Governance Infrastructure and Foreign Direct Investment Flows

Governance Infrastructure and Foreign Direct Investment Flows A Fresh Look at the World and its Regions Rotterdam, 2009 RSM Erasmus University Dr. Jordan Otten David Eberle, 336553 John de Geus, 5079 Pablo Mandelz, 33664

Introduction The notion that the world s economies are becoming increasingly integrated with the greater global economy, a concept commonly referred to as globalization, has become an accepted part of mainstream thinking. Foreign direct investment (FDI) is an important factor in globalization, since it is a means for creating direct, stable and long-lasting economic links between countries (OECD, 2008). The OECD asserts that, within a proper policy framework, FDI assists host countries in developing local enterprises, promotes international trade through access to markets and contributes to the transfer of technology and know-how. FDI also has an impact on labour and financial markets and influences other factors of economic performance (OECD, 2008). This implies that, without a proper governance infrastructure, FDI will not contribute effectively to a country s economic development as described above. However, exactly how governance infrastructure affects FDI flows between countries is a subject of continuing research. Various authors have conducted research on the effect of a country s governance infrastructure on its FDI inflows and outflows. Most notably, Globerman and Shapiro (2002) have examined this relationship for a large number of countries using a combination of indicators from various sources and have determined that governance infrastructure is indeed an important determinant of FDI inflows and outflows. Other, more recent research includes Globerman, Shapiro and Tang (2004), specifically focusing on emerging and transition European countries, and Gani (2007), who focuses on Asian, Latin American and Caribbean countries; these authors likewise conclude that there is a positive correlation between governance infrastructure and FDI inflows and outflows. An important assumption made by the abovementioned authors is that the more developed a country s governance infrastructure, as measured by indicators that find their basis in a country s laws (de jure or rule-based indicators), the fairer the business climate and the less risk investors face. Khanna et al. (2006) question this assumption by researching the relationship between governance infrastructure and globalization using de facto (outcome-based) governance indicators. Page 2 of 42

When these indicators, which are based on the actual situation in a country instead of on its laws, are used, no significant correlation between governance infrastructure and foreign direct investment is found (Khanna et al., 2006). The United Nations Conference on Trade and Development (UNCTAD) regularly publishes country statistics on foreign direct investment. Recent research into the relationship between governance infrastructure and FDI utilizing the most recent statistics (from 2006 to 2008), which include more countries than previous UNCTAD statistics, is not readily available in the literature on this topic. With this paper we aim to add to and enrich the body of knowledge on this topic by employing the most recent statistics and a larger number of governance indicators. For this purpose, we have developed a new measure of governance, the Erasmus Governance Index (EGI), an additive index which incorporates four existing governance indices. Additionally, we use this index to explore the variations in the level of correlation that occur in the world s different regions. Key Concepts For the purposes of this paper we will adhere to the definition of foreign direct investment as presented by the OECD (Organisation for Economic Development and Cooperation), a consortium of 30 wealthy member countries dedicated to the promotion of global economic development. According to the OECD, FDI reflects the objective of establishing a lasting interest by a resident enterprise in one economy (direct investor) in an enterprise (direct investment enterprise) that is resident in an economy other than that of the direct investor (OECD, 2008). Numerous definitions of corporate governance exist, reflecting the diversity of approaches to corporate governance that have been developed over time. Shleifer and Vishny (997) define corporate governance in terms of the principal-agent problem: Corporate governance deals with the ways in which suppliers of finance to corporations assure themselves of getting a return on their investment. Zingales (997) takes the principal-agent perspective a step further, arguing that it Page 3 of 42

only makes sense to discuss corporate governance on the assumption that contracts are inherently incomplete, thus defining corporate governance as: the complex set of constraints that shape the ex-post bargaining over the quasi-rents generated by a firm. Aoki (2000) instead maintains a much broader definition, stating, Corporate governance is the structure of rights and responsibilities among the parties with a stake in the firm. For the purposes of this paper, we will adhere to a more recent definition as presented by the OECD: Corporate governance involves a set of relationships between a company s management, its board, its shareholders and other stakeholders. Corporate governance also provides the structure through which the objectives of the company are set, and the means of attaining those objectives and monitoring performance are determined. (OECD, 2004) Globerman and Shapiro (2002) and Globerman, Shapiro and Tang (2004) use the term governance infrastructure to describe the framework of institutions and rules that shape corporate governance within a single country. For the purposes of this paper, we will use the definition as presented by Globerman, Shapiro and Tang (2004): Governance infrastructure refers to a country s political, institutional and legal environment, as well as to the policies that accompany them. Measuring Governance There are several indices claiming to measure governance infrastructure. They can generally be divided into rule-based and outcome-based indicators of governance (Kaufmann & Kraay, 2008). The advantage of the latter is that they offer direct information regarding a country s current governance environment and hence are of particular relevance to decision-makers considering whether to invest in that country. This study incorporates a relevant selection of indices presented in a paper conducted by Kaufmann and Kraay (2008), which provides an informative overview of 2 major governance surveys. The fact that not all governance indicators are available for a large number of countries leads to a trade-off between the number of indices used and the amount of countries observed. Since the aim Page 4 of 42

of this study is to provide a detailed look at the world s different regions, including a high number of countries is crucial. Moreover, some surveys (e.g. Economist Intelligence Unit) are not available to the public and are therefore not included in this study. Four indices have been included as components of the Erasmus Governance Index (EGI); these are discussed in the following paragraphs. One of the most prominent indices is the Worldwide Governance Indicator (WGI), an aggregate index developed by Kaufmann, Kraay and Mastruzzi. The index consists of six dimensions: voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption. Their values are based on 44 individual variables published by 33 different organizations around the globe (Kaufmann et al., 2009). Another important index is the Gallup World Poll (GWP), a large cross-country survey measuring corporate governance on four different dimensions. Interestingly, the names of these dimensions are equal to some of the WGI s. However, since the correlation between the two indices is low, GWP is a useful component of our additive index, particularly due to the fact that the GWP uses household surveys to gather its data, thus adding a different perspective on corporate governance infrastructure. The third index, labeled HER, concerns economic freedom and is published by The Heritage Foundation. This organization assesses countries on ten different dimensions, including business freedom and property rights. (The Heritage Foundation, 2009). A fourth index, the Empowerment Rights Index (CIR), is based on the Cingranelli-Richards Human Rights Dataset, which provides quantitative information on 5 internationally recognized human rights (Cingranelli & Richards, 2009). This is an additive index comprised of seven relevant human rights indicators, including foreign movement and workers rights. (Cingranelli & Richards, 2008). Page 5 of 42

Since the correlation between the different dimensions within an index, as well as the correlation between the indices themselves (except for the direct correlation between the dimensions with the same name of WGI and GWP), is rather high (see appendix), constructing an additive index for this research seems reasonable (Globerman & Shapiro, 2002). The correlations between dimensions within the same index are especially strong, which suggests that the examination of individual components of a single index would add little value to this research. Each index, which is incorporated into the EGI, uses a slightly different scale. Whereas, for instance, the WGI ranges from -2.5 to 2.5, the CRI uses integers between 0 and 4. Because the EGI is an additive index, all components to be incorporated have to be normalized prior to their summation; otherwise, the values are not comparable and cannot be summed. alizing the indexes results in values between 0 and for each component. Since the total number of individual components used in the EGI is 2, the EGI has a value between 0 and 2 for each country. The amount of investment flowing into and out of a country (FDI and FDO respectively), is likely to be influenced by other factors besides governance infrastructure. Factors such as GDP and human development in general are likely to play a major role when measuring FDI and FDO. If such control variables are not included, the importance of governance infrastructure is likely to be overstated. According to Globerman and Shapiro (2002), GDP and the Human Development Index (HDI), published by the United Nations Development Programme, are good candidate variables to control for other factors determining FDI and FDO. The correlation between the EGI and the HDI is rather high (r=0.678); Globerman and Shapiro present similar findings (r=0.69) (Globerman & Shapiro, 2002). Page 6 of 42

Statistical Framework The aim of this paper is to analyze the impact of governance infrastructure, as measured by the EGI, on FDI and FDO of a particular country. The following equations are used to estimate FDI and FDO: () ln FDI = a + b ln GDP + error (2) ln FDI = a + b ln GDP + b 2 HDI + b 3 EGI + error (3) ln FDI = a + b ln GDP + b 2 HDI + b 3 WGI + b 4 WGI 2 + b 5 WGI 3 + b 6 WGI 4 + b 7 WGI 5 + b 8 WGI 6 + b 9 HER + b 0 GWP + b GWP 2 + b 2 GWP 3 + b 3 GWP 4 + b 4 CRI + error (4) ln FDO = a + b ln GDP + error (5) ln FDO = a + b ln GDP + b 2 HDI + b 3 EGI + error (6) ln FDO = a + b ln GDP + b 2 HDI + b 3 WGI + b 4 WGI 2 + b 5 WGI 3 + b 6 WGI 4 + b 7 WGI 5 + b 8 WGI 6 + b 9 HER + b 0 GWP + b GWP 2 + b 2 GWP 3 + b 3 GWP 4 + b 4 CRI + error Ordinary least squares (OLS) regression is used to estimate the two equations stated above. Since regional differences regarding the impact of governance structure on capital flows are possible, multiple models containing different countries are tested. The standard model includes all countries whereas additional estimations focus on specific regions around the globe, including OECD, Europe, Asia, Africa, Middle East and Latin America. Countries listed in OECD are not included again in other regional models. Each model is first tested trying to explain inflows and outflows respectively by GDP only (see equations and 4). The second test estimates capital flows by GDP, the HDI and the EGI (see equations 2 and 5). The third estimation replaces the EGI with its individual components (see equations 3 and 6). The statistical model presented in this paper uses the logarithm of both FDI and GDP as suggested by Globerman and Shapiro (2002). Without this linear transformation absolute numbers would greatly complicate the outcome and hinder meaningful interpretation of the estimated coefficients. The values for both FDI and FDO are obtained from the World Investment Report published by UNCTAD. Annual flows seem to be a better candidate for a measurement of investment than stocks, since latter tend to change only little from year to year. Flows, however, can increase or decrease Page 7 of 42

rapidly depending on the global economic situation. The recent recession, for instance, led to a sharp fall in FDI worldwide (The Economist, 2009). This research therefore computes average flows between 2006 and 2008, hence mitigating abrupt drops and rises. Data for the chosen governance infrastructure indicators are acquired directly through each publisher s website. For all indices the latest data available is used (see table ). Table : Index Data Index Year Source CIR 2007 CIRI Human Rights Data Project (http://ciri.binghamton.edu/) GWP 2008 Gallup (http://www.gallup.com/consulting/worldpoll/24046/about.aspx) HDR 2009 UNDP (http://hdr.undp.org/en/statistics/) HER 2009 Heritage Foundation (http://www.heritage.org/index/) WGI 2008 Worldbank (http://info.worldbank.org/governance/wgi/index.asp) Results FDI Model The results for the FDI model, consisting of 04 valid observations for all indices, are illustrated in table 2 (see next page). The first model () presents the results for a regression of ln(fdi) on ln(gdp) for 35 countries. The cross-section estimation of this particular model (R² = 0.n) is moderate but still suggests that GDP acts as a control variable for several economic factors to a certain extent. The coefficient on the term ln(gdp), however, is highly significant. Page 8 of 42

Table 2: FDI Model Regression Before testing the model (2), which tests a regression of ln(fdi) on the additive Erasmus Governance Index (EGI) plus the control indices Human Development Index (HDI) and ln(gdp) for 34 countries, a scatter plot analysis helps to determine the residual errors (see figure on the next page). The plot, showing the residual errors of model (2), illustrates that the scatter is not perfectly random; yet it neither points to a clear direction. Hence, the regression estimation can be conducted for model (2) and (3). Page 9 of 42

Figure : FDI Model Scatter Plot In model (2), while the coefficient of ln(gdp) is significant, the same does not hold for EGI and HDI. Still, the cross-section estimation of model (2) (R² = 0.45) has improved as compared to the previous estimation. The same results are observed in model (3) regarding the cross-section estimation. The split in model (3) of the EGI into its components (allind EGI) improves the level of explanation for most of the country groups (the results of the individual indices are reported in the appendix). This effect could be analyzed in future research. Some models, for example model (3) for Latin America or model (2) for Asia, have especially high cross-section estimations (R² = 0.98 and R² = 0.92 respectively). These high levels could be explained by the limited numbers of observations available for the two models (N = 9 and N = 8 respectively). Since the number of countries is fixed, however, the statistical model cannot be improved by adding more observations. The coefficient of EGI in model (2) measures the percentage change of FDI as EGI increases by point. For the region World the coefficient of EGI is estimated at 0.23, which indicates that a better governance infrastructure leads to an increase in foreign direct investment. This result is not significant for the overall FDI model; yet, it is consistent for all regions and is especially significant Page 0 of 42

for the region Latin America. Interestingly, the coefficient for EGI is also higher for Latin America than for most other regions. The cross-section estimation of model (2) is, as indicated above, moderate. Yet, for the regions Asia (R² = 0.92), Africa (R² = 0.63) and Latin America (R² = 0.75) the cross-section estimation is high and indicates a very high level of explanation. Europe (R² = 0.8) has a low level of explanation in model (2) but a higher one in model (3) (R² = 0.63). This effect cannot be explained at this stage and needs be researched in more depth. Besides that, only Middle East shows a weak level of explanation (R² = 0.29), which could be related to cultural aspects even though the low number of observations (N = 9) might influence the results as well. This will be discussed later in this paper. FDO Model The regression results of the FDO model () with ln(fdo) as the dependent variable suggests a high level of explanation of GDP as a control variable (R² = 0.63) (see table 3 on the next page). Ln(GDP) is significant with an estimated coefficient of.565. A % increase of GDP thus leads to a rise of.565% in FDO. The regression model () for FDO as well as for FDI suggests a positive correlation between direct investment flows and a country s GDP. Page of 42

Table 3: FDO Model Regression The FDO model (2) is also analyzed using a scatter plot for the sample world (see figure 2 on the next page). Although the scatter is not perfectly random it does not point to a clear direction. Yet, a weak direction can be observed between an increase in ln(fdo) and a rise in the value of error terms. Page 2 of 42

Figure 2: FDO Model Scatter Plot FDO model (2) indicates significant coefficients both of ln(gdp) and EGI. With a coefficient estimated at 0.529 for EGI the test states that an increase of EGI by point leads to an increase of 0.529% in foreign direct outflows. The coefficient for EGI is significant for the world as a whole, and for the regions OECD and Latin America. Again, as for the FDI model, Latin America exhibits a higher EGI coefficient than the other regions. The HDI, however, is not significant in any region. The cross-section estimation observed in model (2) is high (R² = 0.7). This result shows again an improvement from model () to model (2) in the level of explanation. The cross-section estimation improves even further in model (3) regarding the sample World. The split of the created index EGI into its individual components also leads to a higher R². In model (3) the R² equals 0.79. For the region split a much higher R² is observed with ln(fdo) as the dependent variable. R² for model (2) is 0.84 for OECD, 0.9 for Asia and 0.70 for Latin America. For Europe and Africa the level of explanation is lower but still acceptable (R² = 0.44 and R² = 0.54 respectively). Similar to the FDI model, some regions feature very high cross-section estimates in model (3). The Asian region, for instance, even has R² =.00. Sample size seems, again, to be the underlying reason for these high cross-section estimates. Page 3 of 42

Regional Split While comparing the two models for FDI and FDO with their regional split, the FDO model s level of explanation is clearly higher than the one of the FDI model (see table 4). The average R 2 of the six regions used for the FDI model is 0.49, whereas the average for the FDO model equals 0.67. Table 4: Regional Regression Results Especially for FDI some of the regions exhibit a very low R 2, particularly in Europe. An additional regression tests whether the R 2 differs if the number of the observations for the region Europe would be split accordingly to more narrowly defined regions, namely Europe (New) and Central Asia (see table 5). Table 5: European Split Regression Results Page 4 of 42

The region Europe is split into Europe (New) containing 5 countries (Albania, Belarus, Bosnia- Herzegovina, Bulgaria, Croatia, Cyprus, Latvia, Lithuania, Macedonia, Moldova, Montenegro, Romania, Serbia, Slovenia and Ukraine) and Central Asia consisting of 5 countries (Kazakhstan, Armenia, Azerbaijan, Georgia and Russia). The levels of explanation for the new regions (R 2 = 0.77 for Europe (New) and R 2 =0.99 for Central Asia ) are much higher. This suggests large differences in the effect of governance infrastructure on FDI between these two regions. Another remarkable observation illustrated in figure 3 is that the percentage alteration of in- and outflows are quite symmetrical as the EGI moves up or down, i.e. an increase (decrease) in EGI leads to both a higher (lower) FDI and FDO, except for the region Middle East. Africa, however, exhibits a different trend, as the coefficient for EGI is negative for both FDI as well as FDO. Figure 3: Summary Regional Coefficients 2.50 2.00.50.00 0.50 0.00 Ln FDI Ln FDO - 0.50 -.00 -.50 Africa and the Middle East In Africa, in contrast with most other regions, FDI inflows and outflows decrease as governance infrastructure becomes more developed. We theorize that this is because foreign aid boosts the FDI inflows and outflows of African countries with poor governance. As governance improves, this flow of aid is reduced. Page 5 of 42

The budget of many African countries consists, for a large part, of foreign aid (sometimes coupled with FDI inflows). A condition for receiving this foreign aid is that the countries improve their governance infrastructure (according to Western guidelines which may not be appropriate for underdeveloped African states). However, once the countries reach a certain level of (governance) development, the aid is reduced and there is actually less money in the country's budget (Asiedu, 200). One could conclude from this that as governance infrastructure improves, the consequent reduction of aid to a country means that there is less money available to spend on FDI outflows. Additionally, the institutional reforms that Western countries demand in return for aid actually give rise to new, more sophisticated forms of corruption (Szeftel, 2000). Finally, the countries are being forced to transition from informal relational governance systems to formal governance systems and there is a period of transition from one to the other where neither system works well (Li, 2004). During this period of transition (which can last for years) the governance infrastructure is better according to the governance indicators used in our research, but actual outcomes are worse. In the Middle East region FDI outflows are lower when the governance infrastructure of a country is more developed. In recent years the Gulf States (including the United Arab Emirates and Kuwait) and Saudi Arabia, countries in the Middle East with some of the most highly developed governance infrastructures, have experienced a major economic boom based on real estate (Oxford Analytica, 2009). We theorize that it has been more lucrative for this select group of countries to invest at home instead of abroad. For this reason, these countries display lower relative FDI outflows than the other countries in the Middle East region, despite having a better governance infrastructure. Conclusions Our research reveals that there is a connection between a country s governance infrastructure and the foreign direct investment flowing into its economy. The link between the additive EGI specially created for this research and the logarithm of FDI, however, is not significant except for Latin American countries, where the impact of governance structure on FDI seems to be much stronger Page 6 of 42

than for other regions. The FDO model, on the other hand, produces more reliable results. The coefficient of EGI is significant for the world as a whole, for the OECD region and for Latin America. Special attention should be given to models with extraordinary high cross-section estimates (e.g. the Asian FDO model (3) has R² =.00), where small sample sizes could explain these high values. In general, the coefficient estimation suggests that with better governance structure a country both attracts more FDI as well as invests more abroad (higher FDO). Exceptions are countries in Africa and the Middle East, where coefficients turn out to be negative. We theorize that as African countries reach a certain level of governance infrastructure, a reduction of foreign aid ultimately leads to a decline in FDI. The reason for less FDO with advancing governance infrastructure in the Middle East is less clear at this point and requires further research. Of further interest is also the difference between Eastern European (labeled as Europe (New) ) and Central Asian countries. When combined in one regional cluster the results were not significant. Yet, as soon as the cluster is split, the coefficient estimates of the regions move into opposite directions, and Eastern Europe s coefficient for FDO becomes highly significant. This paper adds value to previously conducted research by incorporating a larger number of major governance indicators and combining them in the additive Erasmus Governance Index (EGI). Also, the use of the most recent data available provides a fresh view on the relation between governance infrastructure and foreign direct investment flows. This actual view, however, also raises a plethora of new questions to be analyzed in future research. Especially regional differences in the directions and magnitudes of the coefficients are an interesting topic to look into in more detail. One possibility could be the incorporation of a measurement of culture in the regression as a control variable; Hofstede s framework of cultural dimensions could be useful in this regard. Page 7 of 42

Bibliography Aoki, M. (2000). Information, Corporate Governance and Institutional Diversity: Competitiveness in Japan, the U.S.A. and Transnational Economies. Oxford: Oxford University Press. Asiedu, E. (200). On the Determinants of Foreign Direct Investment in Developing Countries: Is Africa Different? World Development, 30 (), 07-9. Cingranelli, D. L., & Richards, D. L. (2009 2-September). CIRI Human Rights Data Project. Retrieved 2009 23-November from CIRI: http://ciri.binghamton.edu/ Cingranelli, D. L., & Richards, D. L. (2008). Short Variable Descriptions for Indicators in the Cingranelli-Richards (CIRI) Human Rights Dataset. CIRI. Gani, A. (2007). Governance and Foreign Direct Investment Links: Evidence from Panel Data Estimations. Applied Economics Letters, 4, 753-756. Globerman, S., & Shapiro, D. (2002). Global Foreign Direct Investment Flows: The Role of Governance Infrastructure. World Development, 30 (), 899-99. Globerman, S., Shapiro, D., & Tang, Y. (2004). Governance and Foreign Direct Investment in Emerging and Transition European Countries. Working Paper, Western Washington University. Kaufmann, D., & Kraay, A. (2008). Governance Indicators: Where Are We, Where Should We Be Going? The World Bank Research Observer, 23 (). Kaufmann, D., Kraay, A., & Mastruzzi, M. (2009). Governance Matters VIII: Aggregate and Individual Governance Indicators 996-2008. The World Bank. Khanna, T., Kogan, J., & Palepu, K. (2006). Globalization and Similarities in Corporate Governance: a Cross-Country Analysis. The Review of Economics and Statistics, 88 (), 69-90. Li, S. (2004 5-February). Poor Governance Does not Repel Investors. Retrieved 2009 25- November from FDI Magazine: http://www.fdimagazine.com/news/fullstory.php/aid/585/poor_governance_does_not_repel_investo rs.html OECD. (2008). OECD Benchmark Definition of Foreign Direct Investment, 4th Edition. Paris: OECD. OECD. (2004). OECD Principles of Corporate Governance. Paris: OECD. Oxford Analytica. (2009). GULF STATES: Realigned property sector set to grow. Oxford: Oxford Analytica Ltd. Schleifer, A., & Vishny, R. (997). A Survey of Corporate Governance. The Journal of Finance, 52 (2), 737-783. Szeftel, M. (2000). Between Governance and Underdevelopment: Accumulation and Africa's Catastrophic Corruption. Review of African Political Economy, 27 (84), 304. Page 8 of 42

The Economist. (2009 8-September). Shrinking assets. The Economist. The Heritage Foundation. (2009). 2009 Index of Economic Freedom. Retrieved 2009 23-November from The Heritage Foundation: http://www.heritage.org/index/ UNCTAD. (2009). World Investment Report: Transnational Corporations, Agricultural Production and Development. Geneva: United Nations Publication. Zingales, L. (997). Corporate Governance. Working Paper, University of Chicago. Page 9 of 42

Appendix Correlations Figure 4: Correlations of Individual Components of EGI Page 20 of 42

FDI- Model (3): allegi Index Coefficients I. World Coefficients a Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -.379 2.5 -.549.584 Human Development Index -.78 2.27 -.0 -.808.42 ln GDP PPP (World Bank).854.46.605 5.83.000 WGI: Voice & Accountability 2.397 3.746.56.640.524 WGI: Political Stability.696.928.046.36.79 WGI: Effectiveness -.37 5.9 -.00 -.026.979 WGI: Regulatory Quality.977 4.97.33.47.639 WGI: Rule of Law -2.8 5.364 -.98 -.524.60 WGI: Control of Corruption 3.350 4.902.237.683.496 HER: Overall -2.985 3.200 -.4 -.933.353 GWP: Voice & Accountability -2.66.809 -.88 -.446.5 GWP: Effectivess 2.839 2.005.48.46.60 GWP: Rule of Law -2.355 3.62 -.095 -.745.458 GWP: Control of Corruption 2.859.829.90.563.2 CRI: Empowerment Rights Index -.75.672 -.069 -.427.670 Page 2 of 42

Coefficients a Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -.379 2.5 -.549.584 Human Development Index -.78 2.27 -.0 -.808.42 ln GDP PPP (World Bank).854.46.605 5.83.000 WGI: Voice & Accountability 2.397 3.746.56.640.524 WGI: Political Stability.696.928.046.36.79 WGI: Effectiveness -.37 5.9 -.00 -.026.979 WGI: Regulatory Quality.977 4.97.33.47.639 WGI: Rule of Law -2.8 5.364 -.98 -.524.60 WGI: Control of Corruption 3.350 4.902.237.683.496 HER: Overall -2.985 3.200 -.4 -.933.353 GWP: Voice & Accountability -2.66.809 -.88 -.446.5 GWP: Effectivess 2.839 2.005.48.46.60 GWP: Rule of Law -2.355 3.62 -.095 -.745.458 GWP: Control of Corruption 2.859.829.90.563.2 CRI: Empowerment Rights Index -.75.672 -.069 -.427.670 a. Dependent Variable: ln FDI Page 22 of 42

II. OECD Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) 20.77 2.40.942.366 Human Development Index -47.634 29.285 -.685 -.627.32 ln GDP PPP (World Bank) 2.334.026.834 2.274.044 WGI: Voice & Accountability 47.098 46.780.235.007.336 WGI: Political Stability -3.359 22.074 -.09 -.52.882 WGI: Effectiveness 9.009 26.897.34.335.744 WGI: Regulatory Quality -8.863 39.09 -.245 -.227.825 WGI: Rule of Law 7.064 29.36.293.24.84 WGI: Control of Corruption -.509 26.444 -.580 -.435.672 HER: Overall -28.896 20.585 -.68 -.404.88 GWP: Voice & Accountability -4.705.34 -.768 -.32.23 GWP: Effectivess 3.338 0.349.089.323.753 GWP: Rule of Law -5.684 6.582 -.434 -.946.365 GWP: Control of Corruption 8.077 4.95.24.273.229 CRI: Empowerment Rights Index 5.05 8.525.225.588.568 a. Dependent Variable: ln FDI Page 23 of 42

Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) 20.77 2.40.942.366 Human Development Index -47.634 29.285 -.685 -.627.32 ln GDP PPP (World Bank) 2.334.026.834 2.274.044 WGI: Voice & Accountability 47.098 46.780.235.007.336 WGI: Political Stability -3.359 22.074 -.09 -.52.882 WGI: Effectiveness 9.009 26.897.34.335.744 WGI: Regulatory Quality -8.863 39.09 -.245 -.227.825 WGI: Rule of Law 7.064 29.36.293.24.84 WGI: Control of Corruption -.509 26.444 -.580 -.435.672 HER: Overall -28.896 20.585 -.68 -.404.88 GWP: Voice & Accountability -4.705.34 -.768 -.32.23 GWP: Effectivess 3.338 0.349.089.323.753 GWP: Rule of Law -5.684 6.582 -.434 -.946.365 GWP: Control of Corruption 8.077 4.95.24.273.229 CRI: Empowerment Rights Index 5.05 8.525.225.588.568 a. Dependent Variable: ln FDI b. Selecting only cases for which Country Code = OECD Page 24 of 42

III. Europe Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant).87 29.374.040.969 Human Development Index 6.098 22.63.374.726.500 ln GDP PPP (World Bank) -.29.454 -.047 -.089.933 WGI: Voice & Accountability 66.874 49.589 2.80.349.235 WGI: Political Stability 6.78 4.32.88.432.684 WGI: Effectiveness 9.8 35.405.648.542.6 WGI: Regulatory Quality -24.099 38.779 -.858 -.62.562 WGI: Rule of Law -45.576 49.567 -.489 -.99.400 WGI: Control of Corruption 20.784 8.395.652.255.809 HER: Overall -8.239 8.555 -.227 -.444.676 GWP: Voice & Accountability -5.967 5.056 -.74 -.06.337 GWP: Effectivess -0.979 37.039 -.20 -.296.779 GWP: Rule of Law -5.53 34.056 -.0 -.62.878 GWP: Control of Corruption 5.334 2.977.674.82.290 CRI: Empowerment Rights Index -22.680 6.748 -.565 -.354.234 a. Dependent Variable: ln FDI Page 25 of 42

Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant).87 29.374.040.969 Human Development Index 6.098 22.63.374.726.500 ln GDP PPP (World Bank) -.29.454 -.047 -.089.933 WGI: Voice & Accountability 66.874 49.589 2.80.349.235 WGI: Political Stability 6.78 4.32.88.432.684 WGI: Effectiveness 9.8 35.405.648.542.6 WGI: Regulatory Quality -24.099 38.779 -.858 -.62.562 WGI: Rule of Law -45.576 49.567 -.489 -.99.400 WGI: Control of Corruption 20.784 8.395.652.255.809 HER: Overall -8.239 8.555 -.227 -.444.676 GWP: Voice & Accountability -5.967 5.056 -.74 -.06.337 GWP: Effectivess -0.979 37.039 -.20 -.296.779 GWP: Rule of Law -5.53 34.056 -.0 -.62.878 GWP: Control of Corruption 5.334 2.977.674.82.290 CRI: Empowerment Rights Index -22.680 6.748 -.565 -.354.234 a. Dependent Variable: ln FDI b. Selecting only cases for which Country Code = Europe Page 26 of 42

IV. Asia Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -8.896.000.. Human Development Index 5.35.000.368.. ln GDP PPP (World Bank).03.000.07.. WGI: Voice & Accountability -2.245.000 -.63.. WGI: Political Stability -.52.000 -.93.. WGI: Effectiveness -3.30.000 -.366.. WGI: Regulatory Quality -8.028.000 -.687.. WGI: Rule of Law 7.204.000.663.. HER: Overall.545.000.663.. GWP: Voice & Accountability 3.825.000.428.. GWP: Effectivess.458.000.0.. GWP: Rule of Law -8.622.000 -.55.. GWP: Control of Corruption.383.000.049.. CRI: Empowerment Rights Index 2.20.000.284.. a. Dependent Variable: ln FDI b. Selecting only cases for which Country Code = Asia Page 27 of 42

Excluded Variables b Model Beta In t Sig. Partial Correlation Collinearity Statistics Tolerance WGI: Control of Corruption. a....000 V. Africa Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -5.347 3.597 -.487.55 Human Development Index.63 3.80.5.507.69 ln GDP PPP (World Bank).30.288.864 3.925.00 WGI: Voice & Accountability -2.334 5.027 -.70 -.464.648 WGI: Political Stability 5.584 3.200.495.745.099 WGI: Effectiveness -7.380 8.99 -.50 -.900.38 WGI: Regulatory Quality 7.093 8.200.454.865.399 WGI: Rule of Law -7.486 0.79 -.02 -.78.04 WGI: Control of Corruption 4.353 6.874.873 2.088.052 HER: Overall.098 4.487.005.022.983 Page 28 of 42

GWP: Voice & Accountability.074 2.25.008.035.973 GWP: Effectivess.064 2.488.095.428.674 GWP: Rule of Law -2.633 4.446 -.47 -.592.56 GWP: Control of Corruption.052 2.768.072.380.709 CRI: Empowerment Rights Index -.78 2.55 -.087 -.362.722 a. Dependent Variable: ln FDI b. Selecting only cases for which Country Code = Africa VI. Middle East Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -4.939.000.. Human Development Index 7.672.000.48.. HER: Overall 2.346.000.54.. GWP: Voice & Accountability.807.000.072.. GWP: Effectivess 4.669.000.86.. GWP: Rule of Law -3.934.000 -.469.. CRI: Empowerment Rights Index -2.558.000 -.352.. a. Dependent Variable: ln FDI b. Selecting only cases for which Country Code = MiddleEa Page 29 of 42

Excluded Variables b Model Beta In t Sig. Partial Correlation Collinearity Statistics Tolerance ln GDP PPP (World Bank). a....000 WGI: Voice & Accountability. a....000 WGI: Political Stability. a....000 WGI: Effectiveness. a....000 WGI: Regulatory Quality. a....000 WGI: Rule of Law. a....000 WGI: Control of Corruption. a....000 GWP: Control of Corruption. a....000 VII. Latin America Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -2.773.625 -.706.39 Human Development Index 0.76 3.89.783 2.87.030 ln GDP PPP (World Bank).600.089.669 6.735.00 Page 30 of 42

WGI: Voice & Accountability -3.75 3.053 -.297 -.229.265 WGI: Political Stability 6.24 2.45.577 2.855.029 WGI: Effectiveness 2.056 3.90.7.644.543 WGI: Regulatory Quality 3.5 4.04.239 3.96.09 WGI: Rule of Law -.200 5.277 -.08 -.038.97 WGI: Control of Corruption -2.726 8.702 -.0 -.462.94 HER: Overall -4.85 2.062 -.236-2.030.089 GWP: Voice & Accountability -3.289.57 -.329-2.094.08 GWP: Effectivess -3.69.58 -.232-2.088.082 GWP: Rule of Law -2.674 3.445 -.45 -.776.467 GWP: Control of Corruption.46 3.943.83 2.907.027 CRI: Empowerment Rights Index -4.293.332 -.470-3.222.08 a. Dependent Variable: ln FDI b. Selecting only cases for which Country Code = LatinAme FDO- Model (3): allegi Index Coefficients VIII. World Coefficients a Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. Page 3 of 42

(Constant) -6.89 2.403-2.868.005 Human Development Index 2.487 2.06.09.207.23 ln GDP PPP (World Bank).24.36.63 8.903.000 WGI: Voice & Accountability -.429 3.578 -.069 -.399.690 WGI: Political Stability -3.848.809 -.92-2.27.036 WGI: Effectiveness 2.68 4.68.672 2.696.008 WGI: Regulatory Quality 2.740 3.855.38.7.479 WGI: Rule of Law -2.079 5.055 -. -.4.682 WGI: Control of Corruption -.55 4.562 -.082 -.332.74 HER: Overall -4.836 3.069 -.42 -.576.9 GWP: Voice & Accountability.783.695.098.052.296 GWP: Effectivess -5.797 2.04 -.20-2.84.006 GWP: Rule of Law -4.253 2.967 -.33 -.433.55 GWP: Control of Corruption.274.69.067.787.434 CRI: Empowerment Rights Index.993.627.075.60.543 a. Dependent Variable: ln FDO Page 32 of 42

IX. OECD Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -22.507 3.206-7.020.000 Human Development Index 3.937 4.385.08.898.389 ln GDP PPP (World Bank).0.54.749 7.65.000 WGI: Voice & Accountability 5.773 7.005.788 2.252.046 WGI: Political Stability -2.463 3.305 -.52 -.745.472 WGI: Effectiveness.687 4.027.049.7.868 WGI: Regulatory Quality 7.80 5.856.378.226.246 WGI: Rule of Law -2.32 4.390 -.83 -.529.608 WGI: Control of Corruption -4.259 3.960 -.409 -.076.305 HER: Overall -3.322 3.082 -.35 -.078.304 GWP: Voice & Accountability -.563.667 -.056 -.338.742 GWP: Effectivess 5.769.550.294 3.723.003 GWP: Rule of Law 5.526 2.483.292 2.226.048 GWP: Control of Corruption.253 2.26.032.9.907 CRI: Empowerment Rights Index -2.258.276 -.93 -.769.05 a. Dependent Variable: ln FDO Page 33 of 42

Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -22.507 3.206-7.020.000 Human Development Index 3.937 4.385.08.898.389 ln GDP PPP (World Bank).0.54.749 7.65.000 WGI: Voice & Accountability 5.773 7.005.788 2.252.046 WGI: Political Stability -2.463 3.305 -.52 -.745.472 WGI: Effectiveness.687 4.027.049.7.868 WGI: Regulatory Quality 7.80 5.856.378.226.246 WGI: Rule of Law -2.32 4.390 -.83 -.529.608 WGI: Control of Corruption -4.259 3.960 -.409 -.076.305 HER: Overall -3.322 3.082 -.35 -.078.304 GWP: Voice & Accountability -.563.667 -.056 -.338.742 GWP: Effectivess 5.769.550.294 3.723.003 GWP: Rule of Law 5.526 2.483.292 2.226.048 GWP: Control of Corruption.253 2.26.032.9.907 CRI: Empowerment Rights Index -2.258.276 -.93 -.769.05 a. Dependent Variable: ln FDO b. Selecting only cases for which Country Code = OECD Page 34 of 42

X. Europe Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -9.745 8.7 -.93.287 Human Development Index 3.09 6.66.097.50.637 ln GDP PPP (World Bank).792.405.393.957.08 WGI: Voice & Accountability 36.556 3.795 2.074 2.650.045 WGI: Political Stability 3.739 3.98.53.939.39 WGI: Effectiveness 2.49 9.849.980 2.82.08 WGI: Regulatory Quality -2.272 0.788 -.09 -.2.84 WGI: Rule of Law 8.750 3.789.827.360.232 WGI: Control of Corruption -76.040 22.643-3.222-3.358.020 HER: Overall -5.796 5.62 -.25 -.23.32 GWP: Voice & Accountability 2.488 4.88.753 2.982.03 GWP: Effectivess 5.544 0.304.37.538.64 GWP: Rule of Law -2.73 9.474 -.067 -.288.785 GWP: Control of Corruption 6.23 3.60.368.72.46 CRI: Empowerment Rights Index -8.932 4.659 -.832 -.97.3 a. Dependent Variable: ln FDO Page 35 of 42

Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -9.745 8.7 -.93.287 Human Development Index 3.09 6.66.097.50.637 ln GDP PPP (World Bank).792.405.393.957.08 WGI: Voice & Accountability 36.556 3.795 2.074 2.650.045 WGI: Political Stability 3.739 3.98.53.939.39 WGI: Effectiveness 2.49 9.849.980 2.82.08 WGI: Regulatory Quality -2.272 0.788 -.09 -.2.84 WGI: Rule of Law 8.750 3.789.827.360.232 WGI: Control of Corruption -76.040 22.643-3.222-3.358.020 HER: Overall -5.796 5.62 -.25 -.23.32 GWP: Voice & Accountability 2.488 4.88.753 2.982.03 GWP: Effectivess 5.544 0.304.37.538.64 GWP: Rule of Law -2.73 9.474 -.067 -.288.785 GWP: Control of Corruption 6.23 3.60.368.72.46 CRI: Empowerment Rights Index -8.932 4.659 -.832 -.97.3 a. Dependent Variable: ln FDO b. Selecting only cases for which Country Code = Europe Page 36 of 42

XI. Asia Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -0.736.000.. Human Development Index 20.975.000.96.. ln GDP PPP (World Bank).560.000.243.. WGI: Voice & Accountability 36.50.000.386.. WGI: Political Stability -.307.000 -.06.. HER: Overall -2.42.000 -.467.. GWP: Voice & Accountability 4.045.000.28.. GWP: Effectivess 22.882.000.97.. GWP: Rule of Law -38.382.000 -.236.. GWP: Control of Corruption.652.000.053.. CRI: Empowerment Rights Index -5.30.000 -.374.. a. Dependent Variable: ln FDO b. Selecting only cases for which Country Code = Asia Excluded Variables b Model Beta In t Sig. Partial Correlation Collinearity Statistics Tolerance Page 37 of 42

WGI: Effectiveness. a....000 WGI: Regulatory Quality. a....000 WGI: Rule of Law. a....000 WGI: Control of Corruption. a....000 XII. Africa Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -20.600 9.72-2.2.087 Human Development Index 6.453 4.76.750.5.36 ln GDP PPP (World Bank).786.605.456.300.250 WGI: Voice & Accountability -4.987.862 -.743 -.263.262 WGI: Political Stability -.7 7.485 -.099 -.229.828 WGI: Effectiveness 8.7 8.36.904.09.355 WGI: Regulatory Quality -27.543 2.003 -.288 -.3.247 WGI: Rule of Law -28.706 25.044 -.296 -.46.304 WGI: Control of Corruption -24.428 27.80 -.057 -.878.420 HER: Overall 43.922 2.752.473 2.09.099 Page 38 of 42

GWP: Voice & Accountability 2.380 6.02.064 2.056.095 GWP: Effectivess -.720 7.498 -.680 -.563.79 GWP: Rule of Law 2.636 6.435.834.36.245 GWP: Control of Corruption -23.58 7.602 -.234-3.094.027 CRI: Empowerment Rights Index.66 5.700.927 2.038.097 a. Dependent Variable: ln FDO b. Selecting only cases for which Country Code = Africa XIII. Middle East Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant).036.000.. Human Development Index 46.975.000.87.. HER: Overall -9.96.000 -.387.. GWP: Voice & Accountability -23.892.000 -.63.. GWP: Effectivess -.885.000 -.03.. GWP: Rule of Law -2.32.000 -.428.. CRI: Empowerment Rights Index -.757.000 -.479.. a. Dependent Variable: ln FDO b. Selecting only cases for which Country Code = MiddleEa Page 39 of 42

Excluded Variables b Model Beta In t Sig. Partial Correlation Collinearity Statistics Tolerance ln GDP PPP (World Bank). a....000 WGI: Voice & Accountability. a....000 WGI: Political Stability. a....000 WGI: Effectiveness. a....000 WGI: Regulatory Quality. a....000 WGI: Rule of Law. a....000 WGI: Control of Corruption. a....000 GWP: Control of Corruption. a....000 a. Predictors in the Model: (Constant), CRI: Empowerment Rights Index, Human Development Index, GWP: Effectivess, GWP: Rule of Law, HER: Overall, GWP: Voice & Accountability b. Dependent Variable: ln FDO XIV. Latin America Coefficients a,b Unstandardized Coefficients Standardized Coefficients Model B Std. Error Beta t Sig. (Constant) -4.25 5.39-2.748.05 Human Development Index.488 4.420.430.797.470 Page 40 of 42

ln GDP PPP (World Bank).350.382.765 3.533.024 WGI: Voice & Accountability 7.64 0.26.77.724.60 WGI: Political Stability 8.798 8.77.460.009.370 WGI: Effectiveness 9.799 9.788.443.00.373 WGI: Regulatory Quality 34.025 2.74.809 2.795.049 WGI: Rule of Law.547 6.056.028.034.974 WGI: Control of Corruption -55.258 24.669-2.683-2.240.089 HER: Overall -4.066 7.379 -.25 -.55.6 GWP: Voice & Accountability 6.276 5.58.362.24.324 GWP: Effectivess -6.573 4.66 -.620-3.978.06 GWP: Rule of Law -4.558 9.547 -.42 -.477.658 GWP: Control of Corruption 26.707 2.25.070 2.86.094 CRI: Empowerment Rights Index -2.863 5.679 -.87-2.265.086 a. Dependent Variable: ln FDO b. Selecting only cases for which Country Code = LatinAme Page 4 of 42

Contribution Assessment The team for this paper consists of three members who worked together closely in a dynamic group process. Before embarking on the actual research, each member s team role was defined according to his particular strengths. The tasks were subsequently divided among the team members as illustrated in the table below. To assure progress and avoid postponement of tasks, milestones were set dynamically and objectives were distributed among the team members. This flexible planning approach worked very well and gave each member the flexibility he needed. All team members did their utmost to fulfill their individual responsibilities. Regular meetings assured a structured and coherent approach to the research. Member Team Role Tasks David Eberle John de Geus Pablo Mandelz Coordinator Shaper Completer Innovator Evaluator Completer Contact Person Implementer Specialist Methodology Data Collection Conclusion Layout Introduction Theoretical Framework Literature Review Editing Data Collection Regression Model Design Test Results Group Process Tulder, R. v. (2007). Skill Sheets. Amsterdam: Pearson Education Benelux. Page 42 of 42