UNIVERSITY OF GRONINGEN
How does Chinese
involvement in Africa
affect human
development?
The role of trade, aid and FDI
By Iris Mireille Kerstens University of Groningen Faculty of Economics and Business MSc International Economics and Business
Supervisor: Dr. G.J. (Gaaitzen) de Vries Co-assessor: Prof. Dr. M.P. (Marcel) Timmer
1
Abstract
In my thesis I investigate the effect of Chinese aid, trade and FDI flows on human development and economic growth in African countries. I have collected panel data for 42 Sub-Saharan African countries for the period between 1990 and 2013. My results indicate that Chinese aid, FDI and trade flows positively affect both human and economic development in African countries. The impact on economic development is however larger than the effect on human development. A specific analysis on the effect of natural resource trade indicates that the exports of natural resources to China negatively impacts human development.
2
List of abbreviations
FDI Foreign Direct Investment
GDP Gross Domestic Product
GNI ` Gross National Income
GNI Gross National Income
HDI Human Development Index
IMF International Monetary Fund
ODA Official Development Aid
ODA Official Development Assistance
OECD Organisation for Economic Co-operation and Development
ROW Rest Of the World
SSA Sub-Saharan Africa
UN United Nations
UNDP United Nations Development Programme
3
Table of contents
Abstract ... 1 List of abbreviations ... 2 1. Introduction ... 4 2. Literature ... 83. Methodology and data ... 14
Model ... 14
Data ... 16
4. Empirical results ... 21
Descriptive statistics ... 21
5. Results ... 26
6. Limitations and Future research ... 35
7. Conclusions ... 36
References ... 38
4
1. Introduction
In this thesis I will investigate what effect Chinese aid, trade and FDI flows have on human development in African countries. Additionally I will investigate whether the effect of these flows on economic growth is similar, and if trade in natural resources play a specific role. For this research I will use a Solow type growth model that is similar to Busse, Erdogan and Mühlen (2014). I will compare the results of my economic growth model to their findings, and additionally expand the research by adding my human development model. This will allow me to assess the effect of Chinese activities in Africa not only in terms of economic performance but also on a broader level by looking at the effects on human development.
There are many studies that have researched the effect of aid, trade and FDI on economic growth. Also there are some articles on the effects these factors have on human development. In my research I want to combine these elements and see what the effects are if we consider China specifically. This subject is becoming more and more relevant because in the past decade China has become an important world player with increasing influence on the world economy. China’s increased involvement in Africa and its effect on the economy and welfare in these countries is a hot topic in the economic literature and is continuously under the loop of the United Nations. China’s activities involve depleting large amounts of natural resources out of SSA. The incentives and whether or not SSA countries should welcome these and other Chinese activities is sometimes questioned (Brautigam, 2011; Zafar, 2007). This thesis will add to existing literature, because it looks at the effect of Chinese activities in Africa through three channels; aid, trade and FDI. Also this research considers the effect these factors have on both human development and economic growth. Previous research has only reviewed the effects on the economy in general, not taking into consideration the broader effect it might have on welfare.
China is progressively becoming more important as a world power. Its developing economy is undergoing economic reform and is increasingly opening up to the rest of the world. In order to sustain its high rates of economic growth it is becoming dependent on other economies. Wages of Chinese labourers are rising, and it needs to ensure its future supply of oil and raw materials. In these particular aspects Sub-Saharan Africa1 has become an important region for China, which led to increased involvement. On the contrary Africa is still struggling for improvement of its economy. African economies have suffered from years of negative economic growth rates and only experienced uplift in economic growth since 2000. African companies experience harsh competition
1
5
from Chinese low-cost manufacturing firms but rising domestic demand and labour costs in China might open up a spot for Africa.
The relationship between African countries and China has intensified over the years. Trade between China and SSA has increased drastically in the last decade. Export to China accounted for 12.9 percent of total merchandise trade in 2009, making China Africa’s largest trade partner (OECD, 2011). In 1992 this was only 1.2 percent, showing a large increase of Chinese involvement over the past years. Also in opposite direction involvement has increased, where total merchandise imports from China increased from 1.8 percent in 2001 to 14 percent in 2009 (OECD,2011). Also Chinese FDI has increased over the years. Chinese foreign direct investment flows to Africa increased from only $75 million in 2003 to over $2.5 billion in 2012 (MOFCOM, 2013). When we take Chinese economic cooperation figures into account the increase in Chinese involvement becomes even more apparent. According to China’s national bureau of statistics the value of the turnover on economic projects was $40.8 billion in 2012, where it was only $1.7 billion in 2001.
In the literature it is often emphasized that openness to aid, trade and FDI positively influences per capita income. (Borensztein, Gregorio & Lee, 1998; Makki & Somwaru, 2004) For my research I would however like to further extend this debate by also focusing on a broader level of social development. Therefore I take into account the United Nations Development Programme’s (UNDP) Human Development Index2 as a measure for social welfare. The United Nations claim that this is a better measure for measuring social welfare because GDP per capita only measures income whereas HDI also includes longevity and education. By using both data on GDP and the HDI index as dependent variables in my empirical analysis we can compare the different effects Chinese investments, trade and aid flows have.
At the same time as the involvement with China intensified, Africa’s growth performance also improved. Africa’s GDP per capita between 2000 and 2010 grew with an annual average of 3.1 percent (World Bank, 2015). This growth finally started to accelerate after two decades of negative growth rates in the 1980’s and 1990’s. Also the Human Development Index trends show positive growth rates in welfare in terms of income, life expectancy and education over the years. The Human Development Index of the United Nations rated the African continent’s annual average HDI growth between 1990 and 2000 at 0.52 percent. This was even higher between 2000 and 2013 where annual average HDI growth was already 1.37 percent. Growing from an average index of 0.399 in 1990 to an index of 0.502 in 2013 (index between 0 and 1).
2
6
Trade could positively influence human development because it creates conditions for a better standard of living, which in its turn begets a better educational system, and higher levels of healthcare and social services. An increased level of trade directly influences levels of income which indirectly affects levels of welfare because they are transmitted via income. Also a variety of new goods will be made available through trade. Products like medicine and nutrition products, that were previously unavailable, can positively influence the longevity of the people in Africa. African consumers can benefit from low-cost Chinese product. Therefore an extra inflow of income through trade, and trade of products itself, could positively influence welfare and GDP levels.
Also higher levels of FDI and economic cooperation trough Chinese investments in Africa could go hand in hand with economic growth and higher levels of welfare. Chinese involvement in African economies can increase business activity and stimulate competition which in its turn can create higher levels of productivity for African firms. Also native firms can benefit from knowledge and technology spill overs by cooperating with Chinese firms. Economic cooperation projects that focus on the much needed investments in infrastructure and education improvements can increase welfare in these countries.
Despite the positive effects, China’s engagement in Africa can also negatively affect the welfare and economic development in these countries. Chinese FDI figures show that a large part of the FDI done by Chinese investors is concentrated in the mining industry (China Statistical Yearbook, 2013). Chinese companies need to ensure the future supply of oil and other raw materials necessary to keep producing at high scales and at low costs. Most of Chinese trade and investments are therefore focused on depleting natural resources out of African soil in order to facilitate the high production levels taking place in China. The level of Chinese investments focused on education or health on the other hand are much lower. This high demand for natural resources could create exchange rate overvaluation which makes African products relatively expensive compared to the Chinese products. Also the weak institutions combined with large demand for expensive raw materials could result in rent-seeking activities and corruption. According to Szirmai (2005) an equitable distribution of income is necessary to maximize growth rates and poverty reduction.
7
related. By looking at three levels of interactions at the same time (FDI, aid and trade flows) I can obtain a comprehensive view of China’s impact on African development. Also a focus will be on the effect of China’s special interest in natural resources in the African region. In my analysis I will use panel data for 42 sub-Saharan African countries in the period 1990 until 2013. The following research questions will be answered at the end of this thesis:
Does Chinese FDI increase human development in Africa?
Does Chinese aid increase human development in Africa?
Does Chinese trade increase human development in Africa?
What impact do FDI, aid and trade flows from China have on economic growth in Africa?
Does trade in natural resources impact human and economic development in Africa differently?
8
2. Literature
The sections in this chapter provide an overview of the relevant literature. In the different sections I will elaborate on how previous research has evaluated the effect of aid, FDI, trade on human development and economic growth. Additionally I will discuss previous literature on the Solow model and growth economics. Because the literature on these subjects is very comprehensive I will only discuss previous research that is within the scope of this thesis.
Aid
A lot of research has been done on the general effect of aid on human development and economic growth. The results on the relationship between aid and development are however very diverse and cannot all be discussed, therefore only the most relevant arguments in the literature will be discussed below.
We can argue that aid finances projects and programmes which would otherwise need to be financed out of private sources. If domestic savings are insufficient, receiving aid could relieve crucial bottleneck in the course of economic growth and development (Szirmai, 2005). This way aid can ‘free’ capital for other purposes. Zafar (2007) argues that aid coming from China is bringing the necessary capital to Sub-Saharan Africa. These capital investments finance infrastructure projects such as the building of dams, ports and roads which stimulates economic growth. According to Dollar and Prichet (1998) economic growth stimulates progress in health, education and nutrition which is the key to poverty reduction. However, according to Boone (1995) poverty is not caused by a shortage of capital and will therefore not solve the underlying problem. In his research he finds that aid does not increase investment or growth, and does not reduce poverty measured by improvements in human development indicators. Furthermore, according to displacement theories an increase in aid does not necessarily result in a similar amount of investment, and therefore might not lead to as much investment growth as we would expect.
9
It is also important to consider what kind of aid China is contributing and whether it serves specific objectives. In this area of research there are different views on what these incentives might be. Zafar (2007) argues in his paper that aid is targeted at facilitating trade and getting access to natural resources. Whereas Brautigam (2011) argues that Chinese aid follows diplomatic ties. Brautigam identified that countries with which Beijing has diplomatic ties receive much more aid then countries that do not. And official development aid (ODA) does not appear to be given in larger amounts to resource rich countries. Also Brautigam (2011) argues that most of Chinese finance is not official development assistance. The majority of the share is used to foster Chinese investments, like export credits and non-concessional state loans, which are not considered to be ODA. The OECD defines aid as; “the concessional funding given to developing countries and to multilateral institutions primarily for the purpose of promoting welfare and economic development in the recipient country” (OECD 2008). Because the Chinese objectives stay rather vague it is interesting to research whether Chinese aid flows positively contribute to human development. Since aid effectiveness is largely dependent on reaching ‘the right hands’ (Easterly 2006).
Trade
10
manufacturing, hence industrialisation. However, when industrial development is paired with high exchange rates this process might be retained. Additionally, it is important to review how the large amounts of capital flowing in from the import of natural resources by China are spent. Because it could lead to rent-seeking and corruption, and in this case the benefits of increased capital due to trade are diminished. Also countries that are not natural resource abundant but do require raw material inputs loose income because of higher prices (Zafar, 2007).
Increased trade between China and SSA means that African consumers have access to low cost manufacturing products from China. Low cost of consumption can have a positive effect on income since it could reduce the income share spend on consumption goods and consequently leave more money available for education or healthcare. On the other hand, Chinese activities in SSA might propose a threat as Chinese manufacturing firms could replace African competitors (Busse, Erdogan and Mühlen ,2014). The exchange rate overvaluation arising from large export of natural resources, and the low cost Chinese production poses another threat to African companies. This could lead to lower income in hands of African manufacturing companies which has a negative effect on human development.
FDI
11
The case of Chinese FDI flowing to SSA might be different. Gohou & Soumaré (2011) find that when FDI is used as a tool to get access to raw materials for a firm outside the host country than the scope for job creation and spill-overs is limited. Since Chinese FDI is often aimed at getting access to natural resources for production in the home country this might be the case. Most investments come from Chinese state owned companies and banks who try to supply the high demand of raw materials (Zafar, 2007). Also some of these companies bring Chinese workers to the African continent to work there, which means the positive effect on employment of local workers is limited.
Natural resources
As previously mentioned China has a high need for natural resources to be able to satisfy demands of industrialization and carry on current levels of production. However their domestic factor endowments do not allow for such production which makes China dependent on importing those resources. In this respect Africa is a very important partner for China. When China announced its “going-global” policy in 2001 Chinese ties with Africa were tightened (UNCTAD, 2007). The Chinese government tried to stimulate investment by giving firms access to loans, foreign currencies, beneficial tax rates and other policies to stimulate imports and exports (UNCTAD, 2007). These policy measures are not only targeted at Africa, but they show how the Chinese government encourages international involvement and stimulates FDI, including projects to extract resources which are relevant in this research. The relationship between Chinese natural resources extraction and FDI, aid and trade flows is therefore important to consider.
12
Zafar (2007) shows that trade flows between China and Africa follow a Heckscher-Ohlin model prediction, where Chinese firms import resource scare natural resources and exports labor-intensive manufactures and high tech products. Also their Foreign Direct Investment is directed at countries with high levels of raw materials. Many researches have therefore implied that Chinese investments and economic cooperation projects serve a greater purpose: getting access to the much needed natural resources. Chinese governments supposedly apply a strategy where financial assistance and funding of construction projects are directed at gaining influence on oil distribution and creating a network of reliable allies and suppliers. By investing in the development of oil production in regions where they lack capital and technology themselves they try to avoid buying oil at market prices (Forney, 2004).
This does not however mean that Chinese influence in Africa is a negative development in terms of welfare and economic growth. It could however mean that regions which are natural resource rich benefit more from Chinese involvement. By making a distinction between natural resources imports and export I will show whether this specific part of Chinese-African trade flows influences development and economic growth.
Growth economics
For my econometric model I use a Solow type growth model to estimate the effect of Chinese aid, trade and FDI flows on economic growth and human development in Africa. Described by Szirmai (2005) this neoclassical theory predicts that when markets function correctly and labour and capital can move freely within the economy, rich and poor countries will converge and developing countries will catch up. Therefore we would expect that an increase in capital due to Chinese investments in Africa in the form of aid, FDI or trade will lead to higher growth in African countries. The model works as follows;
13
shift the production function upwards and create large gains in terms labour productivity. In this model technology is assumed to be freely available for all countries. Therefore, this model predicts that per capita income of different countries will converge (Szirmai, 2005).
14
3. Methodology and data
Model
In order to assess the impact of the Chinese activities on economic growth and human development, I use a Solow-type growth model. My research is based on the framework by Busse, Erdogan and Mühlen (2013), and I add human development as an alternative dependent variable. Due to a lack of alternative models to predict human development I assume that Solow’s growth model is also able to predict the HDI. This neo-classical model predicts that countries with low per-capita incomes grow faster than those with high GDP per capita, so that over time per-capita incomes converge. In the case of human development I will assume that countries with a low score on the Human Development Index grow faster than countries with high HDI scores and converge in terms of welfare over time. Therefore high income or HDI levels in the previous period have a negative impact on the upcoming period due to the convergence effect.
When constructing my econometric model I use panel data which allows me to examine several countries (N) over time (T). With regard to the methodology I use a standard OLS fixed effects model, in order to account for unobserved constant country fixed effects. A benefit of this approach is that any omitted variable that is constant over time does not bias the estimates, also when it is correlated with the explanatory variables (Durlauf, Johnson and Temple, 2005). I expect that a fixed effects model is suited because the sub-Saharan African sample is not random. However, in order to assess whether a fixed effects model is indeed appropriate I perform a Hausman test for each regression model3. Results indicate that the fixed effects model is the most appropriate for both the human development and economic growth model.
I use both the Human Development Index and data on GDP as my dependent variables. In this model changes in the log HDI (HDI) or the log real GDP (y) in country i over time t are measured. This is in accordance with literature written by Islam (1995) and Durlauf et al. (2005). I have chosen to also use a log for HDI because it allows me to compare my results to the economic growth model outcomes. Furthermore, this way the results of both models can be interpreted as elasticities because they are log-log models. Additionally, I include a lagged version of the dependent variable in each model as explanatory variable. I assume that the values of GDP and HDI are correlated with their value from the previous year.
3
15
For all explanatory variables logs are taken except for changes in the terms-of-trade and the inflation indicator. I use log transformation and robust standard errors in order to account for heteroskedasticity. Using log variables also allows me to compare my results to the outcomes of a similar research by Busse et al. (2013). I do not use a log for terms-of-trade because it is an index. Furthermore I do not use a log for inflation, as opposed to Busse et al (2013) who do take a log, because there are too many negative observations in my dataset. This is a problem because the log is only defined for positive numbers. The basic models read as follows:
𝐻𝐷𝐼𝑖𝑡 = 𝛼 + (𝛽 + 1) 𝐻𝐷𝐼𝑖𝑡−1+ 𝛾 𝑙𝑛𝑠𝑖𝑡+ 𝜑 ln(𝑛𝑖𝑡+ 𝑔 + 𝛿) + 𝜑′𝑙𝑛𝑋𝑖𝑡+ 𝜆𝑡+ 𝜇𝑖+ 𝜀𝑖𝑡 (1)
𝑙𝑛𝑦𝑖𝑡 = 𝛼 + (𝛽 + 1) 𝑙𝑛𝑦𝑖𝑡−1+ 𝛾 𝑙𝑛𝑠𝑖𝑡+ 𝜑 ln(𝑛𝑖𝑡+ 𝑔 + 𝛿) + 𝜑′𝑙𝑛𝑋𝑖𝑡+ 𝜆𝑡+ 𝜇𝑖+ 𝜀𝑖𝑡 (2) Explanatory variables in this model are the initial HDI and income values, the population growth rate n, changes is technology g, the depreciation rate of the capital stock δ, and the savings rate s. Time specific effects that influence the entire country set are also included with λt. Furthermore, control variables are taken into account by considering the country specific effects μi, and the error term εit . The independent variables of interest; aid, trade and FDI flows, are represented by Xit. Also the additional control variables changes in terms-of-trade, inflation rates, health and the education indicators are included here. In order to be able to use the Solow model to predict human development, I add variables for health and education. In the standard model these are not included but because they may have an impact on human development I add them to the model. I expect that countries which have higher levels of education and life expectancy will grow slower than countries that have low values in these aspects. According to Mankiw et al. (1992) I can assume g and δ to be constant over time at 0.05. Mankiw et al. (1992) argue that g represents the changes is technology which is not country specific, and that the depreciation rate (δ) will not vary much between countries.
Endogeneity
16
variable. This model deals with endogeneity problems and provides extra information about unknown variables and deals with heteroskedasticity issues. The explanatory variables are transformed into lagged variable, which means that they are considered as endogenous. Using GMM estimators does however have some disadvantages. Because when autocorrelation is present endogeneity might not dissolve when using GMM. Because I include a lagged dependent variable in my model autocorrelation could be of consern. Roodman (2009) points out that using GMM can be a great risk because the system is very complex and instruments might appear valid but in reality are not. Using the GMM correctly is complex and could lead to misinterpretation of the results. Additionally, Busse et al. (2013) find that transforming the variables to endogenous with the GMM hardly affected their main results. In the case of aid an explanation could be that Chinese aid is not meant to serve the objective of reducing poverty. According the Brautigam (2011) and Zafar (2007) aid flows are directed at serving larger political objectives or getting access to natural resources. If Chinese FDI and aid is indeed directed at countries that are natural resource rich, this is exogenous. Considering the complexity and appropriateness of the GMM and I have chosen not to add it to my research and solely use a fixed effects model.
Data
A list of all variables and their sources is included in appendix A. In this section I will further explain how the data of this research is collected and which variables are used.
Data collection
Because data on Chinese FDI and economic cooperation before 1990 is not available the scope of my research is limited; therefore I review data from 1990 to 2013 in this thesis. My data set consists of data on 42 countries in Sub-Saharan Africa. North African countries are not considered in this research. Sub-Saharan African countries that were excluded are Somalia, Liberia, Soa Tome and Principe, Seychelles, South-Sudan and Djibouti due to a lack of available data.
17
might actually include flows that should be considered as FDI. However in the absence of superior data the flaws in the statistical methodology need to be considered when interpreting the results.
Dependent variables
As my first dependent variable I use the Human Development Index from the United Nations Development Programme (UNDP) as the measure of welfare. Because my research is focused on the development of welfare, I need a measurement that does not only consider income but also considers the level of healthcare and education in a country. The Human Development Index offers an alternative to GDP per capita which does not consider social wellbeing. It is a summary statistic that assesses a country’s average score in three dimensions: life expectancy at birth (health), mean years of schooling and expected years of schooling (education), and Gross National Income (GNI). The HDI, while imperfect, is the most universal measure of a country’s human development. The measurement is imperfect because factors that influence human development are much broader than what is captured in the index. For example, it does not reflect political participation or gender inequalities (Human Development Reports, 2015). This means that the index can only serve as a representative for some issues of human development. Nevertheless it has often been used in existing literature that measured the effect of FDI on welfare (Gohou and Soumaré, 2012; Sharma and Gani, 2004). Data was received on request from the United Nations Development Programme. A large advantage of this is that all statistics are calculated with the same methodology. Statistics were available for the years 1990-2013, although not for all years and countries.
In order to compare my result with the existing literature and to be able to draw broader conclusions on the effect of Chinese activities I use data on GDP as a second dependent variable. This is in line with Busse et al. (2014), Islam (1995) and Durlauf et al. (2005) who use real GDP per capita as a dependent variable. It is important to consider both outcomes in order to justify how a country can have a high GDP per capita but a low HDI. Differences can be caused by lower life expectancy or educational attainment in a country while having a relatively high GDP per capita. It is interesting to see whether China positively or negatively influences human development by means of aid, trade and FDI and what different effect it has on economic growth.
Independent variables
Aid
18
Africa existed before 1998, data has only been published since then. I have chosen not to extrapolate or interpolate the missing data between 1990 and 1998 because figures for such a long period are not suited to be estimated.
In line with Sanfillippo (2010) and Busse, Erdogan and Muhlen (2014) I use economic cooperation data as an alternative to Official Development Assistance data. According to Brautigam (2011) data on economic cooperation is not the same as ODA, and should not be used as a measure for aid. However, due to a lack of available data on aid flows between China and each African country separately I will use it as a proxy. Brautigam argues that Chinese economic cooperation data illustrates where Chinese companies are getting contracts and do not represent ordinary ODA flows, which also encounter concessional loans and grants. Chinese economic cooperation figures on the other hand represent the turnover generated in projects undertaken by Chinese contractors (NBS, 2015). Brautigam argues that Chinese economic cooperation data might actually represent money flows that are considered as FDI by others. Particularly in the case of mining and infrastructural projects, economic cooperation and FDI are hard to separate (Ajakaiye et al, 2008). In order to account for this issue I will look at both FDI and aid separately but also perform a regression analysis where they are summed together. Data on aid flows from the world are taken from the World Bank and measured as flows of ODA. Data was available for most of the observations.
Trade
All data is collected from the United Nations Commodity Trade Statistics Database. Firstly, I use data on African exports to, and imports from, China. Also data on import and export with the world is collected. Afterwards Chinese trade figures are deducted from the trade with the world row in order to not account for Chinese trade flows double. I make a distinction between imports and exports because I want to identify the effect that each separate trade flow has on welfare and the economy. This data is afterwards normalized by the host country’s GDP.
In order to assess the impact of trade in natural resources between China and Africa I have collected data on natural resource and non-natural resource imports and exports. Unfortunately this specific data was only available for a limited amount of countries and years. But because natural resources are an important part of African trade, I want to assess the effects of trade in these specific commodities. Following Busse, Erdogan and Muhlen (2014) natural resources are defined as mineral fuels, oils, distillation products, inorganic chemicals, precious metals and isotopes4. I make this distinction because these specific natural resources are highly valuable and therefore likely to impact welfare and economic growth. Therefore I am interested if exporting these resources to China
4
19
positively influences welfare and GDP accordingly. All trade variables are again normalized by host country’s GDP.
FDI
When measuring FDI I differentiate between FDI coming from China and FDI from the world, and again normalize all FDI variables by the GDP of the host country. Total FDI flows from the world are taken from UNCTADstat (2014) and are measured as the net inflows of investment to acquire a lasting management interest in an enterprise operating in an economy other than that of the investor (World Bank, 2001). Collecting FDI data from Chinese sources was however more difficult and only available from 2003 and onwards. Chinese outward FDI data is collected from the IMF-OECD reports format, available since 2003. Again I have chosen not to extrapolate data before 2003 because of the large amount of data that in that case has to be estimated.
Control variables
To assess the impact of Chinese activities in Africa on both economic and human development, I use the Solow growth model. To empirically test this model I use several indicators and control variables. Data on GDP is used to estimate income levels and the share of investment as a measurement for the savings rate. As discussed before the depreciation rate of capital and changes in technology are assumed to be constant over time and equal to 0.05 in accordance with Mankiw et al. (1992). In order to control for other factors that could influence the dependent variables I include control variables which I will explain in more detail below.
Changes in terms of trade, inflation rates and population growth
20
Therefore I predict that a positive population growth rate negatively impacts human and economic development.
Education
Education is considered in the Human Development Index as mean years of schooling and expected years of schooling. I include data on the general government expenditure on education as a
percentage of GDP to control for the positive effect spending on education should have on human development. Because this data was not available for all countries and years, I have interpolated the data following a straight-line annual progression in order not to lose too many observation
Health
21
4. Empirical results
Descriptive statistics
My sample consists of 42 sub Saharan African countries for the period of 1990 until 2013. A list of all countries and their abbreviations are included in appendix B. In table 1 you can find the descriptive statistics of all variables. Unfortunately the amount of observations for the resource specific variables is limited. In my main regression analyses these variables will not be included. My primary indicators have at least 245 observations. The Human Development Index data was received directly from the United Nations Development Programme on my request. This means that the data received is from a single source with only one approach of calculation. This is important because the determinants on education and income changed in 2010. The benefit of using this data is that no variations in the effect measured can be caused by differences in the HDI measurement. This does however mean that not all data was available for each country and year.
Table 1: Statistics for sub Saharan African countries, 1990-2013
(1) (2) (3) (4) (5)
VARIABLES N mean Std Dev min max
Ln HDI 868 -0.859 0.263 -1.645 -0.260
Ln GDP (current US$) 1,006 6.438 1.068 4.631 10.06
Changes in Terms-of-trade (2000=100) 994 110.1 37.62 21.22 280.7
Ln investment, % of GDP 946 2.924 0.575 -1.228 5.389
Ln population growth 997 0.852 0.482 -2.207 2.328
Inflation, annual change in % 1,008 62.85 882.6 -31.57 26,766
Ln governmental expenditure on education, % of GDP 442 1.313 0.524 -0.372 3.792
Ln mortality rate 1,008 4.737 0.520 2.660 5.791 Ln prevalence of HIV 960 0.982 1.327 -2.303 3.357 Ln Aid world, % of GDP 959 1.853 1.260 -2.253 4.464 Ln Aid China, % of GDP 529 0.0489 1.723 -6.879 3.342 Ln FDI world, % of GDP 998 2.635 1.570 -13.63 5.577 Ln FDI China, % of GDP 252 -2.101 1.547 -5.519 2.023 Ln exports to ROW, % of GDP 946 3.198 0.702 0.725 5.055 Ln exports China, % of GDP 245 -2.634 4.877 -18.27 3.974 Ln imports to ROW, % of GDP 581 -0.387 3.173 -17.76 3.130 Ln imports China, % of GDP 946 3.532 0.586 1.555 5.288
Ln resource exports to ROW, % of GDP 124 -0.298 2.375 -12.49 4.090
Ln resource exports to China, % of GDP 94 -7.918 5.597 -17.76 2.392
Ln non-resource exports to ROW, % of GDP 944 3.171 0.691 0.723 5.055
Ln non-resource exports to China, % of GDP 245 -2.660 4.788 -17.76 3.974
Ln resource imports from ROW, % of GDP 133 1.134 1.844 -3.321 4.152
Ln resource imports from China, % of GDP 133 2.000 0.970 -1.859 3.504
Ln non-resource imports from ROW, % of GDP 950 3.512 0.604 1.555 5.288
Ln non-resource imports from China, % of GDP 483 -0.824 3.321 -17.76 2.763
Note: the statistics are calculated for the sample of 42 sub-Saharan African countries. HDI data received on request from the United Nations Development Programme.
22
mentioned African and Chinese data can be unreliable. This enhances the risk of outliers which may unjustifiably influence the outcomes of the model. In order to prevent this I have deleted the observations below the first and above the 99th percentile. This way I limit the possibility of outliers driving the results. The summary statistics without outliers are given in table 2.
Table 2: Statistics for sub Saharan African countries without outliers, 1990-2013
(1) (2) (3) (4) (5)
VARIABLES N mean sd min max
Ln HDI 852 -0.858 0.250 -1.528 -0.326 Ln GDP (current US$) 986 22.26 1.261 19.28 26.33 Changes in Terms-of-trade (2000=100) 976 109.5 34.23 30.44 235.4 Ln investment, % of GDP 928 2.928 0.479 1.286 4.688 Ln population growth 978 0.868 0.400 -0.994 1.524 Ln Battle Deaths 234 5,35 1,927 0 10,81
Ln inflation, annual change in % 988,00 15,10 36,87 -13,93 556,94
Ln governmental expenditure on education, % of GDP 433 1.314 0.473 0.130 2.651
Ln mortality rate 988 4.748 0.475 2.773 5.576 Ln prevalence of HIV 929 1.037 1.227 -1.609 3.307 Ln Aid world, % of GDP 940 1.872 1.199 -1.446 3.734 Ln Aid China, % of GDP 519 0.0801 1.599 -5.588 2.911 Ln FDI world, % of GDP 979 2.704 1.119 -1.014 5.105 Ln FDI China, % of GDP 248 -2.106 1.488 -5.223 1.498 Ln investments China, % of GDP 928 3.200 0.658 1.475 4.762 Ln exports to ROW, % of GDP 240 -2.498 4.581 -17.24 3.859 Ln exports China, % of GDP 928 3.533 0.553 2.135 4.822 Ln imports to ROW, % of GDP 571 -0.266 2.769 -16.46 2.459 Ln imports China, % of GDP 122 -0.234 2.084 -5.238 3.766
Ln resource exports to ROW, % of GDP 94 -7.918 5.597 -17.76 2.392
Ln resource exports to China, % of GDP 926 3.173 0.647 1.475 4.738
Ln non-resource exports to ROW, % of GDP 241 -2.590 4.589 -17.24 3.859
Ln non-resource exports to China, % of GDP 131 1.145 1.797 -3.163 4.098
Ln resource imports from ROW, % of GDP 130 2.006 0.902 -1.837 3.405
Ln resource imports from China, % of GDP 931 3.514 0.568 2.135 4.910
Ln non-resource imports from ROW, % of GDP 475 -0.711 2.968 -16.50 2.006
Ln non-resource imports from China, % of GDP 242 0.697 1.224 -2.848 2.982
Note: the statistics are calculated for the sample of 42 sub-Saharan African countries. HDI data received on request from the United Nations Development Programme.
Closer look at the data
23
In graph 1 we can also see how flows of aid, trade and FDI have developed in the years 1990 to 2013. Firstly, we can see that especially the amount of imports from China have increased over the years. This accelerated around 2001 when China entered the World Trade Organization and opened up for international trade and investment. We see that also the aid flows coming from China have substantially increased in value since this period. The steady trend in aid from China between 2000 and 2003 is caused by the missing data for the intervening years. Exports have also increased, however less rapidly. The figures also show how FDI flows have increased much more slowly. However, this could also be explained by the different Chinese measurement of aid and FDI. Therefore I have created an additional investment flow in the graph that indicates the percentage growth of aid and FDI summed together. Since FDI data was only available from 2003 and onwards, this is the also the starting point in this figure.
24
Graph 1: Average growth of total African HDI, aid, trade and FDI flows (1990-2013)
Notes: the figures refer to the average total African % growth in total aid, FDI, exports and imports from China (right hand vertical axis). Calculated by dividing total African flows by total African GDP. The average total African HDI levels are indicated on the left hand vertical axis. Trade data was collected from the UN Comtrade. HDI data was collected from the United Nations Development Programme.
Graph 3 shows us the average growth rates in GDP of all African countries in the sample. This indicates that on average African countries have experienced positive growth rates between 1990 and 2013. The figure also shows us how these growth rates have been relatively stable since 2000. This is unique for Africa which has always dealt with fluctuating economic growth. But these figures show that the growth rates have been relatively high and stable for over a decade. When we compare this graph to graph 1 we see that the positive period of GDP growth is taking place in a similar time frame as the increased involvement with China. It will be interesting to see whether to two can be linked and what the role natural resources is.
0 0,005 0,01 0,015 0,02 0,025 0,03 0 0,1 0,2 0,3 0,4 0,5 0,6 ave rage S SA ch an ge in % Av e rage S SA HD I valu e year
average African HDI level Total aid from China in % of GDP
Total FDI from China in % of GDP Total exports to China in % of GDP
25 Graph 2: HDI growth in % per country (1990-2013)
Notes: data was provided by the United Nations Development Programme. Abbreviations can be found in appendix B. Graph 3: Average total African GDP growth in %
Notes: The figure refers to an average GDP growth in Africa, calculated by taking an average of the annual GDP growth of the 42 African sample countries each year. Data was collected from the World Bank (2015)
-10% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% AG O BE N BWA BF A BD I CM R CPV CA F
TCD COG ZAR CIV GNQ ERI ETH GAB GMB GH
A G IN G N B KE N
LSO MDG MWI MLI MRT MU
S MOZ NAM N ER N G A RWA SE N
SLE ZAF SDN SWZ TZA TGO UGA ZMB ZWE
HDI growth in % per country
0 1 2 3 4 5 6 7 8 9 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13
26
5. Results
Estimation of the human development model
In this section I will look at the effect of trade, aid and FDI on human development in Africa, and assess whether these flows negatively or positively affect the HDI by running several fixed effect regressions. In column 1 of table 3 (page 32) we see the baseline specification of my model, using only the Solow model variables. As predicted the health variable mortality is negative and significant. Showing that 1 percent increase in mortality rates negatively influences human development in African countries by 0.353 percent. This is as expected because increased mortality rates indicate lower life expectancy which is one of the indicators of the HDI. The HIV prevalence indicator is also significant but positive. This is strange because this would indicate that the prevalence of HIV positively influences human development. The education and investment variables are both positive, but not significant. The terms of trade indicator shows a negative effect, however it is insignificant. The estimate on population growth is positive but also insignificant. Contrary to the Solow theory the inflation estimate is positive and significant but very small. The lagged dependent variable is positive and significant with a coefficient of 0.258. The R-squared is relatively high at 0.781. Overall the Solow growth indicators do not show the expected results. This indicates that the model might not fit optimally when used to predict human development.
27
In column three we see the results of adding the aid variables from China and the world to the model. In this regression more observations are measured but the adjusted R-squared slightly decreased. The results indicate that aid from China and the world both have a positive and significant effect on the human development in the sample countries. These results are in line with previous research done by Gomanee et al. (2003) and Ishfaq (2004), who also find a positive and significant relationship between aid and human development in general. However, my research indicates that Chinese aid specifically also has a positive and significant effect, which is a new and rather controversial finding. Also the effect of Chinese aid flows seems to be larger than the effect of Chinese FDI. Zafar (2007) argues that Chinese aid is targeted at facilitating trade and getting access to resources, and Brautigam (2011) argues that aid follows diplomatic ties and does not serve the incentive of poverty reduction. The motivations behind the Chinese aid flows might remain unclear in my research, but my results do indicate that Chinese aid flows positively affect human welfare in African countries. These results indicate that an increase of aid from China is as good as an increase of aid from other countries despite its incentive. This is in line with the findings of Zafar (2007) who finds that China is bringing the much needed capital to finance infrastructural projects. This answers my second research question. This positive effect could be explained by the increased amount of capital flowing into these countries which seemingly reduces poverty and finances projects that increase welfare.
In the fourth column I measure the influence of the trade flows from China and the world. This indicates positive and significant influence of imports from China and the rest of the world on human development in Africa. The number of observations has decreased, but the adjusted R-squared has increased to a value of 0.783. The results imply that the access to Chinese products has impacted human development positively in the sample countries. An explanation for this could be the access to low cost manufactures from China which could reduce the part of income spent on consumption, and possibly leave more income available for other things like education and healthcare. The coefficient for exports to China is also positive and significant. The results show that the exports to China positively influence human development in African countries. Similar findings were obtained by Davies & Quinlivan (2006). They argue that increased amount of African exports could lead to a higher income which in time could lead to better health care and education. In this way having China as an important export partner could positively affect human development, answering my third research question.
28
minimize the discussion on the Chinese measurement issue, a positive relationship can be established with human development. However, the estimates that measure the FDI and aid flows from the world are no longer significant.
Estimation of the economic growth model
In table 4 (page 33) we see a similar regression table as in table 3. I have estimated a similar model with a different dependent variable; GDP. This regression analysis will show what impact the increased Chinese involvement had on economic growth in Africa. I will compare the results to the regression outcomes of the human development model and to similar research done by Busse et al. (2014).
In the first column of table 4 we can see how the basic Solow model fits relatively well. The results show that investment and terms of trade have a positive effect on GDP. Especially the effect of investment is large, where a 1 percent increase of investments results in 0.248 percent increase in GDP. Additionally the results show that inflation has a negative and significant impact of GDP as expected. Population growth also shows a negative coefficient however it is not significant. This is contrary to the literature by Mankiw (1992), Islam (1995), who did find a significantly negative effect of population growth. The lagged dependent variable is positive and highly significant. The adjusted R-squared is relatively low at 0.326, and therefore I add more variables to the model.
29
already only available from 2003 and onwards I chose not to include it5. My results on this specific sample indicate that Chinese aid and FDI flows do have a positive and significant effect on economic growth in African countries.
In the fourth column of table 4 we consider the trade measures. These estimates show us that all trade flows positively and significantly influence economic growth rates in Africa except for imports from the world. The number of observations and countries considered however decreases when we add the different estimates, since data was not available for all countries and years. These results are different from the results found by Busse et al. (2014), who find that imports from China have a negative effect. My results however indicate a strongly significant effect of import of Chinese products on economic growth. Also the coefficient is relatively high compared to the other results. This is interesting and might be explained by the positive effect the import of Chinese cheap manufactures have on income. Also exporting to China and the rest of the world seems to positively affect economic growth in Africa. This is as expected since exporting goods and services will lead to an increased income.
Overall table 4 shows how aid, FDI and trade flows from China and the world positively impact economic growth in Africa, except for imports from the world which is not significant. This therefore answers my fourth research question by showing a positive relationship between the flows in question and economic development in African countries. When we compare the results of table 3 and 4 we see similar outcomes in terms of the positive effect of FDI, aid and trade flows on African human and economic development. However, by the differences in size of the coefficients we see how these flows seem to have a larger impact on economic growth then they do on human development. When we compare elasticity results we see that a one percent increase in FDI from China leads to a 0.0767 percent higher GDP and to a 0.0138 percent higher HDI value. Especially for aid the difference between the effect on economic growth and human development is large. Where a one percent increase in Chinese aid increases GDP by 0.208 percent it only increases human development by 0.0273. This is strange because aid is usually target at increasing welfare more than the economy. However, this could again indicate how Chinese aid is different from other sources of aid. Furthermore the imports from China seem to have a large effect on GDP in African countries at 0.228.
5
30
Considering natural resources
In order to gain some in-depth knowledge on the effect of trade in resources with China, I have split up the export data in the trade of resources and non-resources. This is especially interesting when looking at Africa and China because the underlying incentive and positive effect on economic and human development is often questioned. The results of the regression analysis that include estimates on natural resources are included in table 5 (page 34). Due to the limited amount of data that was available on resource specific trade flows, these regressions only have a relatively small number of observations and are therefore run separately. Also I use only one control variable in the regressions namely the mortality rate indicator, in order to be able to perform regression analysis with the limited amount of observations. The models below are tested with very limited data and with only one control variable the results therefore need to be interpreted with caution. However, they do show some interesting results.
31
GDP as a dependent variable. The results in column 3 show that resource exports to China and the rest of the word negatively affect economic growth it African countries. Also non-resource exports to the rest of the world show a negative and significant coefficient. Non-resource imports from China positively and significantly affect GDP in African countries.
32
Table 3: panel regression results for the impact FDI, aid and trade on HDI in Africa, 1990-2013
(1) (2) (3) (4) (5)
VARIABLES ln HDI ln HDI ln HDI ln HDI ln HDI
ln investment 0.01000 0.0160 0.0447*** 0.0298 0.0182*
(0.0137) (0.0102) (0.0147) (0.0245) (0.00925)
ln population growth 0.00419 0.00285 -0.0218 0.0407*** 0.00835
(0.0172) (0.0162) (0.0263) (0.0126) (0.0147)
Terms-of-trade growth -5.35e-05 0.000539*** 0.000672** -7.25e-05 0.000371*
(0.000205) (0.000170) (0.000260) (0.000303) (0.000188) Inflation 1.91e-06*** 0.000142 0.000312*** -0.000235 0.000112 (5.05e-07) (0.000172) (9.11e-05) (0.000176) (0.000149) ln governmental expenditure on education 0.0158 (0.0197) ln mortality -0.353*** (0.0331) ln HIV prevalence 0.0223** (0.0109) ln FDI world 0.0245*** 0.0134 (0.00772) (0.00974) ln FDI China 0.0138*** (0.00420) ln Aid world 0.0202*** 0.00700 (0.00728) (0.00526) ln Aid China 0.0273*** (0.00657) ln exports ROW 0.0132 (0.0283) ln exports China 0.00979*** (0.00288) ln imports ROW 0.0536** (0.0209) ln imports China 0.0294*** (0.00526) ln investments China 0.0236*** (0.00655)
lagged dependent variable 0.258* 1.935*** 2.246*** 1.858*** 1.980***
(-0.138) (0.195) (0.131) (0.181) (0.221)
Observations 759 237 423 194 231
R-squared 0.781 0.555 0.515 0.783 0.564
Number of C 38 31 41 30 31
33
Table 4: panel regression results for the impact FDI, aid and trade on GDP in Africa, 1990-2013
(1) (2) (3) (4) (5) VARIABLES ln GDP ln GDP ln GDP ln GDP ln GDP ln investment 0.248* 0.101 -0.0210 -0.122 0.0711 (0.127) (0.0865) (0.0985) (0.0748) (0.0813) ln population growth -0.209 -0.0625 -0.0863 0.262*** -0.000490 (0.154) (0.111) (0.126) (0.0837) (0.101) Terms-of-trade growth 0.0104*** 0.00687*** 0.00808*** 0.00131 0.00563*** (0.00281) (0.00123) (0.00138) (0.00165) (0.00133)
inflation -3.34e-05** 0.00313** -3.77e-05 0.00178 0.00315***
(1.56e-05) (0.00116) (0.000522) (0.00134) (0.000998) ln FDI world 0.206*** 0.130** (0.0469) (0.0573) ln FDI China 0.0767*** (0.0205) ln Aid world 0.110** 0.0242 (0.0524) (0.0275) ln Aid China 0.208*** (0.0361) ln exports ROW 0.251** (0.114) ln exports China 0.0428*** (0.0138) ln imports ROW 0.206 (0.151) ln imports China 0.228*** (0.0258) ln investments China 0.161*** (0.0367)
lagged dependent variable 0.614*** 0.767*** 0.893*** 0.734*** 0.805***
(0.0799) (0.0399) (0.0223) (0.0676) (0.0440)
Observations 929 241 454 203 235
R-squared 0.326 0.655 0.642 0.802 0.687
Number of C 42 31 42 30 31
34
Table 5: panel regression results of trade in resources on human and economic development in Africa, 1990-2013
(1) (2) (3) (4)
VARIABLES ln HDI ln HDI ln GDP ln GDP
ln mortality -0.0825*** -0.137*** -0.921*** -1.369***
(0.0176) (0.0103) (0.283) (0.378)
ln resource exports ROW -0.00295 -0.0437**
(0.00255) (0.0202)
ln resource exports China -0.00103*** -0.00640*
(0.000330) (0.00349)
ln non-resource exports ROW -0.00262*** -0.0304**
(0.000821) (0.0134)
ln non-resource exports China 0.000654 0.0305
(0.00544) (0.0536)
ln resource imports ROW -0.00810** 0.123
(0.00364) (0.132)
ln resource imports China -0.00224*** 0.0212
(0.000635) (0.0211)
ln non-resource imports ROW 0.00507 0.0606
(0.00337) (0.125)
ln non-resource imports China 0.00381** 0.0830*
(0.00150) (0.0438)
lagged dependent variable 0.728*** 0.607*** 0.574*** 0.724***
(0.0545) (0.0548) (0.0981) (0.0614)
Observations 64 112 64 121
R-squared 0.823 0.958 0.644 0.776
Number of C 20 17 20 17
35
6. Limitations and Future research
The main limitation of this research relates to data. Data for FDI flows from China were only available from 2003 and onwards. Furthermore, data was often unavailable for all countries and years. Also the quality of African and Chinese data can be questioned, which makes it difficult to fully rely on the outcomes of any research performed with this kind of data. Future research could extend this research by spending even more attention on collecting data on natural resources. By doing so the number of observations in the regression analyses could be improved, and this allows for a more reliable assessment of the effect of trade in natural resources with China. Furthermore, data before 1990 is limited but might be available in the future. Extending this research to a larger time frame could be interesting because especially in the case of human development progress which moves rather slowly. Also it would be interesting to separate the results for different groups of countries, in order to determine if the results differ per country. This would allow us to draw further conclusions on potential differences of the effect of Chinese trade with natural resources rich and scarce countries. Furthermore, it would be interesting to see a similar research on Chinese activities in Latin America. And assess whether Chinese aid, trade and FDI also positively affect human development in Latin American countries since Chinese activities have also intensified in this region.
36
7. Conclusions
In this thesis I have investigated the effect of Chinese aid, trade and FDI flows on human development and economic growth in African countries. Previous research has not focused on the combination of these three Chinese activities. Also human development and economic growth have not been considered as dependent variables in a single research study. My empirical findings can be summarized as follows. As for human development, aid and FDI have a significantly positive impact human development in African countries. Imports from China and the rest of the world also have a significant impact on welfare, and exports to China significantly contribute to human development. As for economic growth, we see similar results. All flows of FDI, aid imports and exports to the world and China significantly contribute to economic growth. However, the outcomes for the economic growth model are much larger than for the human development model. Indicating that Chinese involvement in Africa has a larger impact on economic development. When I separate the effect of trade in natural resources and non-natural resources I find that resource exports to China negatively affects human development. Also non-resource imports from China positively affect economic growth and human development in African countries.
37
affect human welfare. Policies should therefore welcome these kinds of investments, but be aware of how much is directed at human development.
38
References
Adkins and Hill, 2012. ‘Using STATA for principles of econometrics’, 4th edition.
Ajakaiye, O., Mwega, F.M. & N’Zue, F.F. 2008. Seizing Opportunities and Coping with Challenges of China – Africa Aid Relations: Insights from AERC Scoping Studies. African Economic Research Consortium. Policy Issues Paper No. 1.
Asongu Simplice, A., 2012. ‘Institutional benchmarking of foreign aid effectiveness in Africa’. Derived online at: http://mpra.ub.uni-muenchen.de/38095/ MPRA Paper No. 38095, posted 13.
Auty, R., 1994. ‘industrial policy reform in six large newly industrializing countries: the resource curse thesis.’. World development, Elsevier, January 1994.
Barro, R.J. & Sala-i-Martin, X. (2004). ‘Economic Growth.’. MIT Press, Cambridge, USA and London, UK.
Ben-David, D., Nordström, H., Winters, L.A. (1999). ‘Trade, income disparity and poverty’. World Trade Organization, special studies 5.
Boone, P. 1995. ‘Politics and the effectiveness of Foreign Aid’, Working Paper Series No. 5308.
Borensztein, E., J. De Gregorio, and J. Lee. 1998 “How Does Foreign Direct Investment Affect Economic Growth.” Journal of International Economics 45(1998):115–35.
Brautigam, D. 2011. "Aid 'With Chinese Characteristics': Chinese Aid and Development Finance Meet the OECD-DAC Regime," in Journal of International Development, v. 23, n. 5, July 2011.
Brautigam, D. 2011. "Chinese Development Aid in Africa: What, Where, Why and How Much?" China Update 2011, eds. Jane Golley and Ligang Song, Canberra: Australia National University, 2011.
Brems, H. 1970. "A Growth Model of International Direct Investment." The American Economic Review 60 (3): 320-331.
Busse, M., Erdogan, C. & Muhlen, H. 2014. ‘China’s impact on Africa – The role of trade and FDI’. Carter Hill, R., Griffiths, W.E., Guay, C.L., 2012. ‘Principles of Econometrics, 4rd edition’.
Collier, P. and Goderis, B. 2008. ‘Commodity prices, growth, and the natural resource curse: reconciling a conundrum. Growth, and the Natural Resource Curse: Reconciling a Conundrum’. (June 5, 2008).
Davies, A. & G. Quinlivan, 2006. A panel data analysis of the impact of trade on human development, Journal of Socioeconomics, 35(5): 868-876.
39
Dollar, D., Wolff E.N., & Baumol W.J. (1988). The Factor-Price Equalization Model and Industry Labor Productivity: An Empirical Test across Countries. In: R. C. Feenstra (ed.), Empirical Methods for International Trade. Cambridge, MA: MIT Press, 23-47.
Dreher, A., Fuchs, A., 2011. ‘Rogue Aid? The Determinants of China’s Aid Allocation’. CESIFO WORKING PAPER NO. 3581
Durlauf, S.N., Johnson, P.A. & Temple, J.R.W. 2005. Growth econometrics, Handbook of economic growth, volume 1a.
Easterly, W. 2006. ‘The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill And So Little Good.’ Cato Journal, Penguin, City of Westminster, London, Great Britain. Forney, M. 2004. China’s Quest for Oil. TIME Asia Magazine. October 18.
Gomanee, K., Morrissey, O., Mosley P., & Verschoor A. 2003. Aid, Pro-poor Government Spending and Welfare. Centre for Research in Economic Development and International Trade, University of Nottingham, Nottingham. (CREDIT Research Paper 03/01)
Gomanee, K., Girma S., & Morrissey O. 2003. Aid, Public Spending and Human Welfare, Evidence from Quintile Regressions. Centre for Research in Economic Development and International Trade, University of Nottingham, Nottingham. (CREDIT Research Paper 03/13).
Gomanee, K., Girma S., & Morrissey O. 2003. Searching for Aid Threshold Effects: Aid, Growth and the Welfare of the Poor. Paper presented at a Seminar at CASE. Oxford, 28 January
Grossman, G.M., and E. Helpman, 1995. Technology and Trade. In G. Grossman and E. Helpman (eds.), ’Handbook of International Economics’, Vol. 3. Amsterdam: North-Holland, 1279- 337.
Ishfaq, M. 2004. ‘Aid Effectiveness, Debt Capacity and Debt Management in the Economy of Pakistan’. A dissertation submitted for PhD degree to Quad-i-Azam University, Islamabad.
Islam, N. 1995. Growth Empirics: A Panel Data Approach, Quarterly Journal of Economics, 110 (4): 1127–1170.
Jerven, M. & Johnston D. 2015. Statistical Tragedy in Africa? Evaluating the Data Base for African Economic Development, The Journal of Development Studies, 51:2, 111-115.
Jerven, M. & Duncan M. 2012. Revising GDP estimates in sub-Saharan Africa: Lessons from Ghana. African Statistical Journal, 15, 12–24.
Kolstad, I. & Wiig, A. 2011. ‘Better the Devil You Know? Chinese Foreign Direct Investment in Africa’, Journal of African Business, 12 (1): 31–50.
Kotrajaras, P. (2010). ‘Foreign Direct Investment and Economic Growth: A Comparative Study among East Asian Countries’. Applied Economics Journal 17 (2): 12-26
40
McMillan, M. & Rodrik, D., 2011. ‘Globalization, structural change and productivity growth. Working Paper 17143. Derived at: http://www.nber.org/papers/w17143
MOFCOM (Various years, a) China Commerce Yearbook. Beijing: China Commerce and Trade Press. MOFCOM (Various years, b) Statistical Bulletin of China’s Outward Foreign Direct Investment. Beijing: Ministry of Commerce of People’s Republic of China.
MOFTEC (Various years) Almanac of China’s Foreign Economic Relations and Trade. Hong Kong: China Foreign Economic Relations and Trade Publishing House.
NBS, 2014 Foreign Trade & Economic Cooperation, in Classifications & Methods, www.stats.gov.cn/english./ClassificationsMethods/Definitions/index_2.html.
NBS (Various years) China Statistical Yearbook. Beijing: China Statistics Press.
OECD, 2003. ‘Aid Effectiveness and Selectivity: Integrating Multiple Objectives into Aid Allocations’. Pre-Print of the DAC Journal 2003, Volume 4, No. 3
OECD (Organisation for Economic Co-operation and Development), 2008. The Paris Declaration on Aid Effectiveness and the Accra Agenda for Action. Paris.
OECD 2011, OECD Factbook 2011-2012: Economic, Environmental and Social Statistics, OECD Publishing, Paris, http://dx.doi.org/10.1787/factbook-2011-enData trade China-Africa
Page, J. 2012. ‘Can Africa industrialise?’. Journal of African Economies, 21, 86-124
PRIO 2009 Battle Deaths Dataset Version 3.0, Peace Research Institute Oslo, first published in B. Lacina and N.P. Gleditsch 2005 ‘Monitoring Trends in Global Combat: A New Dataset of Battle Deaths’, European Journal of Population, 21 (2–3): 145–166.
Reiter, S.L. & Steensma K.H. 2010. Human Development and Foreign Direct Investment in Developing countries: the influence of FDI policy and corruption. World development, 38(12), 1678-1691.
Rodrik, D. & Subramanian, A. 2009, Why Did Financial Globalization Disappoint? IMF Staff Papers, Vol. 56, Issue 1, pp. 112-138, 2009.
Roodman, D., 2009. ‘Practitioners’ corner: a not on the theme of too many instruments.’ Oxford Bulletin of Economics and Statistics, 71, 1(2009).
Ross, M.L. 1999 ‘The Political Economy of the Resource Curse’, World Politics 51: 297-322.
Sachs, J.D. & Warner, A.M. 1995. revised 1997, 1999. Natural resource abundance and economic growth. National Bureau of Economic Research Working paper No. 5398.
Sharma, B. & Gani, A. 2004. The effects of foreign direct investment on human development. Global Economy Journal 4(2), Article 9.
