Moshibudi Mokgashi 390202 The effects of politics, institutions and corporate governance on South Africa’s FDI flows By MOSHIBUDI MOKGASHI STUDENT NUMBER: 390202 THESIS SUBMITTED IN FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF MANAGEMENT IN FINANCE & INVESTMENT in the FACULTY OF COMMERCE, LAW AND MANAGEMENT WITS BUSINESS SCHOOL at the UNIVERSITY OF THE WITWATERSRAND, JOHANNESBURG SUPERVISOR: PROF. ODONGO KODONGO Moshibudi Mokgashi 390202 2 DECLARATION I, Moshibudi Mokgashi declare that the research work reported in this dissertation is my own, except where otherwise indicated and acknowledged. It is submitted for the degree of Master of Management in the University of the Witwatersrand, Johannesburg. This thesis has not, either in whole or in part, been submitted for a degree or diploma to any other universities. ………………………... …………………….. Signature of candidate Date Moshibudi Mokgashi 390202 3 ABSTRACT This paper explores the linkage between political risk, corporate governance, institutions and foreign direct investment inflows in the context of South Africa. The study was prompted the by the ever-changing political stability of the country, corporate governance and corruption challenges and their impact on doing business in SA. It is conducted using secondary data for World Governance Indicators (WGI) collected from World Bank’s online database and World Competitiveness from the World Economic Forum (WEF). The relationship was estimated using the Generalised Method of Moments (GMM) econometric technique for the period of 1996 to 2019. For political risk and institutions, governance variables were used. These are rule of law, political stability, control of corruption, voice and accountability, government effectiveness, and regulatory quality. For corporate governance, Ethical Behaviour of Firms and Efficacy of Corporate Boards competitiveness variables were used. Trade Openness, Inflation Rate and Gross Domestic Product growth were used as control variables. The findings of this report indicate that weak governance impacts FDI inflows negatively. The econometric estimates show that tolerance for corruption, government ineffectiveness, lack of rule of law, political instability, poor regulatory quality and accountability have negative impact on FDI inflows. Whilst all six variables indicated significant impact on FDI, rule of law and lack of control of corruption show the most impact. The implication is that this should be an area of focus to improve and therefor positively impact FDI. Overall, the government should reduce political instability and policy makers should employ policies and strategies to improve doing business in South Africa to attract and maintain investors. Moshibudi Mokgashi 390202 4 DEDICATION This thesis is dedicated to my mother Ngwakwana Grace Mokgashi and my late aunt Bridgette Selae Mokgashi. The two sources of inspiration for me to strive to achieve only the best, Mom and Aunty, I hope I made you proud. Moshibudi Mokgashi 390202 5 ACKNOWLEDGEMENTS I would like to thank and acknowledge: - The Almighty God, for my life, for loving, protecting and providing for me throughout my life; - My supervisor and lecturer, Professor Odongo Kodongo for your guidance, positive criticism, patience and wisdom; - The class of 2020; I hope to keep the friendships made for a lifetime - My employer, the Exxaro Resources for affording me the opportunity to pursue my studies living their value of Empower to Grow ; - My friends and family, especially my mother Grace, sister Mokgadi Mokgashi and niece Rorisang Mokgashi for supporting me through the challenging new way of learning; - All those who were positive and encouraging through this journey. Moshibudi Mokgashi 390202 6 Table of Contents List of Tables ............................................................................................................................................ 7 CHAPTER 1 ....................................................................................................................................................... 8 Introduction .............................................................................................................................................. 8 1.1. Background and Context ........................................................................................................... 8 1.2. Problem Statement ................................................................................................................... 15 1.3. Proven Gaps in Previous Studies .......................................................................................... 16 1.4. Research Questions ................................................................................................................. 17 1.5. Objectives of the Study............................................................................................................ 17 1.6. Contribution of the Study ........................................................................................................ 17 1.7. Benefit of the Study .................................................................................................................. 18 1.8. Structure of the Thesis ............................................................................................................ 18 CHAPTER 2 ..................................................................................................................................................... 19 2.1. Literature Review............................................................................................................................ 19 CHAPTER 3 ..................................................................................................................................................... 28 Methodology ........................................................................................................................................... 28 3.1. Data and Sources ........................................................................................................................... 28 3.2. Model Specification........................................................................................................................ 34 CHAPTER 4 ..................................................................................................................................................... 36 4.1. Results ............................................................................................................................................. 36 4.2. Discussion ....................................................................................................................................... 44 CHAPTER 5 ..................................................................................................................................................... 48 5.1. Conclusions and Recommendations .......................................................................................... 48 5.2. Study limitations and Further Research .................................................................................... 49 5.3. Ethical Considerations .................................................................................................................. 50 Appendices .................................................................................................................................................... 51 References ..................................................................................................................................................... 56 Moshibudi Mokgashi 390202 7 List of Tables Table 1: Conceptual framework on leadership and corporate governance .................................................... 25 Table 2: Summary of Explanatory Variables ................................................................................................... 36 Table 3 : Descriptive Statistics........................................................................................................................ 40 Table 4 : The results of Augmented Dickey-Fuller test (ADF) for unit root...................................................... 41 Table 5 : Correlation Matrix ........................................................................................................................... 42 Table 6 : Regression Results ........................................................................................................................... 43 Moshibudi Mokgashi 390202 8 CHAPTER 1 Introduction 1.1. Background and Context The Organisation for Economic Co-operation and Development (OECD) defines Foreign direct investment (FDI) as a long-term relationship established between the direct investor and direct investment enterprise brought about by a lasting interest by the resident enterprise (direct investor) in one economy in an investment enterprise into another economy at ownership level of 10% or more.1 FDI is an integral part of the economies of developing countries: by supplementing national savings, it facilitates accelerated economic growth (Sarbajit and Ujjani, 2014). Data from the UNCTAD (2019), plotted in Appendix A, shows FDI to be the top source of external income for developing countries and a steady growth is observed for the period of 2009 to 2018. Higher savings facilitate growth in GDP because the increased domestic capital can be used to accumulate capital and appropriate technology which improves aggregate output and therefore growth (Solow, 1956). The table in Appendix B shows gross savings ratios and growth rates for selected emerging markets over 39 years in periodic intervals of 10 years; the table shows that over time a decrease in gross savings as percentage of GDP results in a decrease in GDP growth for the same period. This is consistent with Prinsloo’s (2000) view that a high savings rate is associated with the ability of a country to sustain and grow its gross capital formation and therefore economic growth. Appendix C shows the savings rate of South Africa over the period of 1970 to 2019 and between the period of 1970 -1983 South Africa’s savings as a percentage of the GDP was well above the world average. However, the period from 1984 to date has seen a decline in gross savings, dropping from highs of 32% in 1980 to just 14.9% in 2019. Consistent with what is observed in Appendix B, declines in gross savings 1OECD definition of FDI Available at https://www.oecdilibrary.org/docserver/9789264045743en.pdf?expires=1595259836&id=id&accname=guest &checksum=DF33305EF382C946FCB06586B32DE09 [Accessed 24-07-2020] https://www.oecdilibrary.org/docserver/9789264045743en.pdf?expires=1595259836&id=id&accname=guest&checksum=DF33305EF382C946FCB06586B32DE09 https://www.oecdilibrary.org/docserver/9789264045743en.pdf?expires=1595259836&id=id&accname=guest&checksum=DF33305EF382C946FCB06586B32DE09 Moshibudi Mokgashi 390202 9 appear to be associated with declines in GDP per capita growth. This can be seen in Appendix D where South Africa recorded a GDP percentage growth high of 6.6% in 1980 (corresponding to the high gross savings in Appendix C) which has since dropped to a low of 0.15% in 2019 (corresponding to the low gross savings in Appendix C). As such, there exists a bidirectional relationship between gross savings and FDI inflows where growth in gross savings attracts more FDI and increase in FDI also helps countries to increase the level of gross savings (Chani, Shahbaz and Salahuddin, 2010). The same result was observed in a Sub-Saharan Africa study by Nurudeen and Karim (2016) who investigated the relationship between FDI, domestic saving and investment as well as economic growth over the period of 1981 to 2011 using VAR estimation and Granger causality tests. There reason this is so is because gross savings of a country indicate that the country can afford to put in place infrastructure required for development and make investment costs less and thus attractive savings (Chani, Shahbaz and Salahuddin, 2010). Appendix E depicts gross savings of the selected emerging countries in the table in Appendix B and Appendix F shows the associated FDI net inflows. Singapore has the highest savings of the six countries and attracts the most FDI as well. Studies appear to suggest that inflows of FDI triggers technology spillovers, assists human capital formation, contributes to international trade integration, helps create a more competitive business environment and enhances enterprise development (Loungani and Razin, 2001) which facilitate GDP growth. For countries like South Africa where excess capital from savings is a challenge to achieve, FDI becomes integral to serving the purpose of capital to use for development and technology advancement to stimulate output and thus GDP. The result of which will be economic growth which will subsequently create interest for investors to consider South Africa for FDI. FDI has, in several studies, been shown to be the growth catalyst that developing countries need. For instance, Aitken, Hanson and Harrison (1997) used panel data of 33 developing countries to investigate the long run relationship between FDI and total factor productivity and found that there is a positive relationship and that countries with higher FDI offer high wages. Dinh, Vo and Nuyeng (2019) investigated the impact of Moshibudi Mokgashi 390202 10 FDI on economic growth in 30 developing countries using panel data for the period of 2000 to 2014 and found that, in the long run, FDI has a positive impact on the economic growth of those countries and further recommended that emerging and developing countries should make effort to attract FDI to supplement domestic investment for the benefit of their economies. Developing countries have been historically at a disadvantage with regards to FDI inflows because of the high risk that they carry due to political uncertainty and poor policies (Henisz and Zelner,2015). This recently took a turn where the gap of the split of FDI between developing countries and developed countries has in the last 10 years continued to decrease with the years 2014 and 2018 seeing developing countries attracting more FDI than developed countries, fifty-six percent and fifty-four percent of world FDI respectively. See Appendix H (UNCTAD, 2019). Data from the United Nations Conference on Trade and Development (UNCTAD) shows that, over the years, South Africa has continued to lag its BRICS2 peers. Research shows that for the period of 2010 to 2016 South Africa only recorded an average of 1.2% of Gross Domestic Product (GDP) growth compared to the global average of 5% (Krugel and Viljoen, 2018). In the same period, South Africa‘s risk score saw a steep incline rendering it riskier in proposition than comparable developing countries.3 The OECD indicates that the attractiveness of FDI inflows to African countries is a subject of government policy, friendliness towards trade, and of course investors perspective. When it comes to investors, perceptions regarding potential for new business in African countries through looking into key economic indicators of economic stability is important. Krugel and Viljoen (2018) indicate that of the 19 key factors that potential investors consider in investment decisions, the two factors that show the highest correlation with FDI inflows are openness to trade and efficiency of government regulation across 132 countries. They also found that specific to South Africa; political stability, quality of governance and property rights are the top three 2 BRICS is a group of five major emerging countries – Brazil, Russia, India, China and South Africa 3 Available at https://www.euromoney.com/article/b14rlrd31jt3j7/south-african-political-risk-is- overshadowing-economic-recovery [Accessed 15-06-2020] https://www.euromoney.com/article/b14rlrd31jt3j7/south-african-political-risk-is-overshadowing-economic-recovery https://www.euromoney.com/article/b14rlrd31jt3j7/south-african-political-risk-is-overshadowing-economic-recovery Moshibudi Mokgashi 390202 11 factors most correlated to FDI inflows. A deterioration in these three factors saw a decline in FDI inflows. 1.1.1. Politics and FDI Politics affect a country’s economic development by either supporting or disrupting it through factors such as the regime type, instability or stability, corruption and trade laws. Investors shy away from countries with political risk because governments of such countries tend to discriminatorily change laws, regulations or contracts governing investments as well making it expensive to run business which reduce investor returns (Henisz and Zelner, 2010). In his study of South Africa’s savings behaviour, Prinsloo (2000) concludes that the decline in gross savings was a result of the net borrowings of the general government attributed to subdued economic growth in the period of 1985 to 1993. It is a known fact that it was during this period that South Africa was at the height of anti-apartheid conflict and political unrest and a lot of sanctions and repressions imposed on the economy resulting in a slow growth and overall underperformance. A study by Busse and Hefeker (2005) revealed that internal conflict, external conflict, and democratic accountability are amongst the top 5 variables that negatively impact FDI at 1% significant level. They describe internal conflict as political violence and civil disorder and external conflict to be risk to the incumbent government. This includes foreign action ranging from non-violent external pressure such as diplomatic pressures, withholding aid or trade sanctions and democratic accountability and responsiveness of the government to its citizens as well as fundamental civil liberties and political rights. All of which describe the state of South Africa during the height of anti-apartheid conflict of 1985 to 1993. In light of Busse and Heferker (2005) study, it is no surprise that in 1985, FDI declined to a low of -0.76 as percentage of GDP followed by a slow erratic growth into 1993. In 1994 sanctions and economic pressures that were imposed on South Africa were lifted following the abolishment of apartheid and implementation of policies that became more liberal and outward-oriented with the aim of attracting FDI (Arvanitis, 2006). As such, this resulted in slow but steady improvement in FDI inflows recording an improvement of 97% from the transition year of 1993 into the democracy year of 1994. Moshibudi Mokgashi 390202 12 Appendix I shows the changes of FDI in South Africa pre-democracy period and period following democracy and the country has been able to better attract FDIs since the shift into a government of democracy. In the years leading up to the present, political issues in South Africa have taken a turn from a regime of apartheid and oppression to a democratic regime. Political differences, cabinet reshuffles, conflicting ideologies and corruption by officials are some of the risk factors that have continued to heighten in the country leaving investors uncertain when it comes to making investment decisions (Stapelberg,2016). Although FDI in the democracy period has indeed been erratic and shows that the country is not able to sustain its highest recorded inflow of 5.9% in FDI (Appendix I). Stapelberg’s conclusions are not accompanied by any empirical work and as such it is important that this study investigates this political risk and FDI link using empirical evidence. The results of which can be used to explain this link further as well as be used as reference for policy makers. 1.1.2. Institutions and FDI Institutions are defined as rules of the game that structure social behaviour and interactions and as such help form stable expectations (Nabli & Nugent, 1989). In the economic context, Wiggings and Davis (2006) define institutions as “rational mechanisms designed to cope with the imperfections of markets including the asymmetry of information held by different actors, the problems that principals have in ensuring that their agents pursue the same goals, etc”. They explain that although institutions can be formed informally through repeated interactions between individuals and organisations, it is those that are formalized by deliberate design through interactions between governments, private enterprise and civil society. This is usually in response to uncertainty, risk and information cost associated with living and transacting that end up widely recognized and legally binding by the state. It is therefore justified that Stapelberg (2006) links political instability and the quality of institutions in South Africa. He argues that policy uncertainties that arise from the institutional quality are a by-product of political differences, conflicting ideologies, cabinet reshuffles and corruption in South Africa and as such have rendered the country undesirable for investment and therefore attracting low levels of FDI. Moshibudi Mokgashi 390202 13 This conclusion by Stapelberg (2006) is not accompanied by any empirical study that explores the institutional variables that link it to FDI inflow and describe the anticipated negative relationship. This therefore is another reason for this study be done in the context of South Africa to fill the empirical evidence gap. North’s theory of institutional quality (2009) stresses that good institutions reduce production and transaction costs and as a result companies can be profitable and influence economic activity which then attract investors and as such FDI. Dunning (2006) expands the concept by arguing that foreign investors prefer locations that offer the best economic and institutional facilities because their decisions on the rate of return to be expected is impacted by the status of institutions and other macroeconomic indicators of that home country. Several institutional variables have been studied to establish their link and impact on FDI. For instance, Wiggings and Davis (2006) identified three institutions that perform economic functions: establishing and protecting property rights, facilitating transactions, and permitting economic co-operation and organisation. Other studies have been done to examine the growth enhancing effect of institutional variables that also included security of properties, rule of law, control of corruption and financial intermediaries on their impact to FDI. The further away the home country is from source country’s desired institutional quality for investment, the less attractive it is to source country and thus impacting FDI for home country (Buchanan, Le & Rishi, 2011). Kandiero and Wahawan (2003) conducted an empirical study on Institutional quality, Openness, and Investment in Africa for the period of 1990 to 2000. It shows that the quality of institutions in African countries (Including South Africa) causes a reduction in overall benefits of trade. One other finding from the empirical study is that increased openness to trade is much more effective when there is less corruption as well as less risk for investors coming from credibility of government and policies that promote firm confiscations and nationalization. Between 2010 and 2011 there was talks of Nationalization of mines in South Africa and FDI decreased by 62% percent from 2009 (Appendix I). Similarly, the Business confidence level also decreased from 41 to 27 in the same years (i.e. 2009 to 2010)4 4 See Appendix J – Trading Economics Business Confidence Index Moshibudi Mokgashi 390202 14 These observations are in line with the findings of the empirical study done by Kandiero and Wahawan (2003). Thus, this makes it important to explore variables of economic institutions and their impact on FDI in the context of South Africa to ensure that a clear and true conclusion is drawn for South Africa. The results can be used to identify improvement gaps on institutional quality and assist policymakers in passing policies that favour FDI inflows. 1.1.3. Cooperate Governance and FDI Isaksson of the OECD (1999) defines Corporate governance as ‘the system by which business corporations are directed and controlled. This is done by specifying the distribution of rights and responsibilities among the different participants in the corporation and a corporate governance framework that comprises of elements that are essential for a successful outcome at all stages in the investment process’. Appiah-Kubi et al (2020) investigated the impact of corporate governance and institutions on FDI inflows in West African countries using panel data over the period of 2009 – 2018. Like Agyemang, Gbettey, Gatsi, and Acquah (2019), they found economies with firms that display high ethical values were likely to attract higher FDI inflows than those that did not. Good cooperate governance fosters firm integrity and reduces information asymmetry and hence improves investor confidence (Shadid and Abbas, 2019). This is in tune with Isaksson’s definition of corporate governance. Anything in the framework and behaviour of participants that puts the success of the investment process at risk would be rendered bad cooperate governance that therefore exposes the investor to risk. One other challenge that firms face with regards to cooperate governance is the agency problem which refers to difficulties’ financiers face in assuring that their funds are not expropriated or wasted on unattractive projects. Although many are the efforts to try and manage the urgency problem, self-dealing remains to be a challenge and a nightmare for corporates and therefore investors. A study by Agosin and Machado (2005) on poor governance in Russia revealed that higher frequency of illegal payments and high pressure from regulatory agencies, enforcement authorities and criminals have negative effects on FDI. Moshibudi Mokgashi 390202 15 South Africa has in the last 10 years recorded some of the worst corporate governance scandals in government, corporates and auditing institutions where those in power used their positions and authority to fraudulently enrich themselves leaving minority shareholders and stakeholders in dismay. The biggest of these include Steinhoff, Tongaat Hulett, Guptas, VBS Bank, KPMG, McKinsey & Company, SAP, ESKOM, Sharemax, EOH, Bosasa, Sasol, Gold Fields and several construction and media companies5. Following these events, South Africa’s Corporate Governance recorded its worst ever performance in 2018 according to the Institute of Internal Auditors SA however the country saw an increase in FDI inflows in that year according to the World Bank. This conflicts with the literature reviewed and stated in this paper suggests. This may be due to several reasons, possibly including the appointment of a new president for the ruling party, ANC, in December 2017. It is thus imperative that this study is indeed carried out to test the link between FDI and corporate governance and the extent of the relationship in the South African context. 1.2. Problem Statement Political, Institutional and Corporate governance risk gives rise to policy uncertainty and unstable economic environment (Stapelberg,2006). Such instability in a host country is not attractive to investors because of the uncertainty regarding return on investment and growth. South Africa is a country whose savings rate cannot sustain the required growth and development and as such foreign capital is essential for its economy (Jordaan, 2016). Historically, the country was one of the best in attracting FDI in the African region until the massive decline during the years of policy uncertainty and loss of business confidence associated with political risk, ongoing corporate governance scandals and talks of nationalization. The lack of understanding of how these factors impact FDI flows in South Africa by policy makers due to low research effort is a problem that continues to see South Africa losing position in attracting FDI. This continual loss of interest by investors is detrimental to the South African economic growth and the livelihood of the people thereof. African countries that have previously 5 Available at https://www.businessinsider.co.za/the-top-south-african-business-scandals-the-past- decade-2020-1 [Accessed 03-07-2020] https://www.businessinsider.co.za/the-top-south-african-business-scandals-the-past-decade-2020-1 https://www.businessinsider.co.za/the-top-south-african-business-scandals-the-past-decade-2020-1 Moshibudi Mokgashi 390202 16 ignored the signalling of loss of interest for investment by foreign investors due to unfavourable, non-incentivizing policies and overall economic stability were impacted negatively and have recorded low economic growth and development as has been evident in the neighbouring Zimbabwe. It is thus important to investigate the relationship between these factors with FDI inflows in the context of South Africa to address the current decline in inflows. 1.3. Proven Gaps in Previous Studies In reviewing previous work done and anecdotal evidence, there is an indication that there exists a potential relationship between FDI and politics, institutions, and governance. Most of the work reviewed thus far is done using panel data and focusing on a few countries of interest by the researchers. Fewer work is done with the interest of focusing on a specific country to review how the status of these variables has impacted FDI inflows over the years. For instance, the study by Appiah-Kubi et al (2020) on corporate governance is focused on group of West-African countries. Busse and Hefeker (2005) is a frequently referenced study for studies that investigate FDI and these three variables and this popular study also focuses on a total of 83 developing countries. One other gap observed specific to the context of South Africa is that some of the highlighted challenges in attracting FDI are not empirically proven. An example is the view Stapelberg (2016) holds on how politics and institutions impact FDI in South Africa but does not provide empirical evidence. A review of South Africa post freedom was done by Goldman Sachs for a period of 20 years. The main recommendations from the report include improving governance and accountability, building investor confidence, common effort by government and institutions to protect the sovereign rating and cost of capital amongst other things to attract FDI6. He further warns that if this is not done, SA may continue to rely on debt which will in turn affect the growth of the country. 6 Coleman C. 2013. Two decades of freedom: What South Africa is doing with it and what needs to be done. Goldman Sachs. Available at https://www.goldmansachs.com/insights/archive/colin-coleman-south- africa/20-yrs-of-freedom.pdf [Accessed 22-07-2020] https://www.goldmansachs.com/insights/archive/colin-coleman-south-africa/20-yrs-of-freedom.pdf https://www.goldmansachs.com/insights/archive/colin-coleman-south-africa/20-yrs-of-freedom.pdf Moshibudi Mokgashi 390202 17 1.4. Research Questions In this thesis, I argue, following Busse and Hefeker (2005), Kandiero and Wahawan (2003) and Appiah-Kubi et al (2020), that factors such as business confidence and sovereign rating are intractably linked to the underlying political risk, institutional quality and corporate governance. Therefore, the thesis seeks to explore the linkages between these underlying variables and FDI flows. Accordingly, the primary research question for this research is: To what extent does Political Risk, Institutions and corporate governance impact foreign direct inflows of South Africa? Sub questions: 1. What is the relationship, if any, between changes in political risk, Institutions and corporate governance and foreign direct inflows of South Africa? 2. How do the key factors from each of these variables impact FDI inflow? 1.5. Objectives of the Study Thus, the objectives of the study are: • To ascertain the impact that Politics, Institutions and Corporate Governance has on FDI influx to South Africa. • Use the results to identify key factors that are important for the South African government and policy makers to address and such create favourable environment to attract foreign direct investment. 1.6. Contribution of the Study There have been several opinions around the issues that are resulting in the decline in business confidence and FDI in South Africa. Political Risk being at the forefront following several corruption cases, cabinet reshuffles, questionable policies as well as corporate governance issues. The study will peruse the relationship of these three determinants to FDI inflow to confirm whether the sentiments are correct or not as well as to what extent. Importantly, based on the results, recommendations will be made Moshibudi Mokgashi 390202 18 as a guide to what the government, institutions and corporates should be doing to positively impact FDI inflows. 1.7. Benefit of the Study The study is relevant to the South African government, Policy Makers as well Corporate firms. This is so because they are at the forefront of influencing FDI inflow when it comes to these three determinates that will be investigated in this study. Addressing these issues and attracting more FDI into the country will potentially assist with economic growth which will help alleviate the increasing unemployment, improved education, quality of public services and management of government finances. 1.8. Structure of the Thesis Following chapter 1, the rest of the study is organised as follows: Chapter 2 focuses on the theoretical framework and literature review. Chapter 3 deals with the empirical analysis, with emphasis on the formulation and estimation of the models to examine the three determinants of focus and their impact on FDI inflows into the country. Chapter 4 focuses on interpretation of the models estimated in Chapter 3 and a discussion of the results Chapter 5 summarizes the main conclusions, policy recommendations and limitations of the study and further research recommendations. Moshibudi Mokgashi 390202 19 CHAPTER 2 2.1. Literature Review Host countries of FDI benefit through various ways. These include the transfer of knowledge and technology from developed countries to developing countries (Borensztein, De Gregorio and Lee, 1998). Productivity is improved by labour training and new skill acquisition and often increased labour demand responding to a shift in production and new ways of work (Ekholm, 2004, Balasubramanyam et al.,1996 and De Mello, 1999). Furthermore, FDI enables access to international production networks and access to international markets7. All of these have the potential of positively impacting economic growth for the host country. Whilst FDI has been proven to positively impact economic growth of host countries, economic growth is also one of the important macroeconomic elements that attracts FDI. Evidence of a link between FDI inflow and economic growth was studied by Iamsiraroj (2016) over 124 of 195 countries in the world, a representation of 63%, in which he finds the following: “A bi-directional relationship between FDI and economic growth. It means that host country growth can determine and be determined by FDI inflows.” “Not only the direct effect from growth per se that could influence FDI inflows but also the level of labour force, trade restrictions and friendly investment climate.” “These elements will influence FDI into the country and potentially stimulate economic growth through FDI inflows, according to a dynamic relationship between FDI and economic growth.” One can already see that host countries need to be economically competitive and able to some extent be in constant growth to attract or benefit from FDI. These elements are exclusively highlighted as they will be extensively elaborated on further in the literature review of the different determinants of importance for this study and their interconnectedness. As previously indicated, these are Politics, Institutions and governance. 7Foreign Direct Investment for Development, OECD, (2002). Available at https://www.oecd.org/investment/investmentfordevelopment/1959815.pdf https://www.oecd.org/investment/investmentfordevelopment/1959815.pdf Moshibudi Mokgashi 390202 20 Evidence from several studies shows that for host countries to benefit economic growth effects from FDI, it is largely dependent on the host country’s domestic conditions. Things such as human capital, technology gap, degree of openness (Forte and Moura,2013); overall level of financial development and access to finance (Suliman, 2014); corruption and lawlessness (Luu, Nguyeng, Ho and Nam,2019) and economic nationalism (Kobrin, 1984) have influence on the link between FDI and economic growth. These are all a result of quality of institutions, political stability and governance in the host country. The role of good institutions when it comes to FDI entails reduction of transaction costs associated with doing business in the host country (Prufer and Tondl, 2008) and North (1990) and facilitation of the absorption of the technology spillovers which can, in developing countries, be influenced by the gap and receptiveness of technology often referred to as relative backwardness (Falvey et al, 2005). Bad institutions on the other hand do the opposite and several researchers have indicated this. For instance, a study by Islam et al (2020) indicates that whilst financial development significantly attracts FDI, financial markets are less attractive to FDI inflows relative to financial institutions. Shukarov (2017) in a study finds that Western businesses entering countries with low quality institutions are faced with high transactions costs due to bribery and other costs compared to countries with higher degree of institutional development. African institutions tend to limit their openness to trade and capital inflow in fear that greater FDI inflow can ultimately lead to foreigners controlling key sectors of the economy and therefor controlling policy (Moss, Ramachandran and Shah, 2004). South Africa is no different when it comes to this ideology and it is this ideology that led to the proposals of nationalisation of mines in 2012. A deep dive into the effects of such political influenced changes regarding business on FDI as well as the overall state of the country’s institutions is therefore necessary to better manage perception as well as ensure that institutions are managed for the better. There are currently very few research papers on the relationship between institutions and FDI in the context of South Africa. It should be noted that institutions in developing countries are not as strong as those in developed countries. It is for this reason that study observations by Sabir, Rafique and Abbas (2019) found that institutions are a Moshibudi Mokgashi 390202 21 significant determinant for FDI inflow in developed countries than developing countries. Perhaps this could be reason for the lack of research papers on this topic for South Africa. Malikani and Chitambara (2017) investigated the link between FDI, Democracy and Economic Growth in the context of Southern African countries for the period of 1980 to 2014 using the GMM estimator. They found that strong democratic institutions are important for economic growth which further stimulates FDI inflow. They further established that to benefit from FDI spillovers, a country’s level of democracy must be relatively high. Huang (2003) indicates that poor institutions do not facilitate or enable a culture of entrepreneurship thus reducing supply of local entrepreneurship when compared to countries with good institutions. This can negatively impact FDI because there exists a blink between entrepreneurship orientation and economic growth (Stoica et al, 2020) which in turn is a macroeconomic determinant of FDI that has been widely covered by studies on macroeconomic determinants of FDI. According to Kandiero and Wadhawan (2003), increased openness to trade and quality of institutions can be explained by the behaviour of government bureaucrats. Their investigation of the interaction between openness to trade and institutional quality to FDI using GLS method of estimation for African countries indicates that trade openness requires good quality institutions for it to be a significant FDI inflow determinant for the host country. A study of Sub-Saharan African countries revealed that quality of institutions affects the level of FDI inflow and is therefore an essential factor that can be used to drive FDI flows inward (Adegboye, Osabihien, Olokoyo, Matthew and Adediran, 2020). An empirical study by Lindelwa and Makoni (2018) had similar findings. Their study highlights the need for other researchers to explore the threshold level that institutional quality needs to reach before it can positively impact FDI inflows, particularly in developing countries. Moshibudi Mokgashi 390202 22 Whilst sound institutions are important for FDI inflows, Politics of a country heavily influence the quality of institutions owing to the leadership and policymakers (Zolcsák,2015). The politics of a country can either be stable or unstable, both these states have different impacts and influence on the institutions and governance. Political instability is likely to negatively affect the economic growth and consequently FDI inflow of a country because the leadership tends to be short-sighted and indecisive on policies resulting in volatility (Aisen and Veiga, 2011). The Butler-Joaquin model by Butler and Joaquin (1998) was developed to show how political risk impacts cost of capital for investments. It explains that if the expected change in political environment is expected to impact the future cashflows of an investment (positively or negatively) the response to this political risk shock will be an increase (if negatively impacting cashflows) the cost of capital and the reverse is true for if the political change impacts cashflows negatively. Political risk impact on FDI research dates back to the work of Root & Ahmed (1979 and Schneider & Frey (1985) who found that political instability has significant impact on FDI. In the 90s, Brunetti and Weder (19980, found that there exists a negative prelateship between institutional uncertainty and private investment. Property protection appears in most studies as factor that has positive relationship with FDI inflow, one such study is of Lee and Mansfield (1996) Studies of Harms & Ursprung (2002), Jensen (2003), and Busse & Hefeker (2005) found that democracy is important factor to multinational corporations in screening for countries to invest. In these studies, democratic rights influence the improvement of property rights which is an attractive factor for investors. The main reason for multinational corporations’ need to consider these is because of the political and economic risk they may be exposed to in future which poses risk to their returns (McGowan and Moeller, 2009). Alfaro et al (2008), indicate that the question of why capital does not flow from rich developed countries to poor and developing countries can be answered by the political risk associated with developing countries. Moshibudi Mokgashi 390202 23 The negative perspective resulting from political insatiability of a country can affect that country’s ability to attract FDI even when policies are oriented for economic growth. Bogran and Clark (1999) demonstrate this in their review of the history of FDI in South Africa where they indicate that whilst the policies where conducive for business, investors rejected South Africa because of the turmoil and political unrest due to apartheid. It is thus important for domestic situation of a country to be holistically in a good state for FDI. Post-apartheid and political turmoil era saw the South African government make more efforts in improving trade openness such as joining the World Trade Organisation in 1994, free flow of funds through the bilateral investment treaty as well as tax treaties as reference by Bogran and Clark (1999). However, because of the downside of these policies on the domestic business, the impact was felt in various ways such as the increase in unemployment to 41% (Joyner, 1998) and the crime rate reaching rates of 63% during the same period.8 It is because of these results and historical colonialism that political ideologies and outlook differ between the leadership in South Africa when it comes to foreign investment thus causing political risk. Inherent to the history of African countries, is colonialism which Moss et al (2004) uses to explain the tendency of African leadership to gravitate to the socialist model and anti-foreign investment ideology. They further explain that from their studies of FDI and development, the socialist ideology and the common objections of foreign investment are not justified as foreign firms do not crowd the local markets as perceived. The political leadership and ideologies of socialism explains the impact of political risk on FDI in the context of investor perception and reluctance to invest in Africa for the early stages of most of these countries gaining back freedom. Many of the African leaders now speak a language of openness to capitalism with few being nationalists (Moss et al, 2004). However, the question remains for many researchers on the modern-day political challenges that pose a risk to investors causing Africa to still not be able to attract more FDI. This knowledge gap relating the issue to political and governance to FDI in the context of Africa and relevant to this study, South Africa, has been given less attention. It is 8 World Development Indicators: Crime Moshibudi Mokgashi 390202 24 for this reason that a lot of sentiments made by some writers are not accompanied by real empirical studies to support their views. As previously stated, few studies have investigated the relationship between political risk, governance and institutions and FDI inflows in the context of South Africa. Khan and Akbar (2013) have attributed this to lack of high-quality time series data for Sub- Saharan African countries (SA is also part of this grouping). It is for this reason that in their study, only 24% of their sample data was from sub-Saharan countries out of a possible 44%. With this background in mind, some researchers have indeed been able to conduct investigations on some of these variables to try and explain FDI inflows in South Africa. One such recent study is that of Meyer and Habanabakize (2018) who looked at the impact that political risk and economic growth had on FDI inflows in South Africa. Using an ARDL model and time series data from 1995 to 2016, they found that the lower the political risk rating, the higher the level of FDI inflows. A recommendation for future research by this study tables investigating other non-economic factors and their relationship with FDI. This study intends to contribute to that gap through exploring the corporate governance and institutions impact on FDI. As with institutions, political leadership has influence on governance through the quality of leadership (Shailer and Greg, 2004). Ethical leadership promotes good practices of co-operation, harmony and empowerment. Coupled with ethical behaviour, good political leadership specifies vision and enables followers to drive the vision (Ali, Tao, Shaikh and Sajid, 2017). With all this in place, good government effectiveness is likely to be the result. It is these learnings that when adopted by corporate leadership, corporate governance frameworks can effectively be used facilitate for good governance practices that are important to investors as well as ensure that the goals of the economy can be met (Ali et al, 2017). They used Table 1 to show a conceptual framework of how corporate governance of a host country is underpinned by its leadership in both politics and corporate. Moshibudi Mokgashi 390202 25 TABLE 1:CONCEPTUAL FRAMEWORK ON LEADERSHIP AND CORPORATE GOVERNANCE Corporate governance is important to investors because they believe that a financial framework under poor corporate governance will not be able to protect their investments (Globerman and Shapiro, 2002). The aim of corporate governance is indeed to protect the interests of shareholders (Cook et al, 2013) and this is important because of the prevailing agency problem that continues to be a challenge worldwide (Valdés and Foster, 2008). Li (2005) defines good governance to be “independent judiciary and legislation, fair and transparent laws with impartial enforcement, reliable public financial information, and high public trust.” “Governance indicators at both national and corporate level may be from different sources however the underlying influence to their quality is the quality of institutions” (Biro,2019). Good institutions are likely to bear good governance which will protect investments, reduce costs whilst lowering uncertainty and therefore improving investor confidence in the host country (Montero, 2008; Mengistu and Adhikary, 2011). Poor governance quality impacts badly on FDI whilst the opposite is true for good governance as investigate by Kuzmina, Volchkova and Zueva (2014). They found that improving governance quality from an average level to top quality across Russia more than doubled the FDI inflow. Moshibudi Mokgashi 390202 26 Several researchers have investigated which elements of governance influence FDI, especially at national level. Schleifer and Vishny (1993) indicate that corruption affects FDI inflows in terms of the total amount of inflow as well as composition; Maskus (2000) notes a positive effect that strong intellectual property has on FDI inflow although it is sector dependent; whilst Younsi and Bechtini (2019) found political stability, government effectiveness and regulatory quality to positively and significantly influence FDI. In the African context, research by Gangi and Abdulrazak (2012) shows that of the six governance indicators only three; Rule of law, government effectiveness and voice and accountability are statistically significant in positively impacting FDI. A review of the FDI-Growth Nexus in Africa indicates that where good governance is interacted with FDI there is positive increase in growth and thus governance structures matter for FDI (Adeleke, 2014). West African countries have over the period of 2010 to 2018 lead all other corners of Africa in pulling in FDI and this is attributed to their firms that adhere to codes and ethics (Addison and Heshmati, 2004; Globerman and Shapiro, 2002). A study by Gathoni (2017) looked at the effect corporate governance has on FDI and a comparison of the standing of corporate governance in Kenya and South Africa. The results of the study showed that a robust corporate governance system has a positive effect on FDI and that corruption seems to be a challenge for both countries that it is potentially affecting foreign direct investment. A cause to pursue the impact of corruption levels is therefore perpetuated by their study which will therefore be incorporated in this study. The findings of the study by Agyemang, Gbettey, Gatsi, and Acquah (2019) are similar to that of Adeoye (2009) and Gathoni (2017) where they indicate that in African countries where there is high firm ethical values tend to attract a great deal of FDI. Thus far, no particular research has been conducted that looks at corporate governance impact on FDI in the context of South Africa. Work that has been done which includes South Africa as a participant in the sample group includes that of Adeoye (2009), Agyemang et al (2019), and Gathoni (2017). Adeoye (2009) used panel data of 33 countries over the period of 1997 and 2002 to estimate the impact that macro-economic level corporate governance has on FDI Moshibudi Mokgashi 390202 27 using generalized least squares (GLS) with random effects estimation method. The results of the study indicated that macro-level economic growth has a positive effect on FDI inflows at five percent level of significance. A recommendation for further research to fill the gaps is to evaluate each of the variables that constitute the corporate governance index to give better guide to policy makers in addressing matters of corporate governance framework. This is indeed to be investigated in this report by looking at the emerging challenges regarding corporate governance in the context of South Africa as this is objective 2 of this research Whilst these studies are proven and useful, they cannot be used to generalize the situation in South Africa, and this is because of the differences that exists between the countries sampled to reach these conclusions observed in the literature. Moshibudi Mokgashi 390202 28 CHAPTER 3 Methodology 3.1. Data and Sources This study is based on a quantitative research methodology. The analysis comprises of the period of 1996 to 2018 for which linkage between political instability, institutional quality and corporate governance to FDI is of interest for this investigation. This is because it is during this period where FDI net inflows show an erratic trend and several writers have attributed this to elements associated with these variables having a negative impact on FDI. Whilst most of the data to be used for this research is available for the said period, data for corporate governance is only available from 2007 because it was previously not recorded as was with the Political Risk and Institutions data. In the context of this study, some of the interesting events that have taken place during this period include changes in leadership in the country, public governance failures, multiple cabinet reshuffles all of which relate to the Politics of the country and therefore suitable to explore the impact that these have had on FDI. When it comes to the Institutional Quality variable, events of interest that took place during this period include indications of nationalisation of mines and politics of land expropriation. Relating to corporate governance, a number of corporate governance scandals from big firms in the country as well as macro level governance concerns occurred during this period as highlighted in chapter 1. 3.1.1. Political Risk, Institutions and Governance Variables Various sources of data will be used for the study. The dependent variable which is Foreign Direct Investment (FDI) will be extracted from the World Development Indicators, a database from the World Bank. 3.1.1.1. Political Risk and Institutions The Political Risk Services (PRS) which provides the International Country Risk Guide (ICRG) with Political Risk Rating at country level and the Worldwide Governance Indicators (WGI) use the same components in evaluating political governance and institutional standing however at different levels of granularity. Whilst the PRS tends to evaluate each of the twelve components, measuring the various dimensions (often referred to as sub-indicators), the WGI uses these sub-indicators which are in total six. Moshibudi Mokgashi 390202 29 Busse and Hefeker (2005) used all 12 components in their analysis of the relationship between political risk, institutions and foreign direct investment and were able to reach conclusive results as indicated in chapter 1 of this report. On the other hand, an empirical study by Buchanan, Le and Rishi (2011) used an index of the six sub- indicators to investigate the relationship between foreign direct investment and institutional quality using panel data of 164 countries and observed a significant positive effect. This study will thus follow the use of the six sub-indicators to evaluate the country’s political risk and institutional quality. They are Voice and Accountability, Political Stability and Absence of Violence, Government Effectiveness, Regulatory Quality, Rule of Law and Control of Corruption which will be discussed in detail. These six sub-indicators are said to be highly correlated and may introduce multicollinearity by various researchers such as Globerman and Shapiro (2002). Busse and Hefeker (2005) also acknowledged this in their study and proceeded to use the components singly in the regression equation to avoid multicollinearity. It is also for this reason that Buchanan, Le and Rishi (2011) used an index of the six sub- indicators in their model equation which follows the recommendation of avoiding multicollinearity through principal component analysis (PCA) by Daoud (2017). In this study the sub-indicators will be added singly to the regression equation to avoid multicollinearity. Secondly, the Durbin-Watson statistic will also be used to check for multicollinearity. The data set for six sub-indicators used will be the 2019 update which incorporates revisions to data for previous years making it the most reliable version (Kandiero and Wahawan, 2003). They are discussed as follows: Political Stability and Absence of Violence (PRS_PV) The four components under this sub-indicator include Government Stability, Internal Conflict, External Conflict and Ethnic Tensions. The PRS (2005)9 defines them as follows: Government Stability - the ability for the government to carry out its policies and stay in office 9 PRS Group (2005) for details on sub-components and aggregation procedures Moshibudi Mokgashi 390202 30 Internal Conflict – political violence within the country and its actual or potential impact on governance by focusing on, for instance, civil war, terrorism, political violence or civil disorder. External Conflict - weighs external conflict, namely the risk to the incumbent government from foreign action, ranging from non-violent external pressure, such as diplomatic pressures, withholding aid or trade sanctions, to violent external pressures, ranging from cross-border conflicts to all-out war Ethnic Tensions - assesses the degree of tensions among ethnic groups attributable to racial, nationality or language divisions Voice and Accountability (PRS_VA) The two components under this sub-indicator are defined by the PRS (2005): Democratic Accountability – relates to the democratic accountability of the government, that is, the responsiveness of the government to its citizens, but also to fundamental civil liberties and political rights Military in Politics - represents the influence of the military in politics, which could signal that the government is unable to function effectively and that the country might have an unfavourable environment for business. Regulatory Quality (RQ) Investment Profile - Reflects perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development (WGI, 2019). These shape factors that affect risk to investment such as contract viability/expropriation, profit repatriation, and payment delays. Expropriation risk: the risk of foreign firms’ domestic assets being confiscated or nationalized. Repudiation of contracts by government reflects the credibility of government in the enforcement of contracts. Government Effectiveness (GE) Bureaucratic Quality - Reflects perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies (WGI,2019). Moshibudi Mokgashi 390202 31 Rule of Law (RL) Rule of Law captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence (WGI, 2019). Control of Corruption (CC) This captures perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as "capture" of the state by elites and private interests. Estimate gives the country's score on the aggregate indicator, in units of a standard normal distribution (WGI,2019). The WGI uses an estimate rating system of -2.5 to 2.5 where -2.5 represent a lowest rating and therefore a bad outlook and 2.5 represents highest rating and therefor a good outlook. 3.1.1.2. Corporate Governance Data for Corporate Governance variables will be sourced from the World Competitiveness Report deduced from the World Economic Forum’s (WEF) Opinion Survey. The WEF shares several indicators that can be used to evaluate corporate governance at country level such as protection of minority shareholder’s rights, efficacy of corporate boards, regulations of the securities exchange and firm’s ethical behaviour. These are defined in the World Competitiveness Report (2017) as follow10. Protection of minority shareholder’s rights The indicator measures the extent to which minority rights of shareholders are protected by the legal system. This indicator is important to investors because it gives them an indication of whether the host country will allow them to exercise freely their rights such as selling of shares or receiving dividends. Efficacy of corporate boards (ECB) Efficacy of corporate boards refers to the extent to which management of firms is accountable to the board of investors on their responsibility of practicing the corporate governance framework. This is a framework that comprises of elements that are 10 Available at http://www3.weforum.org/docs/WEF_TheGlobalCompetitivenessReport2019.pdf [Accessed 13- 08-2020] http://www3.weforum.org/docs/WEF_TheGlobalCompetitivenessReport2019.pdf Moshibudi Mokgashi 390202 32 essential for a successful outcome at all stages in the investment process as indicated in chapter 1 of this study. The index assesses three dimensions of regulation that address conflicts of interest: 1) transparency of related-party transactions, 2) shareholders’ ability to sue and hold directors liable for self-dealing, and 3) access to evidence and allocation of legal expenses in shareholder litigation. Ethical behaviour of firms (EBF) This refers to the corporate ethics of companies and ethical behaviour in interacting with other firms, politicians and public officials. A low ranking of this indictor means that firms are exposed to bad ethical behaviour which is not attractive to investors as was seen in the study by Agyemang, Gbettey, Gatsi, and Acquah (2019) referred to in chapter. The World Competitiveness Report uses a rating system of 1 to 7 whereby a variable rated 1 is not doing well and that rated 7 is doing best. For the purpose of this study and given the context in chapter 1 regarding the corporate governance issues such as the agency issue and corrupt dealings between state and corporates in the last decade, only two of these indicators will be used in this study to investigate the link between corporate governance and FDI relationship. These are efficacy of corporate boards and ethical behaviour of firms. 3.1.2. Economic Variables Gross Domestic Product (GDP) GDP has been widely used by several researchers such as Appiah-Kubi et al (2020) in their study of FDI and Corporate governance in West African countries as a control variable because it fundamentally affects FDI inflows particularly in developing countries. Time series studies by Adelopo, Omoteso and Obalola (2009) and Kisto (2017) used GDP to control for market size which is an indicator of economic activity and therefore business potential which then attracts FDI. Whilst some studies tend to also use Gross National Product, Chakrabarti (2001) argues that this economic variable tends to be less appropriate as it includes foreign locations and therefore overestimates the market for products of multinationals in the home country whilst on the other hand underestimating the market by excluding the Moshibudi Mokgashi 390202 33 earnings of multinationals. As such, this study will use the widely accepted GDP growth as a control variable. Trade Openness (TRADE)) Trade openness is a measure of the sum of a country's exports and imports as a share of that country's GDP (in %). Several researchers such as Busse & Hefeker (2005) and Buchanan et al (2011) who conducted similar research have this to control for openness to trade. Empirical evidence by Chakrabarti (2001) suggests that a positive link between FDI and Trade Openness can be expected. The link between FDI and Trade Openness is due the fact that liberal policies and trade facilities of a country presents opportunity for FDI inflow to that country (Ranjan and Agrawal, 2011). It is for these reasons that Trade Openness will be used as the second economic variable. Inflation (INF) Inflation as measured by consumer prices will also be used as a control variable. The world bank (2020) defines this as the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, such as yearly. Literature suggest that inflation can be used as an indicator for the economic stability and healthy monetary and fiscal policy (Blanchard, 2017). He further explains that this is because inflation is as macroeconomic variable that governments can use to conceal domestic economic issues. Several empirical studies have used inflation to account for macroeconomic imbalances. Such studies include that of Busse and Hefeker (2005), Kisto (2017), Sabir, Rafique and Abbas (2019), Kandiero and Wadhawan (2003), Adeoye (2009) and Dondashe and Phiri (2018). All studies found a negative relationship between FDI and inflation except that of Kisto (2017). He argues that inflation can attract FDI as investors can look forward to higher profit margins. 3.1.3. Omitted Variables A few macroeconomic variables identified in literature have been left out from this model due to previous researchers proving that their impact on FDI is inconsistent or Moshibudi Mokgashi 390202 34 that the variable does not affect FDI particularly in South Africa. These are Exchange Rates, Government Expenditure and Real Interest Rates. The omitted variables are dependent variables which are not captured by the model but are collected in the error term. 3.2. Model Specification As time series data will be used for this study, it is important to choose a model that will fit this set of data and its characteristics. Studies of a similar nature that look at time series data such as that of by Adelopo, Omoteso and Obalola (2009) and Kisto (2017) used Ordinary Least Squares for the regression however this had to be used with several other methods to account for the econometric challenges that these methods come with when dealing with the type of data used for this study. These include problems of autocorrelation and endogeneity as identified by Busse and Hefeker (2005) and Appiah-Kubi et al (2020). The Generalized Least Square (GLS) was used by Kandiero and Wahawan (2003) in their study to evaluate the relationship between corporate governance variables and FDI. However, this was done in combination with other techniques to address problems of serial correlation and endogenous effects similar to the studies that used OLS. Theoretically the GMM is superior to the GLS and OLS model when dealing with endogeneity (Smoocha, 2014). Endogeneity refers to a variable in a statistical model that is changed or determined by its relationship with other variables within the model (Blanchard, 2017). Considering that most of the variables to be explored in this study have a close relationship and therefore high possibility of endogeneity, the model selected for this study is the GMM method. The estimated model specification describing the FDI determinants evaluated in this study is as follows: FDIt = α0 + β1INFt + β2TRADEt + β3GDPt-1 + β4POLINt + β4EBFt + β5ECBt + ԑt where α is the constant coefficient and βn are the estimated coefficients of the independent variables whilst ԑt is the error term. Moshibudi Mokgashi 390202 35 The independent variables are represented in the equation as follows: GDP stands for Gross Domestic Product percentage growth TRADE stands for Trade Openness as a percentage of GDP, INF stands for Inflation percentage POLIN stands for the six sub-indicators of Political Risk and Institutions, EBF stands for Ethical Behaviour of Firms, ECB stands for Efficacy of Corporate Boards The empirical design will be such that descriptive statistics are used to establish the nature of the distribution of the data. This will be followed by evaluating the time series properties of the data through investigating the extent of the correlation and its significance. This problem will be dealt with through adding the explanatory variables singly to the regression equation and serial correlation will be tested using the Durbin- Watson value from the regression results. The Augmented Dickey-Fuller (ADF) test for investigating the stationarity of the data will be done because non-stationarity may result in spurious results. The ADF is widely accepted as the most efficient for testing stationarity in comparison to the other methods where the power of the tests are slow if the process is stationary but with a root close to the non-stationary boundary (Brooks, 2014). It is expected that some of these dependent variables will indeed be non-stationary. As such the widely used differencing method will be used to transform them into stationary variables before entering them into the regression. Similar studies such as that of Meyer and Habanabakize (2018) used a similar approach. Finally, explanation of the resulting coefficients of the regression. Once all preliminary tests have been done and all of the issues found have been dealt with, the regression results will be interpreted looking at the coefficients and their signs as well at the p-values which represent significance. Moshibudi Mokgashi 390202 36 CHAPTER 4 4.1. Results The purpose of section is to examine if the variables in question show the anticipated relationships with FDI as described in this text. That is whether political risk, institutional quality and corporate governance may affect FDI inflow levels in South Africa. The details of the dataset as described in chapter 3 are summarised in table 2 with an outlook of the expected sign of impact on FDI. The anticipated signs are as per the situation pertaining to South Africa when it comes to politics, governance and institutions (i.e. refer to chapter 1 of this report for an overview of the perceptions that this study has justified empirically). TABLE 2: SUMMARY OF EXPLANATORY VARIABLES Variable Data Source Symbol Expected sign Dependent Foreign direct investment World Bank FDI Control variables Trade Openness World Bank TRADE + Gross Domestic Product World Bank GDP + Inflation Rate SARB INF - Independent Political Stability and Absence of Violence PRS PV - Voice and Accountability PRS VA - Regulatory Quality PRS RQ - Government Effectiveness PRS GE - Rule of Law PRS RL - Control of Corruption PRS CC - Efficacy of corporate boards WEF ECB - Ethical behaviour of firms WEF EBF - Moshibudi Mokgashi 390202 37 4.1.1. Control Variables For the control variables a positive relationship is expected for Trade and Openness and GDP whilst a negative sign is expected for inflation. This is because a higher rating for trade openness is linked to less trade barriers and therefore less transaction costs for investors (Busse & Hefeker, 2005). Growth in GDP also signals good economic activity and business climate to investors (Adeoye, 2009). The expected sign for CPI Inflation on the other hand is negative because a high inflation rate is associated economic instability of a country. This is because it is considered a signal for economic challenges which bring forth uncertainty regarding reassurance against economic risks new business might face (Adeoye, 2009; Asiedu, 2002 and Balasubramanyam, 1996). 4.1.2. Independent Variables The six political risk and institutions variables are expected to have a negative impact on FDI for South Africa as indicated by the literature review in this text as well as the prevailing conditions. 4.1.3. Descriptive statistics Table 3 looks at the descriptive statistics of the data. A mean average of 1.5 as a percentage of GDP is observed for FDI over the sample period which is 1 point less than the average of the other BRICS countries11. This immediately indicates that there is room for improvement and a typical 10year average that South Africa should strive to achieve (i.e. 2.5%). In this sample period, South Africa has managed to keep its inflation at an average of 5.6 percent which is within the South African Reserve Bank (SARB) target of 3-6 percent12. However, overall, the target has not been achieved year-on- year and this is observed in the sample minimum (-0.69%) and maximum (10,06%). 11 Available at https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=ZA-BR-IN-RU-CN [Accessed 03-08-2020] 12Available at https://www.resbank.co.za/en/home/what-we-do/monetary- policy#:~:text=What%20is%20the%20inflation%20target,the%20Governor%20of%20the%20SARB.[Accessed 25-07-2020] https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=ZA-BR-IN-RU-CN https://www.resbank.co.za/en/home/what-we-do/monetary-policy#:~:text=What%20is%20the%20inflation%20target,the%20Governor%20of%20the%20SARB https://www.resbank.co.za/en/home/what-we-do/monetary-policy#:~:text=What%20is%20the%20inflation%20target,the%20Governor%20of%20the%20SARB Moshibudi Mokgashi 390202 38 Trade Openness average for the sample period is 57% of GDP which is 13 points up from the prior 10 years. The difference can be attributed to the economic state of the country where in the prior 10 years trade restrictions and sanctions were imposed on South Africa due to the political turmoil and therefor instability as discussed in chapter 1.With respect to normality of the data, all the control variables display normal distribution as is confirmed by the probability of the Jarque-Bera statistic for each one of them. Similarly, the two variables under corporate governance also display normal distribution whilst the six variables considered for political risk and institution appear to not be normally distributed. 4.1.4. Unit Root Table 4 shows the results of the Augmented Dickey-Fuller (ADF) test for the variables in this study. From the table, FDI and inflation are stationary at level test whilst GDP, Trade, Government Effectiveness, Rule of Law, Regulatory Quality and Ethical behaviour of firms are integrated at first order. Political stability, Control of corruption, Voice and accountability and Efficacy of Corporate Boards are integrated at second order. 4.1.5. Correlation Matrix In table 5 we see a correlation matrix that confirms that all six variables of political risk and institutions are almost positively correlated at one percent significance level. This is due to their close relationship of these variables in explaining the political risk state and quality of institutions of a country. For example, regulatory quality depends on the Governments’ effectiveness in formulating economically sound policies and upholding them such that they give a positive outlook and affirm investors on business risk. This problem of correlation will be dealt with through adding the six experiment variables one at a time to the model as discussed in chapter 3. In so doing the integrity and explanatory power of the variables is maintained at acceptable levels (Adeoye, 2009). Further correlation is observed between the dependent variable, FDI, and the six variables of political risk and institutions. This is because of the close relationship between political risk and institutions as discussed in this report. Trade Openness is positively correlated to all the experimental variables (political risk, institutions and corporate governance). Moshibudi Mokgashi 390202 39 The correlation between all the experimental variables is due to the fact that trade restrictions or liberation is a result of policies that are influenced by the political stability and state of institutions in a country (Busse and Hefeker, 2005). As such a positive movement in any of these variables will result in a positive movement in trade (Chakrabarti, 2001). In both these cases described above, the correlation coefficients are less than 0,8 and do not pose major challenges for econometric modelling (Blanchard, 2017). However, all regressions run were tested for serial correlation to ensure that it is indeed not a problem and this is observed in table 5 in the coefficients of the R2. Moshibudi Mokgashi 390202 TABLE 3 : DESCRIPTIVE STATISTICS FDI GDP TRADE INF LAW PSTAB QUAL VOICE GOV CORR EBF ECB Mean 1,50 2,61 56,99 5,58 0,34 0,62 0,64 0,71 0,45 0,39 2,16 2,77 Median 1,13 2,65 58,59 5,69 0,38 0,69 0,70 0,83 0,50 0,42 0,00 0,00 Maximum 5,98 5,60 72,87 10,06 0,67 0,79 0,91 0,83 0,75 0,83 4,95 6,12 Minimum 0,23 -1,54 46,67 -0,69 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 Std. Dev. 1,29 1,81 6,57 2,27 0,15 0,24 0,28 0,28 0,18 0,18 2,32 2,96 Skewness 1,89 -0,24 0,17 -0,46 -0,99 -2,18 -1,42 -2,18 -1,81 -0,67 0,11 0,09 Kurtosis 7,08 2,60 2,75 4,25 4,32 5,92 4,10 5,90 5,55 5,05 1,04 1,02 Jarque-Bera 30,90 0,40 0,18 2,41 5,65 27,43 9,23 27,48 19,67 5,99 3,73 3,80 Probability 0,00 0,82 0,91 0,30 0,06 0,00 0,01 0,00 0,00 0,05 0,16 0,15 Sum 35,95 62,61 1367,73 133,97 8,25 14,84 15,36 17,13 10,75 9,33 49,67 63,76 Sum Sq. Dev. 38,36 75,24 992,69 118,50 0,55 1,34 1,76 1,78 0,75 0,72 117,96 193,27 Observations 24 24 24 24 24 24 24 24 24 24 23 23 Source: Computed by Author, 2020 Moshibudi Mokgashi 390202 41 TABLE 4 : THE RESULTS OF AUGMENTED DICKEY-FULLER TEST (ADF) FOR UNIT ROOT Level Differenced Conclusion T-stat P-Value T-stat P-Value FDI 5,05 0,0005 I (0) GDP 2,58 0,1101 5,42 0,0002 I (1) INF 4,29 0,031 I (0) TRADE 2,41 0,1476 5,4 0,0003 I (1) PSTAB 2,06 0,2608 4,7 0,0029 I (2) CORR 1,98 0,2909 4,97 0,0026 I (2) GOV 3,66 0,173 5,12 0,001 I (1) LAW 1,36 0,5754 4,9 0,0015 I (1) QUAL 1,1 0,6886 8,14 0 I (1) VOICE 1,95 0,303 2,77 0,008 I (2) EBF 1,38 0,5708 4,23 0,0037 I (1) ECB 1,34 0,5909 3,013 0,0752 I (2) Source: Computed by Author ,2020 Moshibudi Mokgashi 390202 TABLE 5 : CORRELATION MATRIX Correlation Probability FDI GDP TRADE INF LAW QUAL VOICE GOV CORR PSTAB EBF ECB FDI 1,00 GDP 0,00 1,00 TRADE 0,17 0,01 1,00 INF 0,33 -0,29 0,28 1,00 LAW -0,46 0,04 0,37 -0,23 1,00 QUAL -0,41 0,23 0,40 -0,21 0,92 1,00 VOICE -0,48 -0,02 0,53 -0,23 0,93 0,89*** 1,00 GOV -0,50 -0,01 0,50 -0,17 0,93 0,90 0,99 1,00 CORR -0,48** -0,07 0,46** -0,11 0,89*** 0,82*** 0,94*** 0,96*** 1,00 PSTAB -0,50 0,05 0,44** -0,24 0,94 0,93 0,98 0,99*** 0,93 1,00 EBF 0,03 -0,22 0,66*** 0,24 0,35 0,24 0,43** 0,37* 0,35 0,31 1,00 ECB 0,01 -0,25 0,6***5 0,23 0,34 0,22 0,43 0,37* 0,35 0,31 1,00 1,00 Source: Computed by Author ,2020 Moshibudi Mokgashi 390202 TABLE 6 : REGRESSION RESULTS Dependent Variable FDI Independent Variables C o n tr o l o f C o rr u p ti o n P o lit ic al S ta b ili ty an d A b se n ce o f V io le n ce V o ic e a n d A cc o u n ta b ili ty R u le o f La w G o ve rn m e n t Ef fe ct iv e n e ss R e gu la to ry Q u al it y Explanatory DCORR (1) DPSTAB (2) DVOICE (3) DLAW (4) GOV (5) QUAL (6) Control DTRADE -0,07 (-1,26) -0.07 (-1,26) -0,08 (-1,40) -0,07 (-1,35) -0,06 (-1,12) -0,07 (-,122) INF 0,30* (-0,56) 0,32* (-0,54) 0,33* (3,60) 0,31* (3,88) 0,21* (2,45) 0,19*** (1,92) GDP(-1) 0,30*** (2,97) 0,40*** (1,86) 0,41*** (1,92) 0,44*** (1,85) 0,40*** (2,00) 0,54* (2,36) DECB -0,83 (-1,50) -0,91*** (1,83) -0,82 (-1,68) -0,81 (-1,71 -0,58 (-2,34)* -0,38 (-0,94) DEBF 0,97 (1,47) 1,04*** (1,82) 0.94 (1,65) 0,96 (1,66) 0,64 (2,07)** 0,44 (0,86) GOVERNANCE -3,15* (-2,40) -2,44* (-2,95) -2,41* (-2,93) -4,14* (-2,24) -5,06*** (-1,98) -3,19** (-2,12) Observations 24 24 24 24 24 24 R-squared 0,56 0,48 0,57 0,54 0,51 0,50 Durbin-Watson stat 2,04 2,01 2,03 2,18 1,73 1,78 Computed by Author, 2020 Moshibudi Mokgashi 390202 4.2. Discussion The results of the regression are presented in table 6. The table is presentation of all the regressions run singly with the different explanatory variables. On average the model fits the regression well when using the six political and intuitional variables (models 1 to 6). We see this in the R squared values of the regressions being above fifty percent which confirms the extent (in percentage) that these variables explain FDI inflows and therefor explanatory power of the regression. The Durbin-Watson values for the models are around 2 which confirms no serial correlation. For this data sample, the anticipated relationships of FDI with the control variables are not all as expected. Because of the bi-directional relationship that exists between GDP and FDI especially in developing countries (Iamsiraroj,2016), this variable was lagged by one period to better capture its impact on FDI inflows for SA. The results show that there is a positive and significant relationship between FDI and GDP (observed at a ten percent significance level in all models except for model 6). This is a finding that has also been identified by studies of Elabadawi and Mwega (1997) and they have concluded that economic growth of SADC countries is one of the reasons they continue to outperform other African countries when it comes to attracting FDI inflow. The basis of their conclusion is that growth potential of a country is a major attraction to investors. Inflation on the other hand shows a positive and significant relationship at one percent significance level with FDI. The study expected the coefficient of Inflation to be negative coefficient. Asongu, Akpan and Isihak (2018) attribute the positive sign to the fact that the BRICS and MINT countries tend to have high inflation and as such plays a lesser role when investors make decisions in these countries. This is because of the nature of macro-economic instability that is inherent to developing countries. Trade on the other hand has no significant relationship with FDI and the coefficient sign in all models contradicts the expectation. Moshibudi Mokgashi 390202 45 The results of models 1 to 6 show that the six political risk and institutions explanatory variables have negative and significant impact on FDI. 4.2.1 Model 1 Model 1 indicates a negative and significant relationship between control of corruption and FDI in SA. This is consistent with studies by Al-Sadig (2009) and Gangi and Abdulrazak (2012) who all found a similar outcome. For the case in South Africa, a one-point increase in level of corruption has the potential of decreasing FDI inflow by 3.15 points according to the regression results. There exists a relationship between quality of institutions and corruption (hence the correlation of 0.82) and as such failure in institutional quality bears higher corruption levels. It is for this reason that we see the efficacy of corporate boards also impacting FDI negatively (although not significant). Whilst South Africa practices worldwide recognised reporting codes and standards to ensure transparency the institutional structures in place to enforce standard compliance are failing (refer to chapter 2 for corporate corruption cases noted). A further disadvantage to the economy resulting from corruption levels in the country is that a study by Ugur and Dasgupta (2011) shows that a one-point increase in Corruption Perceptions Index (by Transparency International13) is associated with a decline in GDP growth of 0.6% of the said country. With this study and many others confirming the positive relationship between FDI and GDP growth, South Africa is losing FDI inflow opportunity because of corruption perceptions. 4.2.2 Model 2 In model 2 we note that if the country’s politics are not stable and there is a tolerance for violence the impact on FDI will be negative. This outcome is consistent with studies done by Schneider and Frey (1985), Gangi and Abdulrazak (2012), Khan and Akbar (2013) as well as Naude and Krugell (2003). 13 Details about Corruption Perceptions Index. Available at https://www.transparency.org/en/cpi/2020/index/nzl [Accessed 24-02-2021] https://www.transparency.org/en/cpi/2020/index/nzl Moshibudi Mokgashi 390202 46 A study by the Reserve Bank of Zimbabwe on FDI in Southern African Development Community14 (SADC15) also found that there exists a negative relationship between FDI and political instability. These findings are in line with literature review that states political instability has negative impact on FDI because of the perception that the returns of investors will be exposed to risk due to uncertainty and possible policy reversals. Egger and Winner, (2005) attribute the negative relationship between political instability and FDI inflow to weak law enforcement, red tape and inefficient regulatory structures for SADC countries. 4.2.3 Model 3 Model 3 also shows a negative relationship between Voice and Accountability with FDI. This observation is also found in studies by Adelopo, Omoteso and Obalola (2009) and Gangi and Abdulrazak (2012). Voice and Accountability enables media and citizens to hold the state accountable for their performance and as such a pre- requisite for this to take place is access to information. Sabir, Rafique and Abbas (2019) found that whilst the relationship between Voice and Accountability with FDI is not significant for developing countries compared to developed countries, FDI flows to democratic countries that promote and enable freedom of speech and independent media. This implies that if media freedom and access to information are limited, the positive spillover of removing information asymmetry will not realize and as such that impacts FDI (Dutta and Roy,2009). This is because with information asymmetry comes a risk to investors in the sense that they do not have reliable information for which they can make decisions. Another downside to information asymmetry is the high cost of equity that will be incurred by companies in the country simply because of this risk (Lambert, Leuz and Verrechia,2012). 14 Foreign Direct Investment In Southern African Development Community, Paper Prepared for the Committee of the Central Bank Governors In SADC By Reserve Bank of Zimbabwe. Available at https://www.sadcbankers.org/Lists/News%20and%20Publications/Attachments/190/FDI%20PAPER%20FINAL. pdf [Accessed 30-10-2020] 15 SADC member states include Angola, Botswana, Democratic Republic of Congo, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Seychelles, South Africa, Swaziland, Tanzania, Zambia and Zimbabwe https://www.sadcbankers.org/Lists/News%20and%20Publications/Attachments/190/FDI%20PAPER%20FINAL.pdf https://www.sadcbankers.org/Lists/News%20and%20Publications/Attachments/190/FDI%20PAPER%20FINAL.pdf Moshibudi Mokgashi 390202 47 4.2.4 Model 4 Model 4 which explains the relationship between FDI inflow and Rule of law implies that if the host country does not have a transparency and legal system does not create economic security for investment is, FDI is impacted negatively. This is no deviation from the literature review and studies with a similar finding include that of Habtemarian (2004), 4.2.5 Model 5 In model 5 we find that the weak government effectiveness of South Africa also has had a negative impact on FDI. Privatisation is impacted by weak government effectiveness. This finding is expected because as with political instability; a weak government effectiveness tends to not follow through with policies that favour FDI inflow. This is consistent with findings by Habtemarian (2004) in Sub-Saharan African countries. 4.2.6 Model 6 For Model 6, Regulatory quality is also found to have a negative impact on FDI inflows and this finding is similar to studies by Khan and Akbar (2013). Whilst most studies have found regulatory quality to not have a significant impact on developing countries, the implication for the study outcome is that South Africa is expected to adopt and maintain market friendly policies to be able to attract FDI. Moshibudi Mokgashi 390202 48 CHAPTER 5 5.1. Conclusions and Recommendations This thesis has examined the relationship between FDI, Politics, Institutions and Governance in the context to South Africa. In general, the finding is that poor governance and political instability have impact on FDI inflow with respect to South Africa. The econometric estimates show that tolerance for corruption, government ineffectiveness, lack of rule of law, political instability, poor regulatory quality and accountability have negative impact on FDI. In answering the question of the extent of impact, the finding is that of all the six variables, lack Control of corruption and Rule of Law impact FDI inflows the most. Macro-economic indicators such as the level of inflation and GDP growth were found to have positive impact on FDI. Trade Openness however was found to be negative and not significant in the context of South Africa during this period. This may be so because the period of the study is just after the lifting of trade sanctions that were imposed on South Africa and new trade laws were introduced and as such investors were not concerned about the future of trade in South Africa. The results also show that government needs to reduce political instability and employ strategies that will assure investors less risk to their returns through policy implementations that are good for doing business. Furthermore, policies implemented should be upheld to improve investor confidence in the stability of the country. To be able to unlock a potential 0.6% improvement in GDP growth hindered by lack of control of corruption, South Africa needs to strengthen the institutions and structures in place to enforce and manage reporting standards as well as the agency problem. Companies also need to ensure strong boards of directors through diversification as well as controlled rotations to minimize the chances of self-dealing that can result from familiarity. Transparency and access to information needs to be a priority especially with the rapid increase of access to internet. Public and private companies need to control the narrative more now than ever before and this requires strengthened efficacy of corporate boards and ethical behaviour of firms to ensure that information shared to the public is representative of truth and accountability and as such eliminates asymmetry. Moshibudi Mokgashi 390202 49 A study by Drabek and Payne (2002) reveals that high levels of non-transparency prevent host countries from reaching their full potential in attracting FDI. Their finding is that on average a one-point increase in transparency ranking could result in up to 40 percent increase in FDI. South Africa’s current transparency ranking is 69 with a score 44 out 100 placing us more to the highly corrupt end of the scale and therefor there is opportunity to improve16. It is very important to note that transparency goes with good governance and as such as alluded to in the previous paragraph, South African government needs to ensure that the institutional structures in place to regulate and enforce good governance are effective. A study by Staats and Biglaiser (2011) indicates that improvement in protection of property, impartial courts and judicial independence could lead to FDI inflow of up to 27 percent in developing countries. To unlock this potential, South Africa should improve protection of property rights to motivate investors, improve business entry through systems with fewer barriers, improve market policy to facilitate for volume growth and finally show conformation and loyalty to contracts to guarantee commitment to investors and motivate them to seek opportunities of growth in the country. 5.2. Study limitations and Further Research Data availability with respect to the two variables representing corporate governance was limited to a shorter period than that of the study. This could be a possibility as to why there is no consistency in significance of this variables in the six regression models that were run. Future research can improve on this work as more reliable data becomes available. The author proposes research into the areas from which South Africa attracts FDI inflows (home countries) and whether those countries find these explanatory variables important. This would assist in targeted policy implementation and improvement to be able to maintain those existing relationships. Similarly, research on potential home countries that provide FDI inflow to peers but not SA could unlock new investment. 16 Details about Corruption Perceptions Index. Available at https://www.transparency.org/en/cpi/2020/index/nzl [Accessed 09-03-2021] https://www.transparency.org/en/cpi/2020/index/nzl Moshibudi Mokgashi 390202 50 Emphasis should be on what elements of politics, institutions and governance are important for them to make investments into these host countries (SADC, BRICS etc). Whilst this research and others confirm that South Africa has potential to improve FDI inflows, a look at South Africa’s share of world output would further be able to justify whether the FDI levels SA is currently attracting matches its output and therefor importance to the global economy or not. 5.3. Ethical Considerations The nature of this study and the data collection methodology is such that there is little to no risk of ethical issues. The data was collected from well-known institutions that publish this data for public consumption. A list of the sources is indicated below: • World Development Indicators (WDI), a database from the World Bank • Worldwide Governance Indicators (WGI) • World Competitiveness Report deduced from the World Economic Forum’s Opinion Survey (WEF) According to the Graduate School of Business Administration of the University of Witwatersrand’s risk level categories, such sources or methodology of collecting data is categorised under no risk. This because there is no contact with human participants as quantitative datasets are used. Moshibudi Mokgashi 390202 51 Appendices Appendix A DEVELOPING ECONOMIES SOURCES OF EXTERNAL INCOME,2009-2018 (BILLIONS OF DOLLARS) Source: UNCTAD World Economic Report 201917 Appendix B GROSS SAVINGS AND GROWTH RATES OF SELECTED EMERGING MARKETS Year Brazil China India Malaysia Singapore South Africa Gross Savings (% of GDP) 1980-90 19.9 36.1 20.9 24.8 39.0 24.4 1991-00 16.7 40.7 24.0 34.7 47.6 16.6 2001-10 17.4 46.2 33.4 35.5 43.9 16.8 2011-19 15.2 46.5 32.6 29.3 45.5 15.8 Average GDP Growth (%) 1980-90 3.0 9.8 5.7 5.9 7.8 2.2 1991-00 1.7 10.0 5.7 7.2 7.3 1.4 2001-10 2.5 9.9 5.9 2.6 3.5 2.1 2011-19 - 0.13 6.8 5.2 3.6 2.5 0.01 GROSS SAVINGS AND GROWTH RATES OF SELECTED EMERGING MARKETS Source: IMF 17 Available at https://unctad.org/en/PublicationsLibrary/wir2019_en.pdf [Accessed 03--08-2020] https://unctad.o