-contractual

mitigation strategies such as maintain a good relationship with government or establish

a joint ventures with local companies to hedge the political risks. Thirdly, certain

companies will choose to use contractual strategies such as buying political risk

insurance products or credit default swaps.

Due to the firm-level data availability, in this study we use the aggregate net FDI flows

instead to examine how political instability and absence of violence influence Chinese

firms. Given the literature that listed above, I assume political stability and absence of

violence is positively related to FDI.

H1a: Level of political stability and absence of violence in host country is positively

related to FDI.

2.3 Government Effectiveness and FDI

The worldwide governance indicators (WGI) defines the government effectiveness as

the measurement to examine the quality of public and civil services, policy formation

Due to the firm-level data availability, in this study we use the aggregate net FDI flows

instead to examine how government effectiveness influences Chinese firms.

Mora, Garibaidi, Sahay and Zettelmeyer (1999) state that government effectiveness as

well as other three factors are major indicators of foreign companies’ investments.

Brindusa (2005) mentions that she considered government effectiveness as an indicator

of good quality of the bureaucracy, as well as political stability and other factors, to

characterize better performance of governments. And all the data she used from 1996,

1998 and 2000 covers 140 observations, which are scored between -2.5 to 2.5, with

lower scores refer to worse outcomes. Shanta (2011) claims that high level of

government effectiveness has positive effect on foreign companies’ investment by using

the fact that when Rwanda government loosed its control in transportation the average

price decrease 75%, which attracts more investments from foreign companies than

before. Adhikary and Mengistu (2011) examine the effect of six factors on the inflows

of foreign companies’ investment covering 15 countries from 1996 to 2007. Their result

also reveals that government effectiveness with other determinants are the important

factors impacting foreign companies’ decision on investment location. Mainly, they

infer that more FDI can result from improving the environment of governance.

Therefore, I assume that:

H2a: Level of government effectiveness in host country is positively related to FDI.

2.4 Control of Corruption and FDI

measurement to examine the level of how much public power has been used for private

benefit, containing both small and significant size of corruption. Currently, there are

not so many literatures that have examined the impact of corruption on firm-level

investments. For the simple reason that “firm-level data are more difficult to assemble”

(Wei 2001). However, several surveys such as world business environment survey and

so on have revealed that corruption do impact company performance. So it is still

important to analyze how firm-level investment affected by corruption.

Generally, there are three relation between corruption and investments: positive,

negative or none-significant connection. Wei (1997) and Campo (1999) mentions that

corruption increases operational cost, raises uncertainty and hence deters company’s

investment. Batra (2003) uses sample data covering 81 countries with 3,100 companies and indicates that corruption has significantly negative effect on companies’ investment

growth. Stefan (2014) also claims that UK companies running business in high level of

corruption regions in the world show negative irregular returns on passage of the UK

Bribery Act 2010. He also mentions that UK companies develop their subsidiaries’ less

into the higher level of corruption area and their annual revenue in those area grow

more slowly. Nishat, Vikram and Chong (2014) states that firm value is outstandingly

lower in higher level of corrupt areas, while companies are less negatively impacted by

corruption if they provide goods and services to the local government.

Some other literatures’ view show in contrast to what I list earlier in this section. Gaviria

(2002) uses sample data covering 29 countries with 2,612 companies and states that no

corruption. Hellman (2002) claims that, all else equivalent, companies that profit from

corruption may increase their investment to expand their business in the host country.

Elizabeth and James (2008) explain that negative impact of corruption on investment

may be offset when corruption brings chances for illegal profits to companies – for

instance, company may pay cash or cash equivalent for lucrative contracts, which will

help company to have access to natural resource at preferencial price or to receive loan

from banks at lower interest rates.

All these previous literature suggest that general effect of corruption on firm-level

investment is not certain. Due to the firm-level data availability, in this study we use

the aggregate net FDI flows instead to examine how control of corruption influence

Chinese firms. And I consider that corruption is negatively related to FDI, which equals

to that control of corruption is positively related to FDI.

H3a: Level of control of corruption in host country is positively related to FDI.

2.5 Health and FDI

DSAED (2010) states in the report that healthier lives help people to live longer and

more meaningful. And better health condition also leads to great benefits such as

improved performance of company. In an early work of Bloom and Canning (2004),

they also indicates that health, which is considered as a kind of human capital, would

improve the performance of economy both on the level of company’s performance and

impact on countries from different background. Bhargava (2001) claims that economic

improvement of developing countries benefit more from health enhancement than that

of developed countries. In addition, Alsan (2006) indicates that there are various

reasons of health’s impact on foreign companies’ investment. For example, higher employee compensation and decreasing worker’s productivity may increase the labor

cost per unit due to bad health condition. Strauss and Thomas (1998) also states that

workers in healthier condition are much more energetic and agile than those who suffer

from illness and disability.

More recently, the outbreak of the Ebola virus has showed fears that fulminating

infectious disease outbursts would reduce the inflows of foreign companies’ investment.

Some evidences support this point of view. Elizabeth, Yi and Isaac (2012) use panel

data of 40 countries in SSA (sub-Saharan Africa) from year 1990 to 2008 to identify

the extent of causal effect of HIV/AIDS on foreign companies’ investment. They reach

to the conclusion that HIV/AIDS do have negative effect on foreign companies’

investment, but the level of effect would decrease as the population infection rate

decrease to 0.1 percent below.

In this thesis, due to the data availability, I choose the factor of improved sanitation

facilities to measure the health’s impact on foreign companies’ investment. Judging

from the general literature listed above, I assume that this factor has positive relation

with FDI.

2.6 Security Risk and FDI

Conventional security risk mainly covers the aspects of terrorism activities and

territorial conflicts. Alberto and Javier (2005) point out that foreign companies’

investment in lower levels have relation with terrorism risk in higher levels. Some

literature argues that the effect of terrorism on inflows of FDI varied from country to

country. Daniel (2006) mentions that MNEs deciding investing in Canada would have

far less concern of terrorism risk than one investing Iraq.

Later a point of view supported by other researchers indicates that untraditional security

risk has newly come to show. Such security risks mainly arises from the area of culture,

race and religion conflicts. Luo and Huang (2009) claim that nowadays some terrorists

and religious extremists in certain countries would conduct terrorism activities in order

to achieve their political aims. For instance, in year of 2004 two employees from Liao

He Oil and Construction Company were killed in the Sudan. Jiang (2012) also mentions

that the untraditional security risk are larger than before for foreign countries. For

example, over 10,000 people’s deaths in the religion conflicts since year of 1999 has

negative impact on the oil investments from china. Moreover, on January 2012, 29

employees from the Sinohydro Corporation Limited were kidnapped by the

anti-government militants in the Southern Kordofan state of Sudan. Due to the firm-level

data availability, in this study we use the aggregate net FDI flows instead to examine

how security risk influences Chinese firms’ investment. Therefore, due to the majority

of related literatures, I assume:

3. Methodology

To identify the impact of environmental, social and governance on Chinese Resource

related investments involving Africa, this thesis would start by employing a multiple

linear regression on FDI. Since it is almost impossible to get access to all the data of

the specific resource related investments of the total Chinese companies in different

African countries from the last 10 years, I measure FDI as a general indicator to

represent the Chinese resource related investments. It is quite obvious that the reason

why China would invest large sum of money in Africa is that over the last decade, the

rapid growth of Chinese economy needs huge amount of resource to support, including

metal, oil, gas and so on. In order to get access to the rights of mining and energy

harvesting, Chinese government and enterprises have made agreements with

government of African countries that Chinese companies would not only invest in the

resource industry but also put large amount of capital into the infrastructural

construction along with other industries such as manufacturing, telecommunication,

agriculture, banking, health and etc. Several literatures and reports have also supported

this point of view. Sarah (2012) from Standard Chartered states that Chinese

investments in African countries show its long-term goal of ‘going global’, which is

finding stable natural resources and matching the benefits of government-owned

companies. Thompson and Olusegun (2014) also claim that Chinese foreign investment

in Africa is significantly stimulated by the increasing needs of resources to support its

illustrate how the ESG factors affect the investment. Previous important FDI

determinants will be contained in the regression. These determinants used by varied

empirical literatures include political stability and absence of violence, government

effectiveness and control of corruption, which have been proven to have significant

relation to the inflows of FDI.

Compared to the previous papers that researched on relevant topics, this thesis is at an

advantage in two ways. Firstly, former literatures tend to research only one or two aspects of determinants’ impact on FDI, while this paper include the factor of health

and security to make the analysis more comprehensive. Secondly, previous papers use

data sample in the way of either covering certain areas of African countries or combine

the data between different continents, such as Asia and Africa, Africa and Middle East

and etc. However, this thesis mainly focus on the twenty Africa countries that Chinese

companies prefer to invest largely because of their natural resources.

The following section will introduce the part of dataset, measures and analysis of model.

3.1 Dataset

The Chinese FDI data in different African countries comes from the Statistical Bulletin

of China's Outward Foreign Direct Investment from year 2003 to 2012. This FDI report

is published by the State Statistics Bureau of China, Ministry of Commerce of China

and State Administration of Foreign Exchange of China. This Statistical Bulletin

generally covers six part related to Chinese foreign direct investment. It is an

over 50 African countries, on the base of year by year. Among the 50 African countries,

I choose 20 African countries which has large inflows of resources related investment

from China base on the information provided by three sources: The China Global

Investment Tracker 2013 from The Heritage Foundation; Chinese Companies in Africa

from the Profundo company (2014) and Africa’s path to growth: Sector by sector from

the Mckinsey company (2010).

The data on the part of governance and improved sanitation facilities to the percentage

of total population are from the World Bank Database, and cover the period 2003-2012.

Another data of the country’s security risk is from iJET Country Security Risk Ratings

(IJT). I include all the African countries that receive Chinese significant resources

related investment for which completed data are trackable, which includes a sample of

20 countries: Botswana, Cameroon, DRC, Egypt, Ghana, Guinea, Ivory Coast, Liberia,

Malawi, Morocco, Mozambique, Namibia, Niger, Nigeria, Sierra Leone, South Africa,

Sudan, Uganda, Zambia and Zimbabwe.

3.2 Data Collection and Measures

This part will introduce the data collection and statistic description regarding the

measures. It will discuss in detail about the dependent variable, independent variables

and control variable. Data-table of all the variables in summary will be illustrated in the

final part.

3.2.1 Dependent Variable

earlier, FDI can be a good indicator to measure the Chinese resources related investment

in certain African countries. And also the amount of FDI inflows is a proper way to

reveal the global understanding of the country’s economic fundamentals and market’s

future opportunities (Andrea 2003). The inflows of FDI from China can be extracted

from the Statistical Bulletin of China’s Outward Foreign Direct Investment in the period

of 2003 to 2012.

3.2.2 Independent Variables

Political Stability and Absence of Violence. Generally there are two factors measuring

the procedure of how governments are elected, monitored and changed (Daniel and

Aart 2010). They are Voice and Accountability (VA) and Political Stability and Absence

of Violence (PV). And Voice and Accountability here measures the extent of the

probability that the government will be destabilized or replaced by democratic election

or military violence, containing both politics related violence and terrorism. Chinese

investment are more obviously affected by the fragile political stability instead of voice

and accountability, which measures the democratic election and freedom of press. For

example, in year 2011 a joint investment of Iron project in Madagascar by two Chinese

companies, which are Wuhan Iron and Steel Corporation and Hony Capital, failed due

to the replacement of the host country. The value of the data of VA ranges from 0 (weak)

to 1 (strong) regarding the performance of government. That is the lower the country

graded, the higher the risk of the political instability that country had.

performance in formulating and implementing the policies (Daniel and Aart 2010).

They are Government Effectiveness (GE) and Regulatory Quality (RQ). In this paper I

examine the impact of GE on FDI. GE measures the quality of public and civil services, policy formation and execution, the fulfillment of government’s promise to its public

policies and so on. The value of the data of GE ranges from 0 (weak) to 1 (strong)

regarding the performance of government. That means the higher the country scored,

the more efficiently the government operated its daily affairs.

Control of Corruption. Two common indicators are generally used to measure the extent

of citizens being respected and of how institutions monitor the social and economic

interactions (Daniel and Aart 2010). They are Rule of Law (RL) and Control of

Corruption (CC). In this study, I identify the effect of control of corruption on FDI. CC

here examines the level of how much public power has been used for private benefit,

containing both small and significant size of corruption. The value of the data of CC

ranges from 0 (weak) to 1 (strong) regarding the control of corruption. That means the

higher the country scored, the more honest that the country has been considered.

Health. As mentioned earlier health, considered as a kind of human capital, would

improve the performance of economy both on the level of individual and macro

economy. In this paper, I use improved sanitation facilities (ISF) as the indicator to

measure the impact of health on FDI. According to the World Bank database, access to

improved sanitation facilities directs to the percentage of the country’s total population

utilizing the ISF. The value of the data of ISF ranges from percentage 0 (weak) to 100

percentage, the more the people using improved sanitation facilities and the less

probability that people getting affected by infectious disease.

Country Security Risk. In this thesis, I use country security risk (CSR) to measure how

FDI inflows is affected by the security risk, which covers the aspect of religion, race

and territorial conflicts. The value of the data of CSR ranges from 0 (weak) to 1 (strong)

regarding the security level of country, which indicates that the higher the country

graded, the better security condition that the host country has. For the convenience of

explanatory, the security index in the dataset is multiplied by -1 and thus higher security

score refers to a higher insecure environment of the host country.

3.2.3 Control Variable

As mentioned earlier, several literatures believes that market size plays an important

role in FDI, and one reason can be that it brings profitability to both local and export

sales and various resources (Pfefferman and Madarassy 1992). Another report claims

that GDP growth rate and size of middle class are the ways to measure market size

(Pravakar 2006). Chakrabarti (2001) also states that country’s openness to trade and

GDP growth rate are most likely correlated with FDI. Asiedu and Lien (2004) further

mentions that openness to trade is outstandingly essential for MNEs which export goods

from the host country to the world. Besides, African countries’ population is also

important correlated with FDI. Due to the data availability and that other control

variables using as labor costs, tax and so on are less statistically significant when

well as the openness rate, as the control variable to test the relative effect of other ESG

related factors on FDI amount in this study.

Table I: Variables in Summary

Variable Name Type Measures Data Source

FDI Main FDI per country Statistical Bulletin of China’s Outward Foreign Direct Investment 2003-2012

PV Main Political Stability and Absence of Violence

Political Risk Services International Country Risk Guide (PRS)

GE Main Government

Effectiveness

Political Risk Services International Country Risk Guide (PRS)

CC Main Control of Corruption Political Risk Services International Country Risk Guide (PRS)

Health

Improvement(HI)

Main Improved Sanitation Facilities

The World Bank

SR Main Security Risk iJET Country Security Risk Ratings (IJT)

GDP Growth Control GDP Growth per country

The World Develop Indicators

Population Control Population per country The World Develop Indicators

Openness Control Goods and services trade openness

United Nations Conference on Trade and Development

3.3 Method of Analysis

In this report I choose multiple linear regression analysis to test the hypotheses. And

the software implemented in the analysis is SPSS 20. The formula can be written as

follows:

Y

=α+β

*X

+β

*X

+β

*X

+β

*X

+β

*X

+β

*X

+β

*X

β

*X

ε

Through this formula, Y represents the dependent variable, while XPV ~ XSR stands for

the independent variables and XGDP Growth ~ XOpenness represents control variables .

ε

means the residual term and α represents the point where the line intersects the axis of

Y. I will test three regression models:

1. To find the connection between the outcome variable and control variables.

2. To find the connection between the outcome variable and hypothesized predictor

4. Results

4.1 Descriptive Statistics

All the results of variables from descriptive statistics analysis are shown in Table 2:

Descriptive Statistics

N Minimum Maximum Mean Std. Deviation

FDI 200 -81491.0000 480786.0000 6426.055000 35890.4264939 PV 200 .4280 .8864 .683657 .1179168 GE 200 .0000 .6250 .333750 .2013265 CC 200 .0000 .6667 .328333 .1353880 HI 200 7.1000 95.9000 34.072000 23.1732413 SR 200 .0000 .7500 .383750 .2376041 GDP Growth 200 -32.8321 33.7358 4.807249 5.6162609 Population 200 1832602.0000 168833776.0000 29407893.345000 33594741.8937139 Openness 200 176.8530 123823.5000 14164.254398 24488.5700164 Valid N (listwise) 200 ----Table 2

The explanation of all the variables from the descriptive analysis can be as listed as

follows: Firstly, the amount of net FDI inflows changes from country to country,

varying from -81,491 to 480,786 (measured in 10K US dollars). The negative FDI are

reasons of disinvestment or reverse investment. Comparing to the average level of the

rest countries in the world, the mean amount of African countries’ governance

4.2 Correlation Analysis

All the results of independent and control variables from correlation analysis, which also includes the Pearson’s correlation coefficients, are shown in Table 3:

Coefficient Correlationsa

Openness GDPGrowth Population PV CC GE HI SR

Openness 1.000 GDPGrowth .170** 1.000 Population -.910 .082* 1.000 PV -.120 .133** .173 1.000 CC -.365 .059 .437 .338*** 1.000 GE .014*** .026 -.116 .484** .141*** 1.000 HI -.108 .081* .084 .050 -.196 -.670 1.000 SR .015** -.094 .053 -.352** -.028* -.413* .463 1.000

a. Dependent Variable: Normal Score of FDI using Blom's Formula b. *Significant at 10% level

c. **Significant at 5% level d. ***Significant at 1% level

----Table 3

From above, I can see that correlations between independent and control variables

reveal different results. There are seven pairs of variables show relation between each

other. GDP Growth is positively related to Openness (β=.170, p<0.05), showing that

more goods and services trade would lead to higher GDP growth rate. Political

stability and absence of Violence is positively related to GDP (β=.133, p<0.05),

indicating that better political stability would lead to higher amount of GDP. Control

of corruption is positively related to Political stability and absence of violence

(β=.338, p<0.01), showing that higher level of political stability would lead to lower

level of corruption. Government effectiveness is positively related to both control of

p<0.05). This shows that enhanced government performance will increase the level of

control of corruption and political stability, correspondingly. Country security risk is

negatively related to political stability (β=-.352, p<0.05), indicating that higher level

of security risk will decrease political stability. In addition, Country security risk is

negatively related to government effectiveness (β=-.413, p<0.1), showing that lower

level of security risk will improve the government efficiency.

4.3 Regression Analysis

In this study, I examine the impact of independent variables on dependent variable.

Before conducting the regression model, the distribution of the residuals is found to be

non –normality. Consequently, I use a normalizing (Blom) transformation to all the

variables. Next, I establish three regressions in this thesis. The first model merely

contains three control variables and dependent variable: GDP growth rate, population,

openness and FDI. I add three independent variables including political stability and

absence of violence, government effectiveness and control of corruption to the second

model. The third model contains all the variables, including independent variables,

control variables and dependent variable.

In order to eliminate the effect of multi-collinearity, I check the analysis of collinearity

statistics in the Table 4 below. There is no significant evidence showing

multi-collinearity from the column of Tolerance and Variance Inflation Factor. Normally, if

the tolerance is higher than 0.1 and the VIF is less than 10, there will be no

Seen from the Table 4, the lowest Tolerance is 0.116 and the highest VIF is 8.586,

showing that no significant multi-collinearity exist in this model.

Coefficientsa

Standardized Coefficients

t Collinearity Statistics

Beta Tolerance VIF

GDP Growth .491 -1.191 .860 1.163 Population .242 2.124 .116 8.586 Openness .162 1.898 .439 7.209 PV -.234 -2.583 .189 5.581 GE .053 4.573 .379 2.603 CC .362 2.183 .184 5.446 HI .263 3.901 .243 4.120 SR -.440 2.656 .422 8.199

a. Dependent Variable: Normal Score of FDI using Blom's Formula

----Table 4

Table 5 below indicates the result of all three models. Since the third model has the

highest R2 (.633), which shows that independent variables accounts for a higher

percentage of variation in dependent variable. Therefore, I choose the third model in

this study.

In model 3, political stability and absence of violence is negatively (β=-.234) and

significantly (p<0.05) related to Chinese net FDI, indicating that political instability overall did not have negative impact on the decision of Chinese company’s investment

decision. This is contrary to what I assume in literature review that level of political

stability and absence of violence in host country is positively related to FDI. Thus the

Table 5: Regression Results Model 1 2 3 R2 _{.333 .487 .633} Adj. R2 _{.110 .237 .401} Number of Observations 200 200 200 Independent Variables

Political Stability and Absence of Violence -.398 -.234

(.035**) (.015**) Control of Corruption .567 .362 (.002***) (.037**) Government Effectiveness .332 .053 (.414) (.284) Health Improvement .263 (.000***) Security Risk -.440 (.012**) Control Variables Openness .177 .229 .162 (.043**) (.067*) (.046**) GDP Growth .517 .505 .491 (.089*) (.025**) (.034**) Population .541 .371 .242 (.021**) (.044**) (.042**) a. Dependent variable: Normal Score of FDI using Blom's Formula

b. Control variables: Openness, GDP growth and Population

c. Model 1 plus political stability and absence of violence, corruption of corruption and government effectiveness

d. Model 2 plus health improvement and security risk e. P value of beta coefficients are listed in parentheses above f. *. Significant at the 0.1 level

g. **. Significant at the 0.05 level h. ***. Significant at the 0.01 level

Besides, the result indicates that government effectiveness is not significant (β=.053,

p>0.1). Thus there is no direct link between Chinese net FDI inflows and government

effectiveness. This is also contrary to what I assume earlier that level of government

effectiveness in host country is positively related to FDI. Therefore the hypothesis 2a

does not hold either.

The standardized coefficients result shows that control of corruption is positively (β=.362) and significantly (p<0.05) related to FDI. This reveals that higher level of

control of corruption in host country will attract more FDI inflows from China. The

result is consistent with the assumption I list earlier. Therefore, the hypothesis 3a holds.

The above result indicates that health improvement is positively and significantly (β=.263, p<0.01) related to FDI. Thus this represents higher level of health

improvement would lead to more Chinese net FDI inflows. This is also consistent with

what I assume earlier that level of improved sanitation facilities in host country is

positively related to FDI. Therefore the hypothesis 4a holds.

From the standardized coefficients result, the last independent variable – country

security risk is negatively (β=-.440) and significantly (p<0.05) related to FDI. This

reveals that higher level of security risk in host country will attract less FDI inflows

from China. The result is consistent with the assumption I list earlier. Hence, the

5. Robustness Checks

5.1 Robustness Regressions

We have tested that control of corruption is positively related to foreign companies’

investments, which indicates that lower level of corruption would attract MNEs to

invest in the host country. However, it is also understandable that corruption may have

positive impacts on foreign companies’ investment if the host country has very strict

rules for MNEs to do business and corruption may accelerate the decision process of

the government. Moreover, it can be concluded that the willingness to participate in the

corrupt activities for the Chinese companies largely depends on the profitability of the

return from the investment in the host countries. Generally, we can assume that Chinese

firms are less deferred by the corruption if potential profits are much higher than the

cost of corruption such as paying cash for projects or contracts.

Thus I further test for robustness by splitting the sample countries into two groups: high

growth and low growth African countries, and repeating the regressions in each

subsample. We use geometric mean method to choose the 10 higher GDP growth rate

African countries from sample data in the period of 2003 to 2012 as: Nigeria, Ghana,

Mozambique, Uganda, Sierra Leone, Zambia, DRC, Namibia, Niger and Egypt. The

rest countries from sample data are the ones with the lower GDP growth rate, which are

Morocco, Botswana, Liberia, Malawi, Sudan, South Africa, Cameroon, Guinea, Ivory

Coast and Zimbabwe. Therefore, we assume H1b: corruption has less negative effect

Table 6: Robustness Regression Model 4 5 6 7 R2 _{.589 .640 .259 .388} Adj. R2 _{.347 .409 .067 .151} Number of Observations 100 100 100 100 Independent Variables

Political Stability and Absence of

Violence -.209 -.148 -.416 -.427 (.029** ) (.043**) (.055*) (.027**) Control of Corruption .361 .406 -.075 -.047 (.000** *) (.000***) (.051*) (.076*) Government Effectiveness .259 .188 .226 .150 (.284) (.174) (.329) (.251) Health Improvement .385 .427 (.098*) (.025**) Security Risk -.126 -.101 (.042*) (.078*) Control Variables Openness .092 .274 .042 .120 (.675) (.081*) (.819) (.052*) GDP Growth .429 .503 .412 .458 (.068*) (.039**) (.096*) (.013**) Population .404 .229 .117 .102 (.003** *) (.045**) (.017**) (.092*)

a. Dependent variable: Normal Score of FDI using Blom's Formula b. Control variables: Openness, GDP growth and Population

c. Model 4&5 represents sample countries with higher GDP growth, Model 6&7 represents sample countries with lower GDP growth

d. Model 5&7 plus independent variables of health improvement and security risk e. P value of beta coefficients are listed in parentheses above

f. *. Significant at the 0.1 level g. **. Significant at the 0.05 level h. ***. Significant at the 0.01 level

From Table 6, we find that control of corruption in higher growth rate African countries

(Model 5 with higher R square) is positively (β=.406) and significantly (p<0.01) related

to the FDI. While control of corruption in lower growth rate African countries (Model

7 with higher R square) is not significant (β=-.047, p<0.1). This indicates that

corruption have much larger negative effects on foreign companies’ investment in the

host countries with higher growth rate than those with lower growth rate. This is

contrary to what we assume earlier, and thus H1b dos not hold.

5.2 Additional Analysis – Variation of Chinese FDI under Major Events

In this thesis, we have also find that political instability overall did not have negative

impact on the decision of Chinese companies’ investment. This could be explained that

for the majority of the African countries, the rapid economic growth and large natural

resources are significantly appealing to Chinese companies’ investment so that most of

the companies are willing to take the risks of political instability, which in the last

decade happens not so often comparing to the last century. However, this may not be

the case for the countries who have been through serious political crisis or chaos.

Further, I will compare Chinese FDI before and after these events in certain African

On January 2009, Madagascar outbroke political crisis during which at least 130 people

were killed. This political crisis lasted until the end of 2013 when the presidential

elections were considered as credible. As mentioned earlier, investments from two

Chinese companies- Wuhan Iron and Steel Corporation and Hony Capital failed in 2011

due to this political crisis. And share price of Wuhan Iron and Steel failed from 4.3

RMB per share in May 2011 to 2.9 RMB per share in December 2011. We can see that political risk do affect Chinese firms’ investment profitability, and hence net FDI.

Shown from the Table 7, Chinese FDI in Madagascar reached to pitch in 2008 and

decreased significantly since the political crisis happened in 2009.

In the late of 2010, a revolutionary activity of demonstrations, strikes and protests

named “Arab Spring” had outbroken in the Middle East and North Africa. Countries

such as Algeria, Egypt, Libya and Tunisia had been affected radically in the area of

politics and economy. Foreign investments decision from MNEs were also impacted

0 10 20 30 40 50 60 70 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Table 7: Chinese FDI in Madagascar before and after political crisis in 2009. Unit: MM US dollars

significantly. Shown from the Table 8 above, we can see that Chinese companies’

investments in Algeria had decreased dramatically from year 2009 to 2011, while

Chinese FDI in Egypt decreased first in 2010 and then increases ever since. Country of

Tunisia had been affected less significantly, because its relatively less amount of natural

resources reservation are not appealing to Chinese MNEs.

As mentioned earlier, over 13 Chinese companies had suffer huge loss due to the chaos

happened in Libya in 2011. Four listed Chinese companies each has over 4 billion US

dollars of contracts with Libya local government at that year. From the February 16th

of the outburst of political chaos till the end of March when all the Chinese companies

evacuated from Libya, share price of China State Construction Engineering

Corporation dropped about 7.1% from RMB 3.77 to 3.51; share price of Metallurgical

Corporation Of China Ltd dropped about 3.9% from RMB 4.15 to 3.99; share price of

China Railway Construction Corporation ltd dropped about 17.8% from RMB 8.21 to

6.87; and share price of China GEZHOUBA Group Corporation dropped about 22.1%

-100 -50 0 50 100 150 200 250 300 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Table 8: Chinese FDI in North African countries before and after "Arab Spring" in 2010. Unit: MM US dollars

from RMB 14.63 to 11.73.This matches with the outcome in Table 8 that Chinese FDI

decreases largely since 2011.Given the above analysis, we may see that the extent of

impact from political risks on Chinese MNEs investment varied in different African

countries.

Another point earlier mentioned in the independent variables of methodology part that

two common indicators are generally used to measure the extent of citizens being

respected and of how institutions monitor the social and economic interactions (Daniel

and Aart 2010). They are Rule of Law (RL) and Control of Corruption (CC). We only

includes control of corruption in the regression analysis, so in the additional analysis

we will discuss how the rule of law impact the Chinese companies’ financial

performance with a specific case.

China Nonferrous Mining Corporation Limited is listed in Hong Kong and operating

business mainly in southern Africa. Among all the 15 projects, 13 of which company

possesses are located in Zambia. According to the article of Zambian Economist in July, Zambia’s mining minister made a statement on June 28th_{, suggesting that mining taxes}

were going to be reduced in its new mining taxation policy. Shown from the Table 9

below, CNMC’s share price increased about 5.3% after June 28th_{. However, several}

days later the same minister said to CNBC Africa during an interview that Zambia will

be benefiting from mining, which indicates that mining taxes will have to increase since

the Zambian government has a huge funding hole of around US $700 million at the moment. Seen from the Table 9, we may see that company’s share price decreased about

would also affect Chinese MNEs financial performance, and hence Chinese FDI.

Source: Bloomberg Markets.

1.65 1.7 1.75 1.8 1.85 1.9 1.95 2 2.05

Table 9: China Nonferrous Mining Corporation Limited's (CNMC) share price. Unit: HKD

6. Discussions

6.1 Conclusions

In this thesis, I use multiple linear regression model to identify the impact of ESG

related factors on Chinese net FDI inflows. I separate the environmental, social and

governance into five aspects: Political stability and absence of violence, government

effectiveness, control of corruption, improved sanitation facilities and country security

risk. I found that Political stability and absence of violence (PV) and country security

risk are negatively related to FDI; Control of corruption and improved sanitation

facilities are positively related with FDI; And Government effectiveness has no direct

relation with FDI. Our results support the findings of Swain and Wang (1997), who

points out that corruption and non-transparent system has negative effect on economic

environment and thus decrease the level of FDI inflows. Our results are also consistent

with the result of Alberto and Javier (2005), who argues out that net FDI in lower levels

have relation with terrorism risk in higher levels. However, for the aspects of Political

stability and government effectiveness, our result do not support the theory previously

mentioned in the existing literature.

6.2 Limitations and Direction for Future Research

Firstly, I use Chinese net FDI inflows as the indicator of resources related investment.

Although generally all the Chinese foreign investment in Africa is significantly

growing and industry expanding (Thompson and Olusegun 2014), certain amounts of

Chinese investment projects in these 20 African countries still exist and have indirect

relation with the resource. Future studies should examine the amount of resources

related investments from China in a certain range of areas, such as metal, oil and gas,

coal and etc. Secondly, in this study I only employ GDP growth rate, population and

openness as control variable. There are other factors which also have effect on FDI and

should be eliminated from affecting the dependent variable in the thesis. For future

study, determinants such as inflation, exchange rate, tax policy, education, income and

so on may also be taken into consideration. Thirdly, I choose 20 specific African

countries which has large inflows of Chinese resources related investments to be the

sample data. But the time period only lasts for 10 years. Since Chinese economy has

been expanded at the growth rate of almost 10% during the last 30 years (Graeme 2010),

future study should focus on a longer time period to examine the impact of different

ESG determinants on Chinese companies’ investment. Lastly, in this study I use

governance, health and security to measure the whole effect of ESG issues due to the

availability of data source. Following studies should also include the aspects of climate

changing, environment pollution, human rights and so on to represent a more

Reference

Andrea v, D., Lotte, G., and Chloe, J. (2012). Bridging the Gaps: Effectively Addressing

ESG Risks in Emerging Markets. Sustainalytics emerging markets’ report, June, 1-35.

Bailey, Warren, & Peter, C. (1995). Exchange Rate Fluctuations, Political Risk, and

Stock Returns: Some Empirical Evidence. Journal of Financial and Quantitative

Analysis, 541-61.

Beck, T., Demirguc, A., & Maksimovic, V. (2005). Financial and Legal Constraint to

Growth: Does Firm Size Matter? Journal of Finance, 60 (1), 137 -157.

Besada, H., Wang, Y., & Whalley, J. (2008). China’s Growing Econimic Activity in

Africa. National Bureau of economic research, May 14024.

Campbell, R. H. (1994). Predictable Risk and Returns in Emerging Markets. Review of

Financial Studies, 1995, pp. 773-816.

Christa, H., & Stefanie, K. (2012). Political risk, project finance, and the participation

of development banks in syndicated lending. Journal of Financial Intermediation, April,

287-314.

Daniel, A. A., Beatriz, M. R., & Antonio, R. D. (2001). Innovation and firm size: an

empirical study for Spanish engineering consulting companies. European Journal of

Innovation Management, Vol. 4 Iss: 3, pp.133 – 142.

David, V. (2013). China’s bribery culture poses risks for multinationals. Bloomberg

Dierk, H. & Peter, N. (2012). FDI and health in developed economies: A panel

cointegration analysis. Kiel Institute for the world economy, February, No. 1756.

Domowitz, I., Glen, & Madhavan. (1998). Country and Currency Risk Premia in an

Emerging Market. Journal of Financial and Quantitative Analysis, 33(2): 189-216.

DSAED. (2010). The role of health in economic development. DSAED Press, July.

Elizabeth, A., & James, F. (2008). The effect of corruption on investment growth:

evidence from firms in Latin America, Sub-Saharan Africa and Transition Countries.

Wheaton College Press, Feb, G31, O16.

Finance and economics. (2003). Political-risk analysis: The new bull market. The

Economist, May, 1-2.

Haglund, D. (2008). Regulating FDI in weak African states: a case study of Chinese

copper mining in Zambia. The Journal of Modern African Studies, 46(04), 547.

Haque, N.U., Nelson, M., & Matheison, D.J. (2000). Rating Africa: The Economic and

Political Content of Risk Indicators. In P. Collier and C. Pattillo (Eds.), Investment and

Risk in Africa, pp. 33-70.

Javorcik, B. S., & Wei, S.-J. (2009). Corruption and cross-border investment in

emerging markets: Firm-level evidence. Journal of International Money and Finance,

28(4), 605–624.

Jiang, X. L. (2012). Study on political risk of Chinese enterprises’ direct investment in

Kolstad, I., & Wiig, A. (2011). Better the devil you know? Chinese foreign direct

investment in Africa. Journal of African Business, 12(1), 31–50.

Lauren, G., Alexander, H., & Lin, J. (2013). China’s trade and investment relationship

with Africa. USITC executive briefings on trade, April, 1-2.

Laza, K. (2007). The spread of political risk: A growing threat to foreign investment.

The Economist, Nov, 1-2.

Lubos, P., & Pietro, V. (2011). Political uncertainty and risk premia. Journal of

Financial Economics, Sep, 520-545.

Luo, H. Y., & Huang, C. J. (2009). Political risk management of the investment in

Africa from Chinese enterprises. Yunnan Business and Economics Press, April,

1674-4543(2009)04-0140-06.

Marcella, A., David, E. B., & David, C. (2004). The effect of population health on

foreign direct investment. National bureau of economic research, June, 10596

Maya, F., Simon, Z., & Yang, G. (2010). Corporate social responsibility and African

Development. Chinese Academy of Social Sciences, April, 1-34.

Nadine, M. (2012). Egypt post-Mubarak legal challenges test foreign investors.

Bloomberg Markets. Oct, 1-2.

Nishant, D., Vikram, N., & Chong, X. (2014). Firms in corrupt environments and the

value of corporate governance. Georgia Institute of Technology press, May, D73, G34.

in Africa. China Academic Journal, 7, 55-60.

Rodolphe, D., & Celine, A. (2008). Public governance, health and foreign direct

investment in Sub-Saharan African. University of Strathclyde Press.

Roland, D. (2012) China’s Foreign investment in natural resources. University of

Westminster Press, December, N.62.

Sarah, B. G. (2012). Beyond trade – China-Africa investment trends. Standard

Chartered Press, February, GR12JA.

Scharfstein, D. S. & Jeremy, C. S. (2000). The dark side of internal capital markets:

divisional rent-seeking and inefficient investment, Journal of Finance, Dec, 2537-2564.

Smarzynska, B. K. & Wei, S. (2002). Corruption and Cross-Border Investment:

Firm-Level Evidence. NBER Working Paper.

Soku, B. & Zhao, X. (2014). Contracts, Governance, and Country Risk in Project

Finance: Theory and Evidence. Journal of Corporate Finance, March, G32, 56.

Sophie, R. (2013). Mining the ESG Ground: Assessing the financial component of

sustainability strategies in the mining sector. Schroders research’s paper, December,

1-22.

Stefan, Z. (2014). Bribes and Firm Value. Finance Department of INSEAD press, April,

2-32.

Stevens, F. Y. (1997). Quantitative perspective on political risk analysis for direct

The World Bank (2008). Economic impacts of sanitation in Cambodia. Research Report,

February 2008.

Thomas, L., & Anne-C.H-T. (2009). The Listed Mining sector & ESG risks. Novethic’s

CSR research, September, 1-60.

Thompson, A., & Olusegun, S. (2014). China in Africa: An evaluation of Chinese

investment. Initiative For Public Policy Analysis Press.

Zhang, X., & Daly, K. (2011). The determinants of China’s outward foreign direct

investment. Emerging Markets Review, 12(4), 389–398.

Zadek, Simon, Chen X-H, Li Zhaoxi, Jia Tao, Zhou Yan, Kelly Yu, Maya Forstater, and

Guy Morgan. (2009). Responsible Business in Africa: Chinese Business Leaders’

Perspectives on Performance and Enhancement Opportunities. Development Research

Centre of the State Council of P.R. China (DRC-ERI), November, 1-22.

Zeng, G. J., Li, J., & Peng, Q. (2008) A study on determinants of public companies’