The Impacts of Ethnic, Linguistic and Religious Fractionalization on State Fragility in Africa

Transcription

1 Journal of Marketing and Management, 7 (2), 47-66, Nov The Impacts of Ethnic, Linguistic and Religious Fractionalization on State Fragility in Africa Aye Mengistu Alemu Assistant Professor of Economics, SolBridge International School of Business, Samsung 1-dong, Dong-gu, Daejeon , South Korea ayem2011@solbridge.ac.kr Tel: Abstract This study assembled a large longitudinal dataset for 43 African countries ranging from 1995 to 2014 (n = 860) with information on ethnic, linguistic and religious fractionalization along some economic and demographic variables, and indices of state fragility in Africa. Random Effect model was used to explore the effects of ethnic, linguistic and religious fractionalization on state fragility. Results show that except ethnic fractionalization that has a weak direct effect on state fragility, no evidence has been found for linguistic and religious fractionalization to directly contribute to state fragility. However, the study has found a strong evidence for the interactive effects of the three domains of fractionalization with political right and civil liberty to significantly aggravate state fragility. This implies that ethnic, linguistic and religious composition do not in themselves drive fragility significantly. Rather, it is lack of political right and civil liberty that act as moderating mechanisms and exacerbate the negative effects of ethnic, linguistic and religious fractionalization on state fragility. This study therefore demonstrates the need to look other factors beyond ethnic, linguistic and religious fractionalization in order to understand what explains the differences in state fragility across countries in Africa. Introduction African ethnic diversity has been blamed for the major source of conflicts and civil wars that are still prevailing in the region. According to Cocodia (2008), one characteristic of ethnic conflicts in African is that the states are originally heterogeneous created by communal contenders, and this genuine circumstance led to a fact that different groups fighting to seek power and control over the other. Similarly, Bailey (1994) states that the political map of Africa is a western colonial creation, drawn by western powers with little regard to the boundaries of historic ethnic homelands or the ethnic compositions of the subject population, and today these artificial or multi-ethnic nations lack the internal political cohesion necessary for survival as nations. With approximately 2500 languages (or 30 per cent of the world s living languages), Africa is one of the most linguistically diverse continents (Adegbija, 1994; Grimes, 2000; Batibo, 2005). Multilingualism is thus seen as a liability and a burden, particularly when considered in the context of the amount of resources needed to promote the use of multiple languages in the domains of education, media, law and administration, business and commerce and international communication. Hence, some researchers argue that ethnic, linguistic and religious heterogeneity within a nation might have negative effects on societal outcomes since it can create barriers to communication and exchange, factions and rivalries, and internal conflict (Easterly and Levine, 1997; Easterly, 2001). Although it seems reasonable to assume that ethnic, linguistic and religious 47

2 Journal of Marketing and Management, 7 (2), 47-66, Nov heterogeneity have led to major conflicts in many African countries, it is still questionable that to what extent these three domains of heterogeneity have played a negative role in African conflicts and sate fragility. State Fragility Index Fragile states, refers to a fundamental failure of the state in assuring basic security, maintaining rule of law and justice, or providing basic services and economic opportunities for their citizens, unstable and weak governance, a persistent condition of extreme poverty, lack of territorial control, and high propensity to conflict and civil war (Bertocchi, 2010). The State Fragility Index and Matrix" is a new measure of state capabilities which is annually published by the Center for Systematic Peace. It rates the global system's 162 countries (with population greater than 500,000) and then, combines scores on the eight indicators and ranges from 0 (no fragility) to 25 (extreme fragility) (Marshall and Cole, 2014). One billion people, including about 340 million of the world s extreme poor, are estimated to live in this small group of between fragile countries, located mainly in Africa, that are falling behind and falling apart (Collier, 2007). The distribution of fragile states in the world is clearly shown in figure 1 below. By the same token, Stewart and Brown (2009) also propose a three-pronged definition of fragility: states may be fragile because they lack authority, fail to provide services or lack legitimacy. By the same token, OECD (2008) characterized fragile states as: unable to meet [their] population s expectations or manage changes in expectations and capacity through the political process. States that fail to meet basic needs and to keep societal expectations and state capacity in equilibrium can also fail to establish reciprocal state-society relations or create a binding social contract. Accordingly, about 26 of 49 countries identified as fragile by OECD (2009) are in Sub-Saharan Africa. As it can be seen in figure-1, the top 5 highly fragile states in the world include Central African Republic, Democratic Republic of Congo (DRC), South Sudan, Sudan (north) and Afghanistan. Also, figure-2 shows the level of state fragility from the least fragile state (Mauritius with state fragility index of 1) to the highest fragile state (Central African Republic with state fragility index of 24) in

3 Figure 1: State Fragility and Warfare in the Global System, mid-2014 Journal of Marketing and Management, 7 (2), 47-66, Nov Source: Marshall and Cole (2014) While many fragile states in sub-saharan Africa have progressed significantly since the 1990s, too many countries have still not been able to break out of the fragility trap. For instance, seven countries Cameroon, Ethiopia, Mozambique, Niger, Nigeria, Rwanda, and Uganda have made relatively more progress in building resilience. These countries strengthened their institutions, maintained macroeconomic stability and increased public investment and improved social services. Nevertheless, nine countries have not made progress and six countries have regressed (Côte d Ivoire, Eritrea, Madagascar, Malawi, Mali, Zimbabwe). On the other hand, Mauritius, Seychelles, Botswana, South Africa and Namibia are found to be in a relatively better position in terms of stability according to their order (European Report on Development 2009). 49

4 Figure 2: State Fragility in Africa (in 2014) Journal of Marketing and Management, 7 (2), 47-66, Nov SFI in 2014 Mauritius Botswana Namibia Morocco South Africa Tanzania Kenya Ghana Benin Gabon Senegal Zambia Madagascar Algeria Equatorial Guinea Egypt Mozambique Libya Congo Republic Togo Sierra Leone Burkina Faso Liberia Malawi Eritrea Cameroon Gambia Angola Rwanda Mauritania Zimbabwe Mali Nigeria Niger Burundi Uganda Chad Ethiopia Somalia Sudan DRC Central African Republic

5 Journal of Marketing and Management, 7 (2), 47-66, Nov Ethnic, linguistic and religious fractionalization The Fractionalization dataset was compiled by Alberto Alesina and associates, and measures the degree of ethnic, linguistic and religious heterogeneity in various countries with more than one million inhabitants. The dataset was used in Alesina et al. (2003) to test the effects of fractionalization on the quality of institutions and economic growth. The index varies from 0 (complete homogeneity) to 1 (complete heterogeneity). As can be shown in table 1, except Papua New Guinea, all other 4 countries at the top of the list of highly ethnically heterogeneous countries in the world are in Africa. On the contrary, North Korea, South Korea, Japan, Tunisia and Italy are the top 5 homogenous countries in the world. Figure 3 shows the extent of ethnic fractionalization from the top ethnically fractionalized country (Uganda with an ethnic fractionalization index of 0.93) to the least ethnically fractionalized country (Tunisia with ethnic fractionalization index of 0.04). Table 1: The top 5 highly ethnically heterogeneous countries and the top 5 least ethnically heterogeneous countries in the world Top 5 Highly Ethnically Heterogeneous Countries Ethnic Fractionalization Index Papua New Guinea Tanzania Democratic Republic of Congo Uganda Liberia Top 5 Least Ethnically Heterogeneous Countries North Korea South Korea Japan Tunisia Italy Source: Compiled from Fearon (2003). Turning to linguistic heterogeneity, 18 of the most linguistically-heterogeneous countries in the world are again from Sub-Saharan Africa (SSA) and followed by India, with a linguistic fractionalization index of The most linguistically homogenous countries in the world are South Korea and North Korea, followed by Yemen. Whereas in Africa, the top linguistically fractionalized country is again Uganda with a linguistic fractionalization index of 0.92 and the least linguistically fractionalized country is again Tunisia with a linguistic fractionalization index of 0.01 (figure 3). Finally, turning to religious heterogeneity, the most religiously-diverse countries in the world are South Africa, the United States, and Australia, and the most homogenous countries in terms of religion are Yemen, Somalia, Morocco, Turkey, and Algeria (Alesina, et al. 2003). Coming to Africa, the top religiously fractionalized country is South Africa with a religious fractionalization index of 0.86 and the least religiously fractionalized country is Somalia with a religious fractionalization index of zero (figure 4). 51

6 Uganda Tanzania Togo Liberia Cameroon Kenya South Africa CDR CAR Nigeria Gambia Ethiopia Mozambique Angola Sierra Leone Burkinafaso Sudan Senegal Namibia Ghana Malawi Morocco Mauritius Zimbabwe Algeria Botswana Mauritania Equatorial Guniea Burundi Seychelles Libya Somalia Madagascar Egypt Tunisia Uganda Liberia Madagascar CDR Chad Cameroon Kenya Nigeria CAR Somalia Angola Zambia South Africa Burkinafaso Tanzania Ethiopia Sudan Mozambique Senegal Ghana Namibia Mauritius Botswana Zimbabwe Equatorial Guniea Algeria Rwanda Burundi Seychelles Egypt Tunisia Journal of Marketing and Management, 7 (2), 47-66, Nov Figure 3: Ethnic Fractionalization in Africa Figure 3: Linguistic Fractionalization in Africa

7 South Africa Ghana CAR Kenya Cote d'ivoire Nigeria Cameroon CDR Mauritius Uganda Angola Ethiopia Botswana Sierra Leone Burundi Rwanda Liberia Sudan Seychelles Egypt Niger Mali Senegal Gambia Libya Mauritania Tunisia Algeria Morocco Somalia Figure 4: Religious Fractionalization in Africa Journal of Marketing and Management, 7 (2), 47-66, Nov In order to get the full picture of the relationship between ethnic, linguistic and religious fractionalization and state fragility, it is essential to also look at the political right and civil liberty condition in those countries of interest. Accordingly, figure 5 & 6 illustrate the political right index and the civil liberty index, respectively. Political Right Freedom House (2016) compiles the political right index for each country every year. The political rights index is measured on a scale of 1 to 7. Countries with a rating of 1 enjoy a wide range of political rights including free and fair elections, competitive political parties, no discrimination among the polity, and the ability of minority groups to participate in the governing process. On the other hand, countries with a rating of 7 have few or no political rights because of severe government oppression, sometimes in combination with civil war. They may also lack an authoritative and functioning central government and suffer from extreme violence or rule by regional warlords. 53

8 Journal of Marketing and Management, 7 (2), 47-66, Nov Figure 5: Political Right Index in Africa in 2015 Tunisia 1 South Africa Benin Zambia Sierra Leone Niger Madagascar Togo Mozambique Cote d'ivoire Zimbabwe Mali Rwanda Libya Egypt Congo Republic Angola Sudan Gambia Eritrea Chad Burundi

9 Journal of Marketing and Management, 7 (2), 47-66, Nov Civil Liberties Similarly, ccountries with a rate of 1 enjoy a wide range of civil liberties, including freedoms of expression, assembly, association, education, and religion. They have an established and generally fair legal system that ensures the rule of law (including an independent judiciary), allow free economic activity, and tend to strive for equality of opportunity for everyone, including women and minority groups. On the other hand, countries with a rating of 7 have few or no civil liberties. They allow virtually no freedom of expression or association, do not protect the rights of detainees and prisoners, and often control or dominate most economic activity. 55

10 Journal of Marketing and Management, 7 (2), 47-66, Nov Figure 6: Civil Liberty Right Index in Africa in 2015 Senegal 2 Mauritius Botswana Tunisia Seychelles Burkina Faso Togo Niger Morocco Mali Cote d'ivoire Uganda Algeria Gabon Congo Republic Libya Ethiopia Chad Burundi Eritrea CAR Somalia The rest of the paper is organized as follows. The next section explains the theoretical background and hypotheses. The research methodology and data set used in this study will be discussed. It then followed by the regression results, main findings and discussion. The last section of the paper includes a conclusion, the 56

11 Journal of Marketing and Management, 7 (2), 47-66, Nov implications of the paper and its main contributions and finally the limitations of the paper and venue of future research will be discussed. Theoretical Background and Hypotheses Mauro (1995) was the first to introduce the concept of ethnic diversity and empirically examined its effect on economic growth. Employing the Ethno-Linguistic Fractionalization (ELF) Index, he concluded that ethno-linguistic fractionalization leads to a greater probability of political instability and impedes economic development. Similarly, Easterly and Levine (1997) concluded that Africa's poor economic performance was attributed to its high level of ethnic diversity. Some other previous studies such as Alesina and Drazen (1991), Alesina and Rodrick (1994), Alesina and Spoloare (1997 ), Shleifer and Vishny (1993 ) revealed both the direct and indirect effects of ethnic fragmentation on economic performance and quality of institutions by claiming that in heterogeneous societies there is a greater likelihood for competition among interest (ethnic) groups for the provision of public goods, leading to poor public policy decisions. By the same token, some studies argue that ethnic, linguistic and religious fractionalization within a nation might have negative effects on societal outcomes since it can create barriers to communication and exchange, factions and rivalries, and internal conflict (Easterly and Levine, 1997; Easterly, 2001). According to Gupta (2005), fragile states often experience crises of legitimacy in the sense that citizens may not accept the state s basic right to rule. State legitimacy and the development of trust between state and society have long been considered a critical dimension of state-building processes in the political science literature. Furthermore, violence in fragile states is often triggered by competition for resources, such as land or water. Obvious cases where high levels of natural resource supply have contributed to the onset of conflict include DRC, Sierra Leone, Angola, and Sudan (Vallings and Moreno-Torres, 2005). However, it should be noted that not all natural resource rich countries have experienced fragility and a good example is Botswana. Instead, violent conflict is usually sparked by the political manipulation of groups competing for power (including control of natural resource supply). Large supplies of natural resource wealth provide strong incentives for those with access to the supplies to skew the distribution of wealth through patronage (Vallings and Moreno-Torres, 2005). On the other hand, Vallings and Moreno-Torres (2005) argue that ethnic, linguistic and religious composition do not in themselves drive fragility. Rather, it is the political manipulation of these factors that can impact on state stability. This manipulation is more likely in states with weak institutions. For instance, Gettleman (2009) noted that Somalia is a paradox-unified on the surface, poisonously divided beneath. It is one of the world s most homogenous nation-states, with nearly all of its estimated 9 to 10 million people sharing the same language (Somali), the same religion (Sunni Islam), the same culture and the same ethnicity. Similarly, in spite of the homogeneity demonstrated here by the Rwandan people, they have had a troubled history of political and social upheaval (Pottier, 2002:5). This implies that in spite of remarkable homogeneity, these two nations remain conflict prone. The absence of political right and civil liberty may lead to horizontal inequality among groups defined by identity and this in turn will lead to resentment and tensions, and can foster group mobilization like what happened in Rwanda, Côte d'ivoire, Sudan, etc. On the other hand, Switzerland is one of the most religiously diverse countries in Europe, with a large number of Catholic adherents (42%); Evangelical Reformed Church (33%); and a fast-growing Islamic population (4%) (Buehler et al., 2009; Maclaren, 2007). Studies have shown a demonstrated overlap between linguistic region and religious affiliation (Buehler et al., 2009). The elites in Africa and other developing countries use political manipulation of ethnic groups that leads to ethnic conflict and ultimately civil wars (Alesina et al., 2003 ; Alesina and La Ferrara, 2005). By the same token, Collier (2000 ) and Easterly (2001 ) show that in advanced democracies countries that have high political rights tend to have 'good institutions' or at least adequate institutional mechanisms to mitigate potential problems that may stem from ethnic/linguistic/religious and other social tensions. That is why 57

12 Journal of Marketing and Management, 7 (2), 47-66, Nov Lindemann (2008) argues that the ability of postcolonial states in Sub-Saharan Africa to maintain political stability depends on the ability of the ruling political parties to overcome the historical legacy of social fragmentation. Inclusive elite bargains involve a ruling party that integrates a broad coalition of key elites by defining inclusive access to state structures (jobs) and state resources (rents). It is not always true that high ethnic fractionalization is associated with high state fragility and vice versa. As can be seen in table 2, India, Canada, Switzerland and Israel have much higher ethnic fractionalization than some African countries such as Burundi, Rwanda, Egypt and Libya. However, these African countries are by far more fragile than India, Canada, Switzerland and Israel. This is perhaps because of ethnic heterogeneity does not in itself causes state fragility. Rather, it is the political manipulation of these factors that can impact on state stability. In other words, the more the countries are un-democratic (less political right and less civil right), the more the politicians and the elites manipulate ethnic, linguistic and religious heterogeneity for their self-interest. On the other hand, despite India has more than two thousand ethnic groups, it is also the world's most populous democracy and perhaps that might be the reason why this country is not as fragile as other African countries such as Burundi and Rwanda that have much lower ethnic heterogeneity compared to India. Table 2: Comparison of some selected African and non-african countries in terms of ethicfractionalization and state fragility index Selected non-african countries Ethnic Fractionalization Index State Fragility Index India Canada Switzerland Israel Selected African countries Burundi Rwanda Egypt Libya Source: Compiled from Fearon (2003) & Marshall and Cole (2014) Likewise, despite a considerable number of African countries mainly from Sub-Saharan Africa are the world s most conflict-intensive region, there are also few countries in the continent that promoted good governance, peace and political stability. For instance, Mauritius, Seychelles, Botswana, South Africa and Namibia are found to be in a relatively better position in terms of stability according to their order (European Report on Development 2009). Thus, the question is that why some African did states experience civil war, while others have managed to maintain political stability? With this background, we can hypothesize the relationship between ethnic, linguistic and religious fractionalization and state fragility as follows. Main Hypotheses Hypothesis 1:- Ethnic, linguistic and religious fractionalization are significantly and positively contributing to state fragility in Africa. 58

13 Journal of Marketing and Management, 7 (2), 47-66, Nov Hypothesis 2:- Political Right and Civil Liberty moderate the relationship between ethnic, linguistic and religious fractionalization and state fragility such that the higher the political right and civil liberty, the lower the impacts of ethnic, linguistic and religious fractionalization on state fragility. Research Methodology The model to investigate the effects of ethnic, linguistic and religious fractionalization on state fragility in Africa is constructed for a panel of 43 African countries ranging from 1995 to 2014 as follows: SFit = β0 + β1 EFit + β2 LFit + β3 RFit +β4 PRit + β2 CLit + βj Zit +αi + δt + εit (1) Where, index i refers to the unit of observation; t refers to the time period; SF refers state fragility; EF, LF and RF refer ethnic, linguistic and religious fractionalization, respectively; PR refers political right and CL refers civil liberty; Z refers other control variables, αi refers individual specific unobserved factors, δt refers time specific unobserved factors, and εit are individual and time specific residuals. Based on hypothesis 2 and also based on the theoretical justification discussed above, ethnic, linguistic and religious fractionalization may induce state fragility via their interactive effects with political right and civil liberty. Therefore, our final model that takes the interactive factors into consideration can be expressed as follows: SFit = β0 + β1 EFit + β2 LFit + β3 RFit +β4 PRit + β5 CLit + β6 EFit PRit +β7 EFit CLit + β8 LFit PRit +β9 LFit CLit + Β10 RFit PRit + β11 RFit CLit +βj Zit +αi + δt + εit (2) According to Rajan and Zingales (1998), the interest centers are mainly around the coefficient of the interaction terms, i.e, β6, β7, β8, β9, β10, and β11. For instance, the interactive effect between ethnic fractionalization and political right on state fragility corresponds to the β6 slope. If β6 is reliable (or "statistically significant ), it means that the effect of ethnic fractionalization on state fragility depends on the level of the political right. The main variables such as ethnic, linguistic and religious fractionalization are time-invariant variables that have the same effects across time can be estimated by REM not by FEM. In line with this, a White s general test for heteroscedasticity was conducted and the result rejected the null hypothesis of homoscedasticity and this can be easily corrected using the usual heteroscedasticity-robust standard errors. The Relationship between State Fragility and the Independent Variables There are many theories about why states fail, or experience fragility, the majority of which are highly contested. At a very broad level, may be triggered by the following factors: i. Ethnic, linguistic and religious fractionalization: Some studies argue that ethnic, linguistic and religious fractionalization within a nation might have negative effects on societal outcomes since it can create barriers to communication and exchange, factions and rivalries, and internal conflict (Easterly and Levine, 1997; Easterly, 2001). However, Vallings and Moreno-Torres (2005) argue that ethnic, linguistic and religious composition do not in themselves drive fragility. Rather, it is the political manipulation of these factors that can impact on state stability. This manipulation is more likely in states with weak institutions. Thus, we hypothesize the lower the political right and civil liberty, the higher the impacts of ethnic, linguistic and religious fractionalization on state fragility. ii. Political right: Weak institutions are the central driver of state fragility (Vallings and Moreno-Torres, 2005). It is expected that the likelihood of a country to be fragile is low for ccountries with a wide range 59

14 Journal of Marketing and Management, 7 (2), 47-66, Nov of political rights including free and fair elections, competitive political parties, no discrimination among the polity, and the ability of minority groups to participate in the governing process. iii. iv. Civil Liberty: Countries with a wide range of civil liberties, including freedoms of expression, assembly, association and a fair legal system tend to strive for equality of opportunity for everyone, including women and minority groups. This will in turn may reduce the chance of a country to be fragile. Real GDP/Capita: - Economic development is not a prerequisite for preventing fragility, but a lack of growth will mean that institution building is more difficult than otherwise (Vallings and Moreno- Torres, 2005). If the economic development of a country is stagnant or in decline, the country will face absolute poverty and this contributes to multi-dimensional social problems and violence. Thus, a significant and negative relationship is expected between performance in GDP growth and state fragility. v. Natural Resources: This variable is captured by natural resources rent/gdp. Collier and Venables (2009) argue that on one hand, natural resources produce rent and hence a target for rent-seeking that increases the incentives for corruption, theft and insurgency, all of which undermine governance and can lead to state fragility. On the other hand, if properly managed, a resource-rich country has the funds to build state capacity, to invest in education and physical infrastructure for economic development. Thus, the impact of natural resources on state fragility depends on the governance quality of the country. vi. vii. Life Expectancy: - The health dimension of human capital is proxied by life expectancy in a given country. Good health is an input to a healthy workforce necessary for economic growth as well as an intrinsic measure of human development. States that fail to ensure adequate health for their citizens are less likely to grow and the probability for such countries to be caught with negative cycles or traps of perpetual poverty and instability would be very high. This in turn would lead to state fragility in that country. Population growth: According to Goldstone (2002), population growth can give rise to conflicts over increasingly scarce resources, such as farmland and grazing land, especially if those conflicts involve elites seeking to take resources from popular groups, or competition between elite factions for control of those resources. The Data and Stylized Facts This study has made an intensive empirical analysis for a panel of 43 countries from north, south, east, west, and central Africa for the years The year 1995 was chosen as our starting point just because of the availability of full annual data coverage on the State Fragility indicator and other important variables. List of countries included in this study as shown in table 3 below. These countries are selected mainly because of the availability of data throughout the years. Table 3: Countries included in the study Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Central Africa, Chad, Congo Democratic Republic, Cote d'ivoire, Djibouti, Egypt, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Kenya, Liberia, Libya, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, Senegal, Seychelles, Sera Leon, South Africa, Sudan, Tanzania, Togo, Tunisia, Uganda, Zambia, and Zimbabwe. 60

15 Journal of Marketing and Management, 7 (2), 47-66, Nov Moreover, the independent variables, which were discussed in the preceding section and their expected relationships with state fragility are summarized in table 4 below. Table 4: Dependent variable: State Fragility (Source: Global Report 2014: Conflict, Governance and State Fragility, Center for Systematic Peace) Independent Variable Expected Data Sources Relationship (+/-) Ethnic fractionalization + Alesina et al. (2003) Linguistic fractionalization + Alesina et al. (2003) Religious fractionalization + Alesina et al. (2003) Political Right (1=most free; 7= least free) + Freedom House (2016) Civil Liberty (1=high CL; 7= no/few CL) + Freedom House (2016) Ethnic Fractionalization X Political Right + Alesina et al. (2003) & Freedom House (2016) Ethnic Fractionalization X Civil Liberty + Alesina et al. (2003) & Freedom House (2016) Linguistic Fractionalization X Political Right + Alesina et al. (2003) & Freedom House (2016) Linguistic Fractionalization X Civil Liberty + Alesina et al. (2003) & Freedom House (2016) Religious Fractionalization X Political Right + Alesina et al. (2003) & Freedom House (2016) Religious Fractionalization X Civil Liberty + Alesina et al. (2003) & Freedom House (2016) GDP Per Capita (Constant LCU) - WDI database Natural Resources Rent/GDP +/- WDI database Life Expectancy - WDI database Population Growth + WDI database Regression Results, Main Findings and Discussion Two tests have been conducted using REM as shown in Reg-1 (column 1 of table 5) and Reg-2 (column 2 of table 5) that includes the interaction effects of ethnic, linguistic and religious fractionalization with political right and civil liberty on state fragility. Accordingly, the regression results from Reg-1 indicates, keeping other factors constant, ethnic fractionalization can modestly and positively affect state fragility. Contrary to exiting perception, the findings from Reg-1 reveals no evidence has been found for linguistic and religious fractionation to induce state fragility. Instead, the results from Reg-1 indicate the differences in state fragility across countries in Africa can be explained in particular by the level of political right and the level of civil liberty of the countries. This implies that keeping other factors constant, lack of or low level of political right and civil liberty can increase the probability of state fragility. The regression result from Reg-1 also confirm the level of development as captured by the country s income per capita and life expectancy at birth can be significant attributing factors to reduce state fragility. Otherwise, if a country is in a persistent condition of extreme poverty, it is most unlikely for such countries to provide basic services and economic opportunities for its citizens and as a result the likelihood of being fragile state is high. Another important variable that significantly and positively contributing to sate fragility in African 61

16 Journal of Marketing and Management, 7 (2), 47-66, Nov countries include population growth. This implies that if a country experience rapid population growth in slowgrowing economies, it may contribute to conflicts over increasingly scarce resources, such as farmland and grazing land, and eventually to state fragility. On the other hand, in contrary to the existing views by some researchers, the results from Reg-1 found no evidence for natural resources as captured by natural resources rent/gdp to significantly affect state fragility. Thus, the findings from Reg-1 indicate the need to look other factors beyond ethnic, linguistic and religious fractionalization in order to understand what explains the differences in state fragility across countries in Africa. In doing so, Reg-2 of column 2 of table 5 includes the interaction effects of ethnic, linguistic and religious fractionalization with political right and civil liberty on state fragility. Table 5: Dependent Variable: State Fragility Index (SFI) State Fragility Index (SFI) Reg-1 Reg-2 Coef. SE Z P> Z Coef. SE Z P> Z Ethnic Fractionalization * Linguistic Fractionalization Religious Fractionation Political Right ** Civil Liberty *** Real GDP/Capita *** *** Life Expectancy *** *** Population Growth ** *** Natural Resources Rent/GDP Ethnicfr*Lack of Pol-Right *** Ethnicfr*Lack of Civil- Lib *** Linguisfr*Lack of Pol-Right *** Linguistfr*Lack of Civil-Lib *** 62

17 Journal of Marketing and Management, 7 (2), 47-66, Nov Religionfr*Lack of Pol-Right ** Religionfr*Lack of Civil-Lib *** Constant Number of observations Number of groups Obser. per group Wald chi2(9) Prob > chi R-sq: Overall Accordingly, β6 which is the coefficient of the interactive effects of ethnic fractionalization and political right implies that the effect of ethnic fractionalization on state fragility depends on the level of political freedom in that country. Thus, β6 is positive and highly significant implies that the lower the political freedom, the more the impacts of ethnic fractionalization on state fragility. Likewise, β7 which is the coefficient of the interactive effects of ethnic fractionalization and civil liberty implies that the effect of ethnic fractionalization on state fragility also depends on the level of civil liberty in that country Similarly, the evidences from Reg-2 confirm β8 and β9 are also statistically significant at 1% level and this implies that the lower the political freedom and civil liberty, the more the impacts of linguistic fractionalization on state fragility, respectively. By the same token, the statistical results from Reg-2 indicate that β10 which represents the interaction terms of religious fractionalization vs. political right and β11 which represents the interaction terms of religious fractionalization vs. civil right are strongly significant, respectively. More specifically, β10 and β11 are statistically significant implies that the lower the political freedom and civil liberty, the more the impacts of religious fractionalization on state fragility, respectively. Overall, the findings from Reg-2 reveal that ethnic, linguistic and religious composition do not in themselves significantly drive fragility. Rather, it is lack of political right and civil liberty that act as moderating mechanisms and exacerbate the negative effects of ethnic, linguistic and religious fractionalization on state fragility. Last but not least, the statistical results from Reg-2 also confirm the coefficients for the level of development as captured by the country s income per capita and life expectancy at birth are statistically significant so as to reduce state fragility. Keeping other factors constant, the population variable which is captured by population growth has been found to be a positive and significant attributing factor to increase the probability of a country s fragility. This may be justified as in rural African countries population growth can also give rise to conflicts over increasingly scarce resources, such as farmland and grazing land, especially if those conflicts involve completion among elites for control of those resources. Last but not least, the statistical results from Reg-2 found no evidence for natural resources wealth to trigger state fragility. This is contrary to the traditional view that natural resource wealth in African countries leads to conflict over control of those resources. This is due to the fact that not all natural resource rich countries have experienced fragility in Africa and a good example is Botswana. Similarly, not all resource-poor African countries are less state fragile and a good example is Somalia which doesn t have significant natural resources and yet Somalia is at the top of fragile states not only in Africa but also in the world. 63

18 Journal of Marketing and Management, 7 (2), 47-66, Nov Conclusion African ethnic diversity has been blamed for the major source of conflicts and civil wars that are still prevailing in the region. Using a REM, this study attempted to explore the effects of ethnic, linguistic and religious fractionalization on state fragility. It was conventionally believe that ethnic, linguistic and religious fractionalization within a nation might have negative effects on societal outcomes since it can create barriers to communication and exchange, factions and rivalries, and internal conflict (Easterly and Levine, 1997; Easterly, 2001). However, the main finding of this study implies that in countries with low level of political right and civil liberty, ethnic, linguistic and religious fractionalization significantly increase the probability of state fragility. This is equivalent to say that the effect of fragmentation on state fragility declines with higher levels of political rights and civil liberty.this implies that ethnic, linguistic and religious composition do not in themselves drive fragility significantly. Rather, it is lack of political right and civil liberty that act as moderating mechanisms and exacerbate the negative effects of ethnic, linguistic and religious fractionalization on state fragility. This study therefore demonstrates the need to look other factors beyond ethnic, linguistic and religious fractionalization in order to understand what factors exactly explain the differences in state fragility across countries in Africa. Therefore, the main implication of these findings include a crucial need for enhancing strong political institutions and government commitments to enhance political rights including free and fair elections, the ability of minority groups to participate in the governing process (including women and minority groups), a fair legal system that ensures the rule of law (including an independent judiciary) and creating equal economic opportunities for all ethnic, linguistic and religious groups. If properly managed, therefore, contrary to the conventional view, ethnic, linguistic and religious heterogeneity can be considered as asset rather than being a liability and can play a significant role to enhance social cohesion, political stability and national unity. References 1. Adegbija, E Language Attitudes in Sub-Saharan Africa: A Sociolinguistic Overview. Clevedon, Philadelphia and Adelaide: Multilingual Matters. 2. Alesina, A. and G. Tabelleni (1989 ), 'Social fragmentation and economic growth: evidence from developing countries ', Journal of International Economics, 17 : Alesina, A. and A. Drazen (1991). Social Fragmentation and Economic Growth: Evidence from Developing Countries, American Economic Review, 81(5) : Alesina, A. and D. Rodrik (1994 ), 'Social Fragmentation and Economic Growth: Evidence from Developing Countries ', Quarterly Journal of Economics, 109 (2) : Alesina, A. and E. Spoloare (1997 ), 'Social fragmentation and economic growth: evidence from developing countries ', Quarterly Journal of Economics, 112(4) : Alesina, A., A. Devleeschauwer, W. Easterly, S. Kurlat, and R. Wacziarg (2003), Fractionalization, Journal of Economic Growth, 8 (2): Alesina, A. and E. La Ferrara (2005), 'Social fragmentation and economic growth: evidence from developing countries ', Journal of Economic Literature, 43(3): Bailey, G.A. (1994). Rebirth of the Non-Western World. Anthropology Newsletter 35(9):

19 Journal of Marketing and Management, 7 (2), 47-66, Nov Batibo, H.M Language Decline and Death in Africa: Causes, Consequences and Challenges. Clevedon: Multilingual Matters. 10. Bertocchi, G., Guerzoni, A., The Fragile Definition of State Fragility. Mimeo. University of Modena. 11. Cocodia, J. (2008) Exhuming Trends in Ethnic Conflict and Cooperation in Africa: Some Selected States. African Journal on Conflict Resolution, 8 (3): Collier, P. (2000), 'Social fragmentation and economic growth: evidence from developing countries ', Economics and Politics, 3(12): Collier, P The Bottom Billion: Why the Poorest Countries are failing and what can be done about It. Oxford: Oxford University Press. 14. Collier, P. and A. Venables (2009), Natural Resources and State Fragility, Oxford Research Papers, Oxford: Oxford Center for the Analysis of Resource Rich Economies. 15. Easterly, W Can institutions resolve ethnic conflict? Economic Development and Cultural Change, 49 (4) : Easterly, W. and R. Levine (1997). Africa's Growth Tragedy: Policies and Ethnic Divisions. Quarterly Journal of Economics, 112: European Report on Development (ERD) Overcoming Fragility in Africa. San Domenico di Fiesole: Robert Schuman Centre for Advanced Studies, European University Institute. 18. Fearon, J. (2003). Ethnic and Cultural Diversity by Country. Journal of Economic Growth, 8 (2): Freedom House (2016). Freedom in the World, Gettleman, Jeffrey. The Most Dangerous Place in the World. Foreign Policy, February 16, Goldstone, J (2002), Population and Security: How Demographic Change Can Lead to Violent Conflict Journal of International Affairs, 56 (1): Grimes, B.F Ethnologue: Languages of the World, Volumes 1 and 2. 14th Edition. Dallas: SIL International. 23. Gupta, S Rebuilding Fiscal Institutions in Post Conflict Countries. Occasional Paper 247. International Monetary Fund, Washington. 24. Lindemann, S. (2008). Do inclusive elite bargains matter? A research framework for understanding the causes of civil war in Sub-Saharan Africa. Crisis States Research Centre, Discussion Paper No. 15. London: London School of Economics. 25. Maclaren, M. (2007), Social fragmentation and economic growth: evidence from developing countries, Federations, 7(1): Marshall, M., and Cole, B Global Report 2014: Conflict, Governance, and State Fragility, Center for Systematic Peace: Vienna and VA, USA downloaded from Mauro, P. (1995), 'Social fragmentation and economic growth: evidence from developing countries ', Quarterly Journal of Economics, 110(3): OECD-DAC State-building in Situations of Fragility: Initial findings. Paris: OECD-DAC. 29. OECD Resource Flows to Fragile and Conflict Affected States, Annual Report Paris: OECD. 30. Pottier, J Re-imagining Rwanda: Conflict, Survival and Disinformation in the Late Twentieth Century. Cambridge University Press, Cambridge, UK. 31. Rajan, R. and L. Zingales (1998). Financial Dependence and Growth, American Economic Review, 88 (3): Shleifer, A. and R. Vishny (1993), 'Social fragmentation and economic growth: evidence from developing countries ', Quarterly Journal of Economics, 108(3):

20 Journal of Marketing and Management, 7 (2), 47-66, Nov Stewart, F., and Brown G Fragile States', Centre for Research on Inequality. Oxford: Human Security and Ethnicity (CRISE). 34. Vallings, C., and M. Moreno-Torres (2005). Drivers of Fragility: What Makes States Fragile? Trade working Paper No. 7, Department for International Development, UK. 66