CHAPTER ONE INTRODUCTION 1

CHAPTER ONE
INTRODUCTION
1.1Background of the Study
Recently, more attention is been placed on governance in which the relationship between quality of governance and economic growth has been under the lens of various economic scholars so as to determining the association which exist between the two variables. Good governance which is also referred to as country level governance by some economists (Stojanovic et al, 2016), highlighted a new view which marks out the role of state in the economy where the joint participation of state and non-state actors, civil society and private sector, is essential in the process of public governance. From the experiences encountered by India and China, countries around the world have understand that economic growth can best be driven by removing barricades placed on the private firm from the participation in the provision of some economic activities carried out by the government. However, no doubt, the level of the quality of governance is crucial for business development, economic growth and the smooth running of the economy, particularly if we keep in mind the main features that exemplify good governance such as: accountability, transparency in policy making, an effective legislative framework where property rights are clearly defined and the predictability of business transactions is ensured (Bota-Avram et al, 2018).

There have been various views of what should be inclusive in definition and measurement of governance. According to Fayissa and Nysiah (2010), “the working definition of what constitutes good governance has evolved over the years.” It has been established that governance is a multidimensional concept. One common definition focuses on outcomes—the extent to which governments enact and implement policies in the interest of all citizens. Another focuses on political institutions and dynamics that determine these outcomes—the extent to which governments have incentives to adopt and enforce policies in the interests of all citizens. Corruption is the most frequently used indication of weak governance outcomes. The ineffective translation of public resources into public services is another. The vulnerability of property and contract rights to opportunistic behavior by government officials and other private actors is third. The security of property rights and the incentives of politicians to guarantee secure property rights are most closely associated with economic growth (Wilson, 2015).

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Good governance, defined as the quality management and orientation of development policies is assumed by many economists, having positive influence on economic performance. The question is what content the literature gives to the concept of governance? The UNDP (2002) defines good governance as striving for rule of law, transparency, equity, effectiveness /efficiency, accountability, and strategic vision in the exercise of political, economic, and administrative authority. According to the World Bank, good governance is evaluated by the implementation capacity of governance principles of a country, providing a framework for market development and economic growth. Several econometric studies (Kauffman et al. (2005); Knack et al, (2002)) tested the relationship between good governance in the sense of “market-enhancing governance” (stimulus institutions market) and showed a positive relationship between good governance and economic growth. However, does a good governance policy allow developing countries to achieve minimum economic growth and political reforms in order to reach a level of development similar to that of industrialized countries?
Governance in African countries and in particular ECOWAS countries on the average is often considered as poor. According to a news report by the Allafrica (2000), it was reported that the executive secretary of the Economic Community of West African States made the intention of the organization on adopting the strategy to isolate member states with bad records of governance and democracy from some certain vital negotiations known. But yet with this entire move, in these countries, governance indicators have not been improved during the last decade, they still stand under the standard level of other developing countries. Moreover, corruption in these states has been deteriorated affecting transparency of institutions and the effectiveness of the public officers due to increase bureaucracy. In this context, it is expected that improvement in governance may not have the intended impact if governance indicators cannot exceed low levels which means a nonlinear relationship of fiscal policy on growth.
The Economic Community of West African States with the treaty signed and established in 1975 and currently having 15 countries as a member states which is arranged in no particular order than alphabetical order are: Benin, Burkina Faso, Cape Verde, Cote d’Ivoire Gambia, Guinea, Guinea Bissau, Ghana, Liberia, Mali, Niger, Nigeria, Senegal, Sierra lone, Togo. Though haven’t gained much attention in that time, governance in the West African Countries started in the year 1957 when Ghana achieved her independence, followed by Guinea in the year 1958 and lastly Cape Verde getting her independence in 1975. With the achievement of independence coupled with human and natural resources possessed by most of these countries there were high hopes for the region to recover from the developmental neglect and exploitation from the colonials. However, what the region has experienced in the area of growth and development from then till now has been so disappointing. After blowing up the high hopes that surfaced in the 1980s, Most of the ECOWAS countries responded with structural adjustment programmes (SAP) supported by the World Bank and the International Monetary Fund (IMF). The main aim of the programmes was to bring about growth, restoring price stability and reducing external imbalances. However, the SAP strategies and events have not produced the desired results. The outcomes of SAP has been blamed on several issues such as (1) the programmes been introduced somewhat brutally without adequate preparation (2) because of socio-cultural and socio-political reasons (3) both one and two, they have led to the rapid accumulation of debts, leading to heavy indebtedness, and exacerbating the poverty in many of the countries. The yoke of debts—debt burden—has crowded out investments in many instances, and reduced most of these countries to heavily indebted poor countries (Sikod and Teke, 2012). Economic theory of growth and development has predicted convergence. Even though the evidence from the Asian tigers justify the stand of this economic theory, there have been divergence in the West African countries with per capita income stagnating when some developing countries doubled their per capital income (Wilson, 2015). From the beginning of 2000 until the 2014 before the crash in the oil price the real growth rate of the ECOWAS region has consistently been above 5% and these have been mainly driven by the Nigerian economy, this is shown in the table below.

Table 1: Key indicators of the region’s economy
Indicators 2006-2008 2009-2011 2012-2014 2015 2016 2017
Nominal GDP
(billion of US dollars) 367,0 480,5 665,7 637,4 571,4 583,3
Real Growth rate (%) 6.5 7.3 5.6 3.1 -0.2 2.0
Real Growth rate
(excluding Nigeria) 4.6 4.5 6.4 4.7 5.3 6.6
GDP per capita (US dollar) 1331,8 1606,7 2054,4 1866,1 1628,5 1618,2
Population (million) 274,8 298,4 323,7 341,6 350,9 360,5
Source: Annual Report of ECOWAS (2016)
From Table 1, one can see the effect of the global economic recession on the growth rate of the region which has been on record for about decades. The main trigger of the slowdown in the region’s growth is attributed to the fall in oil prices which declined by at least 50% starting from the last quarter of 2014 and Nigeria (the largest oil producer in Africa) that is referred to as the economic powerhouse of the region is mainly dependent on oil. This resulted in the sharp fall in export revenues and into sharp depreciation of the naira which gravitated into decline of the real GDP which falls to -1.7% as to 2.7% achieved in 2015.
There can be growth without development but there can never be development without growth and this make growth a concern of every economy that desires development in which the West African countries are among. Economic growth rates in the West African Countries have been insufficient and this is highly connected with industries and firms producing and trading in highly localized markets and do not achieve the economies of scale to a sufficient level to attract broad-based investment that could accelerate growth. Different constraints have been highlighted to have been involved in putting the economy in an uncompetitive position in the international market such as inefficient transportation and trade barriers along corridors and at borders, a heavy reliance on family and informal sources of financing, and an insufficient supply of reliable and affordable power or energy which good governance can help ameliorate. Linkages between country level governance and the business environment development that can bring about growth have been identified by different scholars (Bota-Avram, et al, 2018). Cule and Futon (2013), states that the impact of good governance on economic growth is backed up by the idea that one believes that an economy with a high regulatory control, stable and consistent political environment and a reduced level of bureaucracy will be able to provide an effective business framework which will trigger growth. Moreover, a previous study also shows that the quality of governance also reflects in accountability and transparency of the bureaucrats (Demirguc-Kunt et al, 2006). All these key elements enhance the performance of the economy both at the micro and macro level.

In the reverse manner, some authors do not believe that governance have a significant impact on the growth of the economy (Sach et al 2004). However, it has been established that an improved business climate is a major factor in attracting both national and international investors into the economy which will increase economic growth. Growth that is inclusive will occur in a business democratic environment when the government is socially accountable in the delivering of her services and responsive to the needs of its citizens. The main objective of this study examines the impact of the dimensions of governance on the economic growth of the countries in the ECOWAS region and also explores the causal relationship between the two variables.

1.2Statement of the Problem
In 2016, growth in the ECOWAS countries in the period of widespread economic downturn with Nigeria and Liberia recording a negative growth rate, though some of the countries in the region such as Cote d’Ivoire recorded a high growth amidst it. It must be noted that the economy in this region depends on few of the countries such as Nigeria accounting for over 70 percent of the GDP in the region, followed by Ghana and Cote d’Ivoire. However, with rising prices of commodities in the international market coupled with the facts that most of the countries in the West African region solely rely and depend on exportation of natural resources raise the expectations of meeting the growth projection, though this leave them vulnerable to external shock. Beyond these, there is a role of the government in coordination and representing the opinion of her people which gives governance a high attention.
There has been contradicting views on the relationship between governance and growth. Improvement of the quality of governance on the growth of the economy has been stressed by various scholars, and has observed positive association between governance and growth (Wilson, 2015). Several scholars have explored the causal relationship between country-level governance on growth but few studies have examined the possible reverse causality. Also, the studies have only been carried out for some selected countries in Sub-Saharan Africa which did not cover all countries in ECOWAS. Also Emara and Chiu (2015) carried out the studies on Middle Eastern and North African countries. This study covers the gap by conducting the study of relationship between the quality of governance and growth in the ECOWAS region.
African has been suffering from governance crises with several international organizations giving her a less than fair rating on the average in the quality of governance with visible examples of profound governance issues as we have in war and violence happening in Liberia and cases of corruption as recorded in Nigeria. None of the countries under the Economic Community of West African States have 100% standard governance and it is not found in any economy of the world. Property rights and contracts are never and nowhere 100% secure (Dixit, 2015). The indexes used in measuring and rating the governance of most developing countries have been called into questions as to whether they truly depicts what is actually happening because for instance in the economy of the ECOWAS countries, there are countries such as Nigeria and Cote d’Ivoire having low governance rating yet they have a relatively high economic growth in the region. Governance can be divided into three dimensions which are economical, institutional and political (Haq and Zia, 2009). To add to the body of knowledge this study examines the impact of each of the dimension of governance and growth and also examine if there be any causality.

1.3Objectives of the Study
The broad objective of this study is to examine the impact of governance on the economic growth of the countries under the region of Economic Community of West African State. The specific objectives of the study therefore include:
To investigate the relationship between governance and growth
To examine the impact of governance on economic growth
To examine the effect of economic growth on governance
1.4Research Questions
What is the relationship between governance and economic growth?
Does high quality of governance bring about growth?
Does high economic growth enhance improvement in the quality of governance?
1.5Research Hypotheses
Hypothesis I
H0: There is no significant relationship between the three dimension of governance and growth
H1: There is significant relationship between the three dimension of governance and growth
Hypothesis II
H0: High quality of governance does not significantly affect growth of countries in ECOWAS region
H1: High quality of governance significantly affects growth of countries in ECOWAS region
Hypothesis III
H0: High level of economic growth does not enhances the improvement in the quality of governance in the countries of the ECOWAS region
H1: High level of economic growth enhances the improvement in the quality of governance in the countries of the ECOWAS region
1.6Justification of the Study
Growth in the economy, which is mostly achieved by inducing investment both internally and externally need stability of policy, violent free environment well defined contract laws and some other key elements which can only be provided by good governance. The issue of the economic dimension of governance and transparency has become a centre of attraction drawing the attention of various governments in the region thereby including as a core components in the programmes they embarked on in their country. “Thus the promotion of good governance is at the centre of many of the reforms being undertaken by the government, particularly in the areas of public expenditure management, the liberalization of markets, and the introduction of transparency and accountability into public life” (Sikod and Teke, 2012), such as declaration of asset when assuming public post.
However in West Africa region, government of countries such as Liberia, Sierra Leone and Guinea Bissau have had a hard time driving long-time peace to their economy which is because of institutional corruption, greed and ethnic discrimination with skewed notions in the country (Musah, 2009). This study provide some in-depth insight into the dimensions of governance issues and economic growth in the 15 countries of the Economic Community of West African State, and will also contribute to the on-going debate on governance and economic growth by increasing the level of consciousness about the implication of governance.
1.7Scope of the Study
This study focuses on examining the impact of the dimension of governance on the growth and also the reverse case on the countries in the Economic Community of West African State region between 1996 and 2016. The choice of the region is based on the motive of having a robust findings and the time frame chosen is based on the availability of governance data.
1.8Organization of the Study
The chapters included in this paper are: Chapter one of the study, which focuses on the background of the study, statement of the research problems, objectives of the study, justification of the study, statement of research question, statement of hypotheses, scope of the study and organization of the study. Chapter two deals with literature review which reviews the relevant existing literature on the research topic. Chapter three consists of the theoretical framework and research methodology used for the research study. Chapter four presents the results obtained from the estimation and also the interpretation of the result. Chapter five comprises of the summary, conclusion, and policy recommendation.

CHAPTER TWO
LITERATURE REVIEW
2.1Conceptual and Theoretical Review
This chapter presents a review of related literature on the dimensions of governance and growth in the ECOWAS member states, and extended globally.

2.1.1Conceptual Review
Beyond governance which is used to capture economic growth in this study, there are other factors that affect the economic growth of a country or region, this study will dwell on governance and the description of the chosen variables is given. This explanation will help us to choose the appropriate methodology to examine the relationship between governance and economic growth. Therefore, the focus of this section is to establish a theoretical and conceptual framework for the quantitative research.

2.1.1.1Governance
In recent time in the study of economics, more attention is now been given to the institutional arrangement in the quest of growth and development. Specialists in the Medieval English Society were the first to use the concept of governance as referred to in the growth and development economics literature. They characterized governance as cooperation between the different sources of powers. The term governance today is widely used in different international development literature and there are numerous interpretations given to describe the concepts (Haq and Zia, 2009). Governance means process of decision making and the process by which decisions are implemented. The quality of governance is determined by the impact of this exercise of power on the quality of life enjoyed by the citizens. Governance can be used as several contexts such as international governance, corporate governance, national governance and local governance. Government is one of the actors in governance. Although no consensus exists in terms of defining governance, a common idea among scholars is that governance means more participation in the political and decision-making process by non-governmental institutions (Agere, 2000; de Ferranti et al 2009; Lovan et al 2004). Thus, under governance, government is one of several players—rather than the only player—managing a nation’s affairs (Frahm and Martin, 2009; Kettl, 2002; Lovan et al., 2004).
Asian Development Bank (1995) identified four basic elements of good governance such as accountability, participation, predictability and transparency. McCawley (2005) categorizes governance issues at the macro and micro level. The macro level includes constitution, the overall rule of size and resources of the government, and relationship between legislators, the judiciary and the military, while micro issues of governance includes commercial firms, social institutions and civil society affairs. According to de Ferranti et al. (2009), “governance describes the overall manner in which public officials and institutions acquire and exercise their authority to shape public policy and provide public goods and services”. Governance “represents the overall quality of relationship between citizens and government, which includes responsiveness, efficiency, honesty, and quality”. Similarly, the United Nations defined governance as “the process of decision-making and the process by which decisions are implemented (or not implemented)” (UNESCAP, 2009, p. 1). United Nation Development Programme (1997) defines governance as the exercise of economic, political and administrative authority to manage a country’s affairs at all levels. It comprises mechanisms, processes and institutions through which citizens and groups articulate their interests, exercise their legal rights, meet their obligations and mediate their differences. The United Nations also introduced characteristics of good governance practices as a global standard to be adopted by governments that receive their aid. According to the United Nations, “good governance has 8 major characteristics; it is participatory, consensus oriented, accountable, transparent, responsive, effective and efficient, equitable and inclusive, and follows the rule of law” (UNESCAP, 2009). International Country Risk Guide (ICRG) covering 140 countries from 1980 to the present analyses and forecast risk for international investors. It includes 22 components that are grouped into three categories of risk: political, financial and economic. The political risk assessments are made on the basis of subjective analysis of the available information, while the financial and economic risk assessments are made solely on the basis of objective data. In determining the component rating, political risk contributes 50 percent to the rating while the other two categories contribute 25 percent each.

The current use of the governance concept may be traced to a World Bank study in the year 1989 on Africa that defined governance as “the exercise of political power to manage a nation’s affairs”. Later, the World Bank (1992) defined governance as “the manner in which power is exercised in the management of a country’s economic and social resources for development”. The Organization for Economic Co-operation and Development (OECD), on the other hand, defined governance as “the exercise of authority in government and the political arena”. In line with this definition, “Good public governance helps to strengthen democracy and human rights, promote economic prosperity and social cohesion, reduce poverty, enhance environmental protection and the sustainable use of natural resources, and deepen confidence in government and public administration” (Tarschys 2001, 28).
World Bank aggregate governance indicators data set developed by Kaufmann, et al. (2005) hereafter called the KK Data sets, is a set of worldwide measures of six composite dimensions of governance perception indicators for 105 countries. These indicators are oriented so that higher value correspond to better outcomes, on a scale refers to the point estimates range from –2.5 to 2.5. These estimates are also rescaled and ranked in percentile (0–100). The lower percentile is ranked as worse off governance indicators whereas upper percentile is ranked as best governance for any given country. These perceptions may often be more meaningful than objective data, especially when it measures public faith in institutions. These averages of governance indicators are considered to capture institutional quality. These dimensions can be classified into three clusters with two indicators in each group is given as:
Governance from Economic Dimension
Under economic governance the two indicators representing this indicator are Government effectiveness and Regulatory quality. These two indicators summarize various indicators that include the government’s effectiveness which shows the quality of public services, the quality of 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. The thrust of this index is on the input required government to be able to produce or implement good policies and quality delivery of public good. The ‘regulatory quality’ governance indicator covers the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development and this includes measures of the incidence of market unfriendly policies such as price control or inadequate bank supervision, as well as the perceptions of the burdens imposed by excessive regulation in area such as foreign trade and business development. The figure 2.1 shows the state of deviation of growth represented by LGDP and economic dimension of governance of the 15 countries of ECOWAS.

Figure 2.1Economic dimension of governance and growth

Source: Author using WGI data from World Bank
Governance from Political Dimension
Political governance covers these two; voice and accountability and political instability and violence. The political governance indicator is intended to capture the process by which government is selected, monitored and replaced. First indicator which is voice and accountability measures political, civil and human rights and independence of the media. It includes a number of indicators measuring diverse aspects of political process, political rights and civil liberties. It expresses to which extent a country’s citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association and free media while the ‘political instability’ indicator captures the likelihood that the government in power will be destabilized or overthrown by possibly unconstitutional or violent means, including military cop, terrorism etc. The figure 2.2 shows the state of deviation of growth represented by LGDP and political dimension of governance of the 15 countries of ECOWAS.

Figure 2.2Political dimension of governance and growth

Source: Author using WGI data from World Bank
Institutional Dimension of Governance
Institutional dimension of governance is captured by Rule of law and Control of corruption. The final dimensions of governance indicators are summarized in broad terms as the respect of citizen and the state of institutions that govern their interactions. Rule of law, summarizes several indicators that measure the extent to which agents have confidence in and abide by the rules of the society. It measures the quality of contract enforcement, the police and the courts as well as likelihood of crime and violence. These indicators also measure a society’s success in developing environment in which fair and predictable rules form basis for economic and social interactions. Control of corruption measures its perceptions conventionally defined as the exercise of public power for private gains. This aspect of corruption differs somewhat, ranging from the occurrence of additional payment to get things done, to assess the effect of corruption on business environment, to measure grand corruption in political arena or in the tendency of elites to engage in state capture. The presence of corruption is often a manifestation of a lack of respect on the part of both the corruptor and the corrupted for the rules that govern their interaction, thus represents a failure of governance (Haq and Zia, 2009). The figure 2.3 shows the state of deviation of growth represented by LGDP and political dimension of governance of the 15 countries of ECOWAS.

Figure 2.3Institutional dimension of governance and growth

Source: Author using WGI data from World Bank
2.1.1.2Economic Growth
Over the years, there has been no universally agreed- upon definition of economic development, however, there is a unionism in the view of development researchers that economic development brings about improved living standards for people and is necessary for a strong long-term national economy. Economic development implies both the improvement of people’s health, education, and general well-being and the presence of positive economic indicators, such as economic growth and low unemployment rates (Adams &Mengistu, 2008). Sustainable development is another issue related to economic development because, without strong long-term economic growth, an economy will be in danger of collapsing during any economic or political crisis (Blair and Carroll, 2008; Mayer-Foulkes, 2009; Nafziger, 2006). Economic development is important because it has implications on people’s lives (Adams and Mengistu, 2008; Chong and Calderon, 2000). With economic development, people will have better education and healthcare and be more productive. Economic development also affects crime rates and political stability as better-developed nations tend to have lower crime rates and greater political stability than less-developed countries (Kaufmann and Kraay, 2002; Abdellatif, 2003; Adams and Mengistu, 2008). Consequently, economic growth concerns all nations trying to increase their GDP per capita in order to increase their citizens’ well-being (Adams and Mengistu, 2008; Mankiw, 2009; UNDP, 2010). Although scholars continue to debate whether it is a consequence of human development or a precondition for human development, economic growth is considered an important component of economic and human development. Smith (2007) found that human development and economic development need each other; as such, countries cannot concentrate on one and ignore the other. According to Smith (2007), “there is in effect a virtuous circle of human development and economic development, each enhancing the other”. In addition, the United Nations Development Program (UNDP) demonstrated that economic growth, education, and health are the key parts of human development, with each part dependent on the others. According to the UNDP (2000), “resources generated by economic growth have financed human development and created employment while human development has contributed to economic growth”.
Economic growth is the increase of real gross domestic product (GDP) or other measurements of aggregate income (AlBassam, 2013). According to the World Bank (2004), economic growth is “quantitative change or expansion in a country’s economy”. In addition, the World Bank (2004) contended that “economic growth is conventionally measured as the percentage increase in gross domestic product (GDP) or gross national product (GNP) during one year”. After acknowledging the existence of the relationship between economic growth and human development, the Human Development Reports (HDRs, 2010) indicated that the direction of the relationship is not clear cut. According to the HDRs (2010), “even if there is a causal relation, the direction is unknown: higher incomes could improve quality of life, or improvements in health and education could make societies more productive”. In addition, both the income distribution among citizens and the quality of goods and services produced are as important for any nation as increasing income levels. According to Petrovskiy (2009), “from a human development perspective, the quality of economic growth is just as important as its quantity”.
Furthermore, economic growth has been linked to governance improvement (Albassam, 2012; Furubotn and Richter, 2005; Kaufman and Kraay, 2008; Mantzavinos, 2001). Kaufmann and Kraay (2002) argued that governance quality and economic growth are positively related. In their evaluation of the worldwide governance indicators (WGI) from 1996 to 2002, they found that “per capita incomes and the quality of governance are strongly positively correlated across countries”. Accordingly, the relationship between economic growth and quality of governance impacts international aid assistance from countries such as the U.S. and the U.K. and from international organizations such as the World Bank and the IMF. According to Mehanna et al (2010), “the issue of causality between governance and economic development is crucial and has many implications from an international agency perspective; resolving this issue would assist international organizations in their choices between prioritizing pro-growth or institutional policies”. Therefore, the power and direction of the relationship between economic growth and governance have been and will continue to be the subject of disagreement among policymakers and those in academia (Acemoglu et al., 2001; Alkire, 2010).
2.1.1.3Governance and Economic Growth
Different literature has linked several theories especially the institutionalist’s theories proposed by the School of Compatibility among others, to the study of the relationship of governance and economic growth (Porras and Vázquez, 2015). These theories argue that development depends on the ability to secure property rights and enforce contracts in the economies. Governance, and eventually democracy, promotes growth because they ensure property rights, business transactions, social rights and the provision of public goods. Studies of the relationship between governance and growth cannot be separated from the debate about the role of government in the economy. Particularly, Bevir (2007) and Aguilar (2010) argue that the concept of governance reflects the transformation in the vision of the State that occurred after the reforms of the public sector of the years 1980 and 1990. As it is known, these reforms reduced the importance of bureaucracy and introduced market criteria in decision-making. For this reason, studies on governance have normative connotations. Governance has changed the way of formulating and implementing public policies; these are no longer considered as unilateral processes. Public decisions, from the perspective of governance, involve coordination, discussions, understandings, negotiations, agreements and public, private and social commitments. In this context, Bevir (2011) highlights the importance of markets, social networks and non-state actors for public policy.
In line with the above view, Bevir emphasizes that governance defines rules that go beyond those defined by the formal powers of the State. Traditionally it is assumed that an improvement in governance and institutions can promote economic growth (Acemoglu, Johnson and Robinson, 2005; Acemoglu and Robinson, 2006 and 2012). So, some argue that consolidated governance and democratic institutions can enhance the growth of developing economies. However, it should be recognized that this regulatory requirement has been questioned by Glaeser et. al. (2004) and by advocates of the “School of Conflict” and “skeptical school”. Therefore, despite popular belief, there are still theoretical debates on the relationship between governance and economic growth.

2.1.2Theoretical Review
2.1.2.1Growth-Governance Hypothesis
The Gross Domestic Product (GDP) has been the most frequently used measure of economic growth in the academic literature and most studies have revealed that there exist a strong positive association between high-quality governance institution and good economic performance (Wilson, 2015). Change in the quality of a country’s quality of governance can be brought about by improvements in the economic performance through different mechanisms. The first mechanisms are the relative payoff to investments in the formal governance, rather than resting on informal mechanisms, which will in turn increase with a country’s level of economic activity. Personal ties and repeated interactions is the drive of economic exchange and activities when it is localized and grooming stage (Dixit, 2004). As the volume of trade and economic activities grows and the economic develops, it may enhance the relative efficiency of formal governance mechanisms, creating stronger incentives for public investments in improved governance institutions (Li, 2003).

Furthermore on growth to governance hypothesis, the cost and technical requirement of quality governance reforms are lacking in many countries (Rodrik, 2007). Therefore, the advocates of a program of good enough governance argue that growth can often be sparked by relatively minor reforms that encourage investment and that such growth can allow developing countries the time and resources to establish higher quality governance institutions at a later stage of economic development (Grindle, 2007). There are some studies where the linkage between economic growth and good governance is contradicted. For instance, some authors (Stojanovic et al, 2016) argue this conclusion with the examples of certain countries such as Cambodia, China and Vietnam in which the economic growth is clearly demonstrated despite of the lack of good governance. There are few studies that investigated the governance-to-economic performance causal relationship and contradict the conclusions resulted from most of the previous papers which argue the positive influence of good governance on economic development. The author of this study confirms a positive correlation between governance and economic growth, but he concludes that is not necessary to consider the quality of governance as a key factor, determinant for economic performance. Furthermore, they demonstrated a positive influence of economic growth on subsequent quality of governance; therefore he suggests that more attention should be paid to analyze the possible reverse causality between governance and economic growth, particularly in cross-country analyses (Wilson, 2015).

Another channel which have been identified that growth may cause governance is through creating a constituency of business and customers with the interest and ability to demand such improvements. In many developing countries, especially those experiencing transition from a centrally planned economy to a market oriented economy, state-owned firms are the one occupying the central stage in the economy despite being inefficient (Li and Xia, 2008). In other to still maintain their position, these firms have strong incentive to resist reforms that would tighten their soft budget constraints and expose them to competition from new, more efficient private enterprises. In such a setting, growth in the non-state sector – largely supported by informal, network-based forms of governance in place of missing or ineffective formal institutions – may be required to give private economic interests the economic and political power to effectively advocate the governance reform. A substantial number of historical case studies support the hypothesis of causality from economic growth to improvements in governance. For instance, Chang (2003) investigated the historical development of several aspects of governance in the now developed countries of Western Europe and North America. Many of these features of good governance – including a professional bureaucracy, effective corporate regulation, an impartial and independent judiciary, consistent and impersonal enforcement of contracts and protection of property rights, efficient broad-based tax collection, and modern social welfare institutions were shown to have been implemented in the most advanced countries only in the late 19th century (or in many cases well into the 20th century), by which time these countries had already enjoyed half a century or more of industrialization and sustained economic growth. Goldsmith (2007) acknowledges the same trend in his finding for Jamaica, Mauritius, Argentina and United States. His finding suggests that quality of governance at least in the modern sense, was not required for economic growth to take off at the early stages, pointing to the fact that improvement in governance may have been as a result of economic growth.
2.1.2.2Governance to Growth Hypothesis
Causality may operate between governance and growth through several potential ways. The first is through the professionalization of the bureaucracy and this goes a long way to give the bureaucrats with the assurance of merit-based career and predictable paths in the civil service which bring about a definite stable stay in office which enhances investment in public infrastructure with long-term payoffs rather than short-term which is in form of consumption (Wilson, 2015). A bureaucracy that one can bank upon is capable of motivating private businesses to engage in long-term investment since the perceived risk that comes along change in power or government policies will be highly reduced.
The correlation between the quality of governance and economic growth has been the objective of some cross-country empirical studies (Wilson, 2015; Sharma, 2001) while only few paper are addressing the impact of bureaucratic professionalization and effectiveness of government on economic growth (Cingolani et al, 2015). Economists and policymakers have common ground about the role of institutions in influencing economic development, while several authors in their individual research finds that “There is increasing recognition that corruption and other aspects of poor governance and weak institutions have substantial, adverse effects on economic growth” (Sharma, 2007). In the attempt to answer the question of how important is good governance for economic growth, another author admits the answer is best highlighted by the “oft-cited aphorism that good governance promotes growth and that growth further improves governance”. Sharma (2007) also admits there are a lot of econometric papers whose findings emphasize a strong correlation between long-term economic development and good governance, thus the quality of governance definitely influences long-run economic performance outcomes. Also, the bureaucratic consistency accomplished through organized rule-based decision-making should also increase the success of major infrastructure projects that involve partnership between different levels of governments. Bureaucratic professionalization is of many advantages such as encouraging investment that are productive and also reduces the chances of corrupt practices. The strong positive correlation across countries between GDP per capita and the quality of governance was also admitted by the World Bank (Kaufmann and Kraay, 2002). Even more, researchers from the World Bank proposed in their working papers an empirical strategy to analyze this correlation from two directions, namely: (a) a strong positive causal effect running from a better quality of governance to GDP per capita; and (b) a weak and even negative causal effect running in the opposite direction from GDP per capita to the quality of governance. Their results provide sufficient evidence on the significance of good governance for economic performance and the same study also points out the non-existence of “virtuous circles in which a higher GDP per capita determines further amelioration in the quality of governance (Bota-Avram et al, 2018).

Governance to growth relationship can also be seen from the standpoint of institutional and policy perspective, economic growth can thrive well in the society when laws and regulations are enforced effectively with an impartial judicial system. In examining China (Zhoung, 2001), role played by educational attainment in selecting and assessing public officials have been on the increase which is as a result of the significant legislative and bureaucratic changes taking place over the past few decades thereby improving China’s quality of governance. In line with the movement towards the global benchmark, room is given to independent regulatory agencies to oversee key industries yet they are still subject to political interference (Wilson, 2015).
2.1.2.3Endogenous Growth Theory
Convergence was predicted by the Neo-Classical using Solow’s view of 1956, and the view is that economies will move toward their steady state growth which implies that in the long run, there will be convergence in their per capita income but to the lack of empirical backup for convergence has been a challenge to their models. Endogenous growth theory came up with some conditional factors that can bring about convergence and these factors are related to institutions. Knowledge spillovers as it is referred to by the new growth theories is the based on the assumption that any sectors in the developing countries can catch-up with the latest trend of technology whenever in innovates and the adaptation of this new innovation rest on the institutional arrangement which governance is key part. Technological changes when seen as endogenous its innovation can be driven by the governance based on the incentive and transaction cost which goes a long way to determine how fast technological changes will actually progress (Siddiqui and Ahmed, 2009).
The quality of governance brings about growth by minimizing the risk of doing business and this in turn brings about channeling of resources to innovation by prevent diversion of resources and preventing predatory rent seeing activities. When governance can give an enabling environment which is free of diversion, the productive units are assured of being rewarded fully to the amount of what they have inputted into the production and this reduces diversion of resources by individual and increases investment in knowledge.

2.2Empirical Review
Growth has been seen as the foundation for development and the role of good governance has been attributed a major role in achieving a sustainable growth. Different academic literature has attempted to explore the relationship between governance and growth. In the empirical literature there are several studies that have examined the relationship between governance and economic growth. Among these studies are found the Knack and Keefer (1995), Mauro (1995) and Alesina (1997). These studies were pioneers in evaluating the mentioned relationships focusing on the role of institutions and property rights. They pioneered the use of cross-sectional indicators for statistical validation. Furthermore, they were the first to consider governance as a multidimensional phenomenon and that its analysis could be separated from the political regimes. Contemporary studies on the relationship of governance and economic growth emphasize the role of property rights and the quality of institutions. These studies tend to find positive relationships between the development of institutions and growth. Among these studies are those of Knack and Keefer (1995) and Ndulu and O’Connell (1999), who evaluate how risk and violence are linked to economic growth. Rivera-Batiz (2002), meanwhile, analyzes the relationships between democracy, good governance and economic growth. The mentioned studies are important because they tend to support the institutionalist theories (see North, 1990 and 2005; Olson, 1996). Furthermore, they are important because they complement other studies on economic development. In particular, they complement studies that emphasize the role of political rights and civil liberties (Grier and Tullock, 1989), democracy and social conflicts (Keefer and Knack, 2002) and the importance of social capital (Gutierrez-Banegas and Ruiz-Porras, 2014).
Alesina et al (1996) studied the effects of political instability on per capita GDP growth of a sample of 113 countries over the period 1950-1982. The major result of their paper is that political instability has negative effects on economic growth. That is, political instability lessens the growth. Moreover, their results suggest that regime changes affect growth adversely. The same findings were reported by Feng’s (1997) study. Using the three stages least-squares estimation for data covering 96 countries covering the period of 1960-80, the author’s findings demonstrated that political instability has significant and negative effects on economic growth.

Kaufmann, Kraay, and Zoido-Lobatón (1999) studied more than 150 countries, provides empirical evidence that good governance matters a great deal for economic outcomes. Kaufmann and Kraay (2002) conducted another study of 175 countries for the period 2000/01, asserting that good governance is necessary for high per capita income and economic development. The same result was concluded by Knack (2002). It is worth mentioning that Kaufmann (with other authors) has examined the impact of governance on economic outcomes in many studies. In each one of them, he comes to the same conclusion stated above.

Moreover, the results of a study by Calderoân and Chong (2000) have confirmed that there is strong causality from institutional quality to economic growth. The authors’ results have also shown that economic growth causes institutional quality. Although their findings indicate that policies that attempt to improve the state’s institutional quality by securing propriety rights, controlling corruption, and limiting uncertainty need considerable time to achieve the desired goal, these policies are important for economic growth. In addition, such a study has shown that institutional reforms have high influence on economic growth, especially for the very poor countries. Furthermore, by answering the question: why do some countries produce so much more output per worker than others? The results of Hall and Jones (1998) have revealed that a country’s long-run productivity, capital accumulation, and thereby productivity per worker are influenced the most by institutions and government policies.

Using the PRASH Model, Campos and Nugent (2000) analyzed the relationship between the growth volatility and political stability of Argentina over the period of 1896-2000. The authors’ findings suggest that “informal” political instability, such as assassinations, directly and negatively affects economic growth; and “formal” political instability affects economic growth indirectly.

In a cross-sectional analysis of all developing countries, Chauvet and Collier (2004) found that those countries suffering from poor governance, on average, experience 2.3 percentage points less GDP growth per year relative to other developing countries. There are also other recent findings that suggest a strong causal effect running from better governance to better development outcomes. In spite of such a broad array of support for the positive impact of good governance on economic growth, there are only few studies that show results to the contrary. For example, an important challenge to the significance of good governance for the economic growth of African countries comes from Sachs et al. (2004). In an empirical analysis, they show that the differences in performance among African countries cannot be explained by differences in the quality of their governance once differences in their levels of development have been accounted for and thus conclude that a focus on governance reforms is misguided.

Acemoglu et al (2005) examined the link between institutions and long-run growth. They argue that when political power is allocated to groups that enforce propriety rights, when there are few rents that can be sought by the groups in power, and when there are effective constraints on power-holders, there will surely be a causality from economic institutions to economic growth.

Drury et al (2006) studied the connection between corruption, democracy, and growth in more than 100 countries for the period 1982-97. The authors’ findings show that corruption does not have any significant impact on growth in democracies, whereas corruption has strong negative effects on growth in non-democratic countries.

Dam (2006) reviewed the relationship between the rule of law and the economic growth of China. The author argues that China is currently facing the same type of governance issues that Asian Tigers have experienced. Asian Tigers’ economic growth has been negatively affected by such governance issues. The author contends that these issues may affect China’s economic growth as they have affected Asian Tigers’ growth. Dam avers that China’s governance weaknesses are associated with many problems, such as a weak judiciary. Additionally, the author concludes that there is nothing in China’s experience from which one can conclude that institutions and rule of law are not important for economic development.

Morita and Zaelke (2007) have studied the link between the rule of law, good governance, and economic development. The authors argue that rule of law and good governance are important to achieve sustainable development. They also emphasize that good governance and sustainable development goals will not be achieved just by making laws and regulations, rather by enforcing those regulations and laws by governments.

In sharing his ideas on governance with World Bank economists, Rodrik (2008) said that governance is an important tool for development. He suggests that it is a good instrument to achieve better economic outcomes and enhance a country’s policy making. Rodrik also distinguishes between governance as a means and as an end. The author advises economists not to try to address governance as an end because it is the political scientists’ task. For governance as a means, however, he argues that only countries having governance as binding constraint can give a governance reform the priority to boost their economic growth.

Amirkhalkhali and Dar’s study investigates the connection between regulatory quality and economic growth of the 23 OECD countries over the period 1996-2008. They use a generalized version of the production function model of Solow. Their findings suggest that regulatory quality and economic growth are positively correlated. That is, a better regulatory quality leads to a high growth rate. The authors argue that regulatory quality has an impact on economic growth through its effects on total factor productivity.

Guisan (2009) examined the link between government effectiveness, education and economic development by comparing European countries to the U.S. and Canada over the period of 2000-07. The author’s results have shown the importance of government effectiveness to economic development.

Many studies have been done to determine the relationship between corruption and growth at the macro-level. One such study has been conducted by Hodge et al, (2009) where the authors used an econometric methodology that can take into account the multidimensional nature of, as well as the inherent endogeneity in, the relationship corruption-growth. Overall, their results have shown that corruption has negative impacts on investment in human capital, physical capital, and political stability, which means that corruption indirectly impedes growth.

Huynh and Jacho-Chavez (2009) have used a nonparametric method to analyze the relationship between governance and growth. Their findings indicate that three of the six indicators of governance: voice and accountability, political stability, and rule of law are economically and statistically significant, while regulatory control, control of corruption, and government effectiveness are insignificant. The authors state that their empirical results support the findings of Glaeser, La Porta, de Silva, and Shleifer (2004) that poor countries get out of poverty and grow through good policies pursued by a dictator.

Using the studies by Knack, Stephen, Baliamoune-Lutz and Stefan Lutz, and alongside the study of Transparency International in Cameroon, Sikod and Teke (2012) established that there is a direct relationship between governance and economic performance, and that Cameroon has lagged behind in development in a major part because of bad governance which led them to give a policy implication that as governance indicators improve, the economic performance will also improve.
Another study was conducted by Cebula and Foley (2011) to test three hypotheses, one of which is about how quality government regulation affects per capita real GDP. By using panel data and PLS estimation for OECD countries over the period of 2003-06, the authors conclude that better regulatory quality is positively associated with economic growth because it has a positive effect on the way market functions, and it allows for the avoidance of unnecessary costs of managing businesses in the marketplace.

Ahmad et al (2012) used panel data over the period of 1984-2009 for 71 developed and developing countries to test whether corruption affects growth. Their study demonstrates that the relationship between corruption and long-run economic growth is hump-shaped. Their results also suggest that the quality of public institution has a crucial impact on any country’s growth performance. They conclude that there are many ways though which corruption can lessen economic growth, such as lowering domestic and foreign direct investment, and overblown government expenditure.

Another study was done by Aisen and Veiga (2013) to determine the impact of political instability on the growth. The authors used the system-GMM estimator for linear dynamic panel data models on a sample covering 169 countries for the period of 1960-2004. Their results have proved that political instability and lower GDP per capita are strongly associated. Political instability has negative effects on economic growth by reducing the rates of productivity growth, and lowering capital and human accumulation.

The study by Han et al (2014) determines whether countries with below-average governance grow slower than countries with above-average governance. Their results show that government effectiveness, political stability, control of corruption, and regulatory quality are more significantly positively correlated with economic growth than rule of law and voice and accountability. The results also indicate that the studied Asian countries’ above-average governance grow faster than those with below-average governance.

Emara and Jhonsa (2014) used the Two-Stage Least Square method for a cross-sectional dataset of 197 countries to investigate the interrelationship between the improvement in the quality of governance and the increase in per capita income. Their findings show that there is a strongly positive and statistically significant causation from the quality of governance to per capita income. The results also prove a positive causation in the opposite direction. The authors used their results to interpret the relationship between the studied variables for 22 MENA countries. They contend that one of their surprising results is that even though most of the studied MENA countries had low performance on all six indicators of governance, these MENA countries’ income per capita is relatively higher than the rest of the countries in the sample.

Wilson (2015) tested the casual relationship between quality of governance and economic growth in China at the provincial level and found out that under some certain circumstances, successful economic growth can be achieved without reliance on the improvements in formal governance institutions and that such economic growth can in turn support subsequent governance improvements.
2.3Implications of the Review for the Current Study
A major question that has been raised repeatedly with respect to the applicability of industrial policies that were so successful in East Asia to Africa economies is “governance.” Therefore, from the above literatures one can conclude that the effects of governance on economic growth might be positive or neutral. This study draws from the case study of Wilson (2015) which examined the bi-causal relationship between governance and growth in China and from a similar work conducted by Bota-Avram (2018) in studying some countries from the Sub-Saharan African countries. Both authors acknowledge the importance of the double feedback relationship between institutions and economic performance in their informal analyses. The motivation for this paper is initiated by the gap of applying this study to the region of West African countries in line with the important policy questions from the recent empirical evidence that was summarized above in which question such as: Does each of the dimensions of governance improve the economic growth of a country? Can appropriate policies make a contribution of governance more efficient? Why would some countries benefit more from a governance reform than other countries? These are challenging questions to answer because there are many interrelated factors that affect the long-term economic growth of a country. In spite of that, one empirical fact is that nations can improve economic growth by adopting appropriate policies. Therefore the goal of this study measure association and causality between the dimensions of governance and economic growth in the Economic Community of West African State countries homogenously and heterogeneously.

CHAPTER THREE
THEORETICAL FRAMEWORK AND METHODOLOGY
3.1Theoretical Framework
This study is based on the governance-growth hypothesis. Professionalization of the bureaucracy is an important factor that drives economic growth according to the study not leaving out institutional and policy perspective (Wilson, 2015). The hypothesis presented that long-term investment that in turn drive growth can be promoted by a stable and trusted bureaucracy. Also, the effective enforcement by an impartial system of governance will bring about a conducive environment that can spur innovation and investment for economic growth. Also by growth-governance hypothesis, economic growth can spur increase in the quality of governance. When an economic starts at thriving and begins to growth, there is a high demand for more of property rights, rule of law and other necessary policy to sustain growth. Following the hypothesis, the model is specified using the Toda-Yamamoto’s approach following the VAR system:
Governancei,t = a1,i+ k-1ki?1,ikGovernancei, t-k+ k-1ki?1,i(k)(GDP)i,t-k+ ?1,i,t (3.1)
GDPi,t = a2,i+ k-1ki?2,ikGDPi, t-k+ k-1ki?2,i(k)(Governance)i,t-k+ ?2,i,t(3.2)
3.2Methodology
To test for the causality of quality of governance and economic growth in the countries of the region of ECOWAS, Granger Causality test will be used to test relations from governance to economic growth and from economic growth to governance. VAR/VECM model is used for this study and reason been that to test the potential for causal relationships to differ across countries in the ECOWAS region which may be as a result of difference in their private-state interaction and level of government participation at the onset of the country’s economy. Taking this heterogeneity into consideration, heterogeneous panel vector autoregressive VAR/VECM model is adopted to examine the focus of the study, governance and growth nexus, a case of ECOWAS.

3.2.1Model Specification
In achieving the objectives of this research by carrying a robust analysis on determining whether governance causes growth and vice-versa, this study adopts the model of Wilson and made adjustment to the model by including three dimensions of the governance variables which are: Economic Dimension, Political Dimension, and Institutional Dimension. i = 1,….,N for countries and t = 1,….T for time to test for the cross country level heterogeneity.

GDPi,t = a1,i+ k-1ki?1,ikGDPi, t-k+ k-1ki?1,i(k)(ECO)i,t-k+ k-1ki?1,i(k)(POL)i,t-k+k-1ki?1,i(k)(INS)i,t-k+ ?1,i,t (3.3)
ECOi,t = a1,i+ k-1ki?1,ikECOi, t-k+ k-1ki?1,i(k)(GDP)i,t-k+ k-1ki?1,i(k)(POL)i,t-k+k-1ki?1,i(k)(INS)i,t-k+ ?1,i,t (3.4)
POLi,t = a1,i+ k-1ki?1,ikPOLi, t-k+ k-1ki?1,i(k)(GDP)i,t-k+ k-1ki?1,i(k)(ECO)i,t-k+k-1ki?1,i(k)(INS)i,t-k+ ?1,i,t (3.5)
INSi,t = a1,i+ k-1ki?1,ikINSi, t-k+ k-1ki?1,i(k)(GDP)i,t-k+ k-1ki?1,i(k)(ECO)i,t-k+k-1ki?1,i(k)(POL)i,t-k+ ?1,i,t (3.6)
Where ai are the country level effects, (GDP)i,t which represent economic growth and (ECO)i,t economic dimension of governance,(POL)i,t represents political dimension of governance, (INS)i,t represent institutional dimension of governance. Both GDP and the dimensions of governance are stationary variables and ?i,t are normally distributed error terms with mean zero.

The heterogeneity of the cross-country is accounted for in the model by allowing the coefficients ?(k)and ?(k), and the lag length Ki, to vary across the countries. Finding a significant effect in the model when the two null hypotheses for causality are tested and the coefficients ?1,i=(?1,i(1),..…,?1,iki) and ?2,i=(?2,i1,….,?2,1ki) are found to be zero for all country against the alternative will be considered as evidence for the presence of the corresponding casual relationship in at least one country in the sample.

The research design for the study is based on multiple regression, which is estimated by employing the VAR/ VECM model. This study made use of secondary data covering the period of 1996 to 2016. The Panel Unit Root test summary consisting of Levin-Lin-Chu (to check for the common unit-root among the variables), Im-Pesaran-Shin and Augmented Dickey-Fuller (ADF-Fisher) will be used to check that variables are stationary and will be used to determine the order of integration for each of the variables that has the alternative hypothesis that at least one cross-section unit is stationary. The Pedroni Residual Co-integration test will be used to check if there is long-run equilibrium relationship among the variable. For absence of co-integration, we run the VAR model. We will also use Wald coefficient test to check for the joint significance or short run effect. For the heterogeneous causality tests, block-bootstrapped p-values will be calculated to test for cross-sectional dependency and will also use heterogeneous panel Granger causality tests introduced by Dumitreseu and Hurlin (2012) to take account of the individual coefficient of each country in the region.

3.2.2Sources and Measurements of Data
The 15 countries under the region of ECOWAS shall be the focus for this study, using only secondary data which span across the period of twenty-one years i.e. 1996 to 2016. To capture economic growth this study uses GDP and in measuring quality of governance, data were sourced from The Worldwide Governance Indicators which is based on the information provided by various organizations worldwide. The quality of governance used will be the un-weighted average of the six governance indicators which are: Voice and accountability; Political stability and absence of violence; Government effectiveness; Regulatory quality; Rule of law; Control of corruption.
Table 3.1Descriptions of Variables and Data Sources
S/N VARIABLES MEASUREMENT(S) SOURCES
1 Governance Aggregated governance measure which consist of an average value of six governance indicators World Bank
2 Gross Domestic Product Growth, measured by GDP which is the market value of all goods and services produced within a country World Bank
3 Economic Dimension of Governance It is the average of both government effectiveness and regulatory control World Bank
4 Political Dimension of Governance The average of voice and accountability and political stability and absence of violence World Bank
5. Institutional Dimension of Governance The average of rule of law and control of corruption World Bank
Source: Author’s Computation
CHAPTER FOUR
PRESENTATION AND ANALYSIS OF RESULT
4.1Presentation of Results
Since it is never known aprior whether the selected variables for the study are integrated, co-integrated or stationary in its trend, the empirical result presented in this analysis starts with the test of the time-series in order to examine the stationarity problem and the possible presence of unit roots in series with Panel Unit-root test which will be followed by Pedroni Residual Co-integration Test.

4.1.1Panel Unit-root test for time series
Testing the time-series of GOV and GDP is the first step of the Toda-Yamamoto approach on Granger causality in order to investigate the presence of unit root in series and to determine their order of integration.

Table 4.1 Panel Unit-root Test
Variables Method At level At First Difference Comment
Statistics Prob. Statistics Prob. I(1)
GOV Levin, Lin & Chu t* -1.13175 0.1289 -11.3861 0.0000 IPS W-Stat -0.24576 0.4029 -8.92861 0.0000 ADF – Fisher Chi-Square 29.2187 0.5061 129.803 0.0000 PP-Fisher Chi-Square 27.8819 0.5767 131.371 0.0000 GDP Levin, Lin & Chu t* -1.57985 0.0571 -8.29686 0.0000 I(1)
IPS W-Stat 3.18975 0.9993 -10.1800 0.0000 ADF – Fisher Chi-Square 40.2207 0.1007 151.787 0.0000 PP-Fisher Chi-Square 28.4673 0.5457 182.115 0.0000 ECO Levin, Lin & Chu t* -2.27893 0.0113 I(0)
IPS W-Stat -5.28602 0.0000 ADF – Fisher Chi-Square 87.0778 0.0000 PP-Fisher Chi-Square 107.403 0.0000 POL Levin, Lin & Chu t* -3.11586 0.0009 I(0)
IPS W-Stat -3.39338 0.0003 ADF – Fisher Chi-Square 70.4734 0.0000 PP-Fisher Chi-Square 73.0552 0.0000 INS Levin, Lin & Chu t* -1.93248 0.0267 I(0)
IPS W-Stat -4.42176 0.0000 ADF – Fisher Chi-Square 76.6938 0.0000 PP-Fisher Chi-Square 77.1543 0.0000 Source: Author (Using Eviews 9)
Table 4.1 gives the summary of the panel unit root test; the null hypothesis is that there is unit root which means the variables are not stationary. Rejecting the null hypothesis means validation of an alternative hypothesis that there is no unit root which is based on the significance level of 5%. In the GOV time-series, the Levin-Lin Chu t* statistics for common unit root p-value (0.1289) > 0.05, we fail to reject the null hypothesis of the presence of unit root. Also the individual unit root test conducted by the Im-Pesaran-Shin, Augmented Dickey-Fuller (ADF-Fisher) and PP-Fisher statistics at the p-value of 0.4029, 0.5061 and 0.5767 respectively which is greater than the specified significance level of 5%, we fail to reject the null hypothesis. Since it is not stationary at first level, we apply first order difference to make it stationary at d=1. With the result from first difference, all the method of the test statistical p-value < 0.05 and we therefore fail to accept the null hypothesis that there is presence of unit root.

In the GDP time-series, the Levin-Lin Chu t* statistics for common unit root p-value (0.0571) > 0.05, we fail to reject the null hypothesis of the presence of unit root. Also the individual unit root test conducted by the Im-Pesaran-Shin, Augmented Dickey-Fuller (ADF-Fisher) and PP-Fisher statistics at the p-value of 0.9993, 0.1007 and 0.5457 respectively which is greater than the specified significance level of 5%, we fail to reject the null hypothesis. Since the GDP is not stationary at first level, we apply first order difference to make it stationary at d=1. With the result from first difference, all the method of the test statistical p-value < 0.05 and we therefore fail to accept the null hypothesis that there is presence of unit root.
In the ECO time-series, the Levin-Lin Chu t* statistics for common unit root p-value (0.0113) which is significant at 5%, we fail to accept the null hypothesis that there is the presence of unit root. Also the individual unit root test conducted by the Im-Pesaran-Shin, Augmented Dickey-Fuller (ADF-Fisher) and PP-Fisher statistics at the p-value of 0.0000, 0.0000 and 0.0000 respectively which is in accordance with the specified significance level of 5%, we will reject the null hypothesis that there is presence of unit root which means the data are stationary
Also in the POL time-series, the Levin-Lin Chu t* statistics for common unit root p-value (0.0009) < 0.05, we fail to accept the null hypothesis of the presence of unit root. Furthermore, the individual unit root test conducted by the Im-Pesaran-Shin, Augmented Dickey-Fuller (ADF-Fisher) and PP-Fisher statistics at the p-value of 0.0003, 0.0000 and 0.0000 respectively which is lesser than the specified significance level of 5%, we fail to accept the null hypothesis that there is presence of unit root.
Also in the INS time-series, the Levin-Lin Chu t* statistics for common unit root p-value (0.0267) < 5%, we reject the null hypothesis of the presence of unit root. Also the individual unit root test conducted by the Im-Pesaran-Shin, Augmented Dickey-Fuller (ADF-Fisher) and PP-Fisher statistics at the p-value of 0.0000, 0.0000 and 0.0000 respectively which is greater than the specified significance level of 5%, we therefore fail to accept the null hypothesis that there is presence of unit root.
4.1.2Panel Co-integration Test
With all the variables not stationary of the same order, we will make them stationary at the same order because all the variables must be stationary of the same order before we can run a co-integration test. We will conduct the Pedroni Residual Co-integration Test to ascertain if they have inner long-run relationship as this will determine the direction of whether to use VAR or VECM. One of the conditions for running the panel con-integration test is that the variables be stationary at the same level which the variables used for this study satisfied.

Table 4.2: Pedroni Residual Co-integration Test
Methods Common AR Coefficient
Statistic Prob. Weighted
Statistic Prob.

Panel v-Statistic -2.948185 0.9984 -2.612984 0.9955
Panel rho-Statistic 2.437032 0.9926 2.188641 0.9857
Panel PP-Statistic 1.111001 0.8667 0.583068 0.7201
Panel ADF-Statistic 1.792682 0.9635 1.173057 0.8796
Individual AR Coefficients
Group rho-Statistic 3.618577 0.9999 Group PP-Statistic 1.079420 0.8598 Group ADF-Statistic 1.459869 0.9278 Source: Author (Using Eviews 9)
The null hypothesis of Pedroni Residual Co-integration test is that there is no co-integration among the variables and the decision rule is that the probability value that conforms more to either the null or alternative hypothesis out the 11 probability values in the test. From Table 4.2, the result shows that co-integration does not exist. The p-value > 0.05 which is not significant, we will fail to reject the null hypothesis that there is no co-integration among GDP, ECO, POL and INS which means that there is no long-run relationship between the variables of dimension of GOV and GDP of the 15 ECOWAS countries between 1996 – 2016; therefore, we will be using VAR for the system of analysis.

Due to the interest in the Granger non-causality tests, we will first establish the number of correct lags because the number of lags has a significant influence on the results of the Granger non-causality test. Using the lag order selection criteria, the LR test statistic (LR), Final Prediction Error (FPE), Akaike Information Criterion (AIC) and Hannan-Quinn Information Criterion (HQ) techniques reveals a maximum lag length of 2 for each of the variables.

4.2Discussion of Results
Table 4.3 VAR with GDP as the dependent variable
Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 0.130641 0.055438 2.356544 0.0192
GDP(-2) 0.123511 0.045428 2.718829 0.0070
ECO(-1) -0.019452 0.032145 -0.605124 0.5456
ECO(-2) -0.024710 0.029527 -0.836845 0.4034
POL(-1) 0.042174 0.015700 2.686279 0.0077
POL(-2) 0.006540 0.015071 0.433985 0.6647
INS(-1) -0.012252 0.030482 -0.401936 0.6881
INS(-2) 0.006576 0.028915 0.227412 0.8203
C 0.031770 0.004085 7.776377 0.0000
R-Squared 0.103450 F-statistic 3.764488
Adj. R-Squared 0.075969 Prob(F-statistic) 0.000340
Durbin-Watson stat 2.108795 Source: Author’s Compilation with Eviews 9
The result presented in table 4.3 shows the vector autoregression model with GDP as the dependent variable. The result shows that the lag of GDP has both positive and with p-values (0.0192 and 0.0070) less than 5 %, it has significant impact on GDP. In the first period, 1% increase in the GDP will bring about 13% increase in subsequent period, while in the second period, 1% increase in GDP will bring about 12% increase in GDP. The result also reveals that the economic dimension of governance (ECO) in the two lag has a negative impact on GDP but it is not significant. Also, the result shows that POL (political dimension of governance) has both positive and negative effect on GDP. In the first period, 1% increase in the POL will bring about 4.21% increase in GDP and with p-value of 0.0077 which is significant, while in the second period, 1% increase in POL will bring about 0.6% increase in GDP but with p-value of 0.6647, it does not have a significant impact. The result also reveals that the institutional dimension of governance (INS) in the two lag has both positive and negative impact on GDP but it is not significant.

The R-square shows that the model captures 10% variation in the dependent variable and this portion is explained by the regressors in the model. This became a major concern but looking at the previous studies conducted on this subject matter, the R-square(s) of the studies were low. Mira, R. and Hammadache, A. (2017) using cross-country panel data of 42 countries also arrived at a low R-Square of 17%. Bota-Avram et al (2018), using 136 countries which is inarguably a very much larger sample than what is covered in this study have a R-square of 45%. From the figures in chapter two, the level of disparity between growth and the three dimensions of governance for countries in the region of ECOWAS is indeed wide which also shows that only a little can be explained of growth by this dimensions of governance in various countries of ECOWAS.

Table 4.4Wald Coefficient Test for ECO, POL and INS on GDP
Variables Test Statistic Value Df Probability
ECO Chi-square 0.906416 2 0.6356
POL Chi-square 7.237414 2 0.0268
INS Chi-Square 0.261126 2 0.8776
Source: Author’s Compilation with eviews 9
The Wald Test for joint significance of both lag of the GDP to see if the lags combined together can cause the GDP, the null hypothesis is that the both lag of GDP cannot cause improvement in the level of growth. From Table 4.4, the p-value > 0.05 for the economic dimension of governance (ECO), we will therefore accept the null hypothesis which means that ECO in the first and second period cannot jointly bring a significant effect on the GDP. Also from the result, the p-value (0.0268) < 0.05 for the political dimension of governance (POL), we will therefore reject the null hypothesis which means that POL in the first and second period can jointly bring a significant effect on the GDP. Lastly from table 4.4, the institutional dimension of governance (INS) cannot also jointly bring significant effect on GDP.
Table 4.5 VAR with ECO as the dependent variable
Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 0.341198 0.167068 2.042272 0.0421
GDP(-2) 0.042411 0.136903 0.309788 0.7570
ECO(-1) -0.505749 0.096873 -5.220760 0.0000
ECO(-2) -0.008973 0.088983 -0.100839 0.9198
POL(-1) -0.025827 0.047313 -0.545873 0.5856
POL(-2) 0.093953 0.045417 2.068687 0.0396
INS(-1) 0.106503 0.091861 1.159399 0.2474
INS(-2) 0.080994 0.087138 0.929487 0.3535
C -0.026807 0.012312 -2.177282 0.0304
R-Squared 0.510710 F-statistic 34.05328
Adj. R-Squared 0.495713 Prob(F-statistic) 0.000000
Durbin-Watson stat 1.694825 Source: Author’s Compilation with Eviews 9
Analyzing the result in table 4.5 which shows the vector autoregression model with ECO as the dependent variable. The result shows that the lag of GDP has a positive and negative impact on the economic dimension of governance (ECO). In the first period, one percent increase in the GDP will bring about 34% increases in the economic dimension of governance and it has a significant impact. Furthermore, the second lag of GDP shows a negative relationship and non-significant on ECO. Also from the result, ECO has a significant effect from itself and POL in the first period and second period respectively. The Institutional dimension of governance effect is not significant. The R-square suggests that the model is okay as about 51% variation in the dependent variable is explained by the regressors in the model. Growth in the ECOWAS region can enhance an improvement in the economic dimension of governance.

Table 4.6 VAR with POL as the dependent variable
Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) -0.399353 0.243870 -1.637566 0.1027
GDP(-2) -0.230054 0.199838 -1.151206 0.2507
ECO(-1) -0.637204 0.141406 -4.506211 0.0000
ECO(-2) -0.212251 0.129889 -1.634088 0.1034
POL(-1) -0.199231 0.069063 -2.884775 0.0042
POL(-2) 0.046732 0.066295 0.704910 0.4815
INS(-1) 0.344147 0.134090 2.566539 0.0108
INS(-2) 0.303153 0.127196 2.383357 0.0179
C 0.032925 0.017972 1.832010 0.0681
R-Squared 0.386900 F-statistic 20.58819
Adj. R-Squared 0.368108 Prob(F-statistic) 0.000000
Durbin-Watson stat 1.728673 Source: Author’s Compilation with Eviews 9
Analyzing the result in table 4.6 which shows the vector autoregression model with POL as the dependent variable. The result shows that the lag of GDP has negative impact on the political dimension of governance (POL). In the first period, one percent increase in the GDP will bring about 39.9% decreases in the political dimension of governance though it does not have a significant impact. But this displays the picture of things in the ECOWAS countries. As GDP increases political stability is been threaten and terrorism emerges as group of citizens coarse together to share in the national cake by forceful means thereby bringing a decrease in the quality of governance from the political dimension. Furthermore, the second lag of GDP shows a negative relationship and non-significant on ECO. Also from the result, ECO has negative impact on POL but not significant. The Institutional dimension of governance effect has a significant. A 1% increase in INS will bring about 34% and 30% increases in POL in the first and second period respectively. The R-square shows that about 38% variations in the dependent variable is explained by the regressors in the model. Growth in the ECOWAS region cannot enhance an improvement in the political dimension of governance.

Table 4.7 VAR with INS as the dependent variable
Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 0.120975 0.176308 0.686154 0.4932
GDP(-2) 0.181352 0.144475 1.255247 0.2105
ECO(-1) -0.364232 0.102231 -3.562840 0.0004
ECO(-2) 0.044411 0.093905 0.472933 0.6367
POL(-1) 0.045972 0.049930 0.920740 0.3580
POL(-2) 0.047799 0.047929 0.997285 0.3195
INS(-1) -0.088730 0.096942 -0.915292 0.3609
INS(-2) 0.080874 0.091958 0.879467 0.3800
C -0.017050 0.012993 -1.312262 0.1906
R-Squared 0.448572 F-statistic 26.53954
Adj. R-Squared 0.431670 Prob(F-statistic) 0.000000
Durbin-Watson stat 1.755360 Source: Author’s Compilation with Eviews 9
Analyzing the result in table 4.7 which shows the vector autoregression model with INS as the dependent variable. The result shows that the lag of GDP has a positive on the institutional dimension of governance (INS) though it is not significant. In the first period of the economic dimension of governance, it has a negative and significant impact on the institutional dimension of governance. A 1% increase in the economic dimension of governance will bring about 36% decline in the institutional quality of governance. In the second period of ECO, it has no significant impact. The political dimension of governance (POL) does not have a significant impact on institutional dimension and also, the institutional dimension do not have a significant impact in itself. The R-square shows that the model captures about 44% variations in the dependent variable are explained by the regressors in the model.
Table 4.8Wald Coefficient Test for GDP on ECO, POL and INS
Variables Test Statistic Value Df Probability
ECO Chi-square 4.956290 2 0.0839
POL Chi-square 5.395264 2 0.0674
INS Chi-square 2.697270 2 0.2596
Source: Author’s Compilation with eviews 9
The Wald Test n table 4.8 is to test for joint significance of both lag of the GDP to see if the lags combined together can cause the changes in the three dimension of governance individually; the null hypothesis is that the both lag of governance cannot cause growth. From Table 4.8, the p-value > 0.05 for all the variables of dimension of governance, we will therefore reject the null hypothesis which means that GDP in the first and second period can jointly bring a significant effect on the economic, political and institutional dimension of governance.

Table 4.9: Granger Causality/Block Exogeneity Wald Tests
Dependent variable: GDP
Excluded Chi-Square Df Prob.

ECO  0.906416 2  0.6356
POL  7.237414 2  0.0268
INS  0.261126 2  0.8776
ALL (GOV)  12.66152 6  0.0487
Dependent Variable: ECO
GDP  4.956290 2  0.0839
Dependent Variable: POL
GDP  5.395264 2  0.0674
Dependent Variable: INS
GDP  2.697270 2  0.2596
Source: Author’s Compilation with eviews 9
The Table 4.9 shows the result of the VAR Granger Causality/Block Exogeneity Wald tests with the null hypothesis that no granger causality and it reveals the a tangible evidence of Granger causality runs from the overall country-level governance (ALL GOV) to GDP with the p-value ; 0.05 which is significant in the joint test. Breaking down the governance index into the three dimensions; economic and institutional dimension of governance do not have evidence of causality running from them to GDP but there is causality running from political dimension to GDP. The causality from GDP to the three dimension of governance is not confirmed since the p-value is ; 0.05. From this, it is established that it is a unidirectional casual relationship that exist and it is from the quality of governance to growth.

Table 4.10 Pairwise Dumitrescu Hurlin Panel Causality Tests
Null Hypothesis W-Stat. Prob.

 POL does not homogeneously cause LGDP  5.47220  4.40329 1.E-05
 LGDP does not homogeneously cause POL  3.06879  1.03172 0.3022
 INS does not homogeneously cause LGDP  4.04943  2.40739 0.0161
 LGDP does not homogeneously cause INS  4.46812  2.99474 0.0027
 ECO does not homogeneously cause LGDP  4.03327  2.38473 0.0171
 LGDP does not homogeneously cause ECO  2.42999  0.13559 0.8921
Source: Author’s Compilation with eviews 9
Allowing for difference in the coefficient of each nation in the region Table 4.8 shows the result of heterogeneous panel granger causality for each cross-section unit independently and the result still tally along with the result of the VAR Granger Causality/Block Exogeneity Wald tests. It shows evidence of Granger causality from country-level governance to GDP for p-value < 0.05 but the causality from GDP to quality of governance is not confirmed since the p-value is > 0.05. Unidirectional casual relationship is what still exists and it is from the quality of governance to growth.

The result obtained in this analysis is mainly as a result of nations in the ECOWAS such as Cape Verde that has a high level of quality of governance yet they have a very low GDP. While some countries such as Cote d’Ivoire and Nigeria with low level of quality of governance and yet maintain a high GDP, accounting for over 70% of the GDP of the Region and being referred to as the economic powerhouse of West Africa (ECOWAS Annual Report, 2016). So there is causality when we homogeneously test between the quality of governance and growth because of this happening highlighted as regarding countries with low quality of governance and high growth that contradicts economic theory or it could be a pointer that the yardstick in capturing governance as given by Kaufmann et al (2005) cannot be fully relied upon to explain growth for the ECOWAS region community.
4.3Comparison of Results with Previous Findings
The result of this study conforms partly to economic literature and it has confirmed some of the results previous studies such as Bota-Avram et al (2018), Calderoân and Chong (2000), Chauvet and Collier (2004) which provide evidence of Granger causality from quality of country level governance to economic growth and so also the causality from economic growth to country-level governance is not confirmed. The result of this study, same as their own result follows the empirical findings of the World Bank researcher that concludes there is no significant influence of economic growth on the quality of governance establishing unidirectional causality because in the opposite direction there is a strong positive causal effect from governance to economic growth. It is noted also that the result of this study is at variance with Sachs et al. (2004) which find out an important challenge to the significance of good governance for the economic growth of African countries. He further asserts that the differences in performance among African countries cannot be explained by differences in the quality of their governance once differences in their levels of development have been accounted for and thus concludes that a focus on governance reforms is misguided. Wilson (2015) which uses heterogeneous Granger causality introduced by Dumitreseu and Hurlin (2012) allowing for potential differences in the causal relations across provinces showing a significant and positive effect of quality of governance on economic growth and also confirms a bi-directional causality between the two variables which is at a variance with this study.
CHAPTER FIVE
SUMMARY, CONCLUSION AND RECOMMENDATION
5.1Summary of Findings
This study focused on the dimensions of governance and growth nexus: a case of ECOWAS between the period of 1996 and 2016. Cross-country data is what is mainly used by many studies in the analysis of governance and growth due to the fact that the measure of governance in based on perceptions. In the period of this study, both economic, political and institutional dimensions of governance seems to have a minute change with the economic growth of the countries in the region witnessing different ups and downs in its movement. The findings of this study show a positive and significant effect of the political dimension of governance on economic growth while the economic and institutional dimension of governance does not have a significant effect. In the opposite relation there is no significant effect of economic performance on the quality of governance in any of its three dimensions.
The model used in estimating in this study employed gross domestic product (GDP) and Governance broken down into its three dimensions; Economic (ECO), Political (POL) and Institutional (INS). Due to no co-integration among the variables, the vector autoregressive model, Wald coefficient test, block exogeneity Wald test and Dumitrescu Hurlin Granger Causality test were employed as the estimation techniques in the study and the findings include: (i) There is a unidirectional causality running from the quality of governance as a whole to GDP (ii) Improvement in economic growth do not generate improvement in any of the dimension of governance which implies there is no significant effect of causality running from economic growth to economic, political and institutional dimension of governance (iii) The panel unit root test indicates that the gross domestic product(GDP) variable used in capturing growth was not stationary at level but when first differenced, they were all stationary; and (iv) The Pedroni Residual Panel Co-integration test confirmed that there is no long run associated between the the dimensions of governance and economic growth.

5.2Conclusion
The dimension of governance and growth nexus in the region of the ECOWAS between 1996 and 2016 has been explored in this study at both theoretical and empirical level covering the period of twenty-one years. The model estimated revealed that the political dimension of governance has a positive effect on growth but it is significant while there is negative effect from the economical and institutional dimension of governance but its effect is not significant. The causality test shows a significant impact of governance on growth but no reverse case. Theoretically and logically, the improvement in the resources available to the government ought to bring about causality running to quality of governance from growth but the region has exhibited something different. Amidst the four traps identified by Collier (2007), the bad governance trap is one that many of the ECOWAS countries fall a victim of as shown in the estimate of their quality of governance being in deficit. However, improvement in the quality of the political dimension can spur economic growth. Therefore, this study rejects the null hypothesis and concludes that there is a relationship between governance and growth. The most important lesson drawn from this study is that in examining the improvement of the quality of the dimension of governance in the ECOWAS economy, growth is not a factor that can initiate it.
5.3 Recommendations
Based on the result obtained in this study of the dimension of governance and growth nexus: a case of ECOWAS between 1996 and 2016, the following policies are recommended in the study:
Policy makers both domestic and external may have to place significant emphases on the maintenance of the voice and accountability, political stability, government effectiveness, regulatory quality, rule of law, and control of corruption in other to attract more foreign direct investment (FDI) into the region of ECOWAS.

Political dimension of governance which constitute voice and accountability and political stability and absence of corruption needs more attention as it has a significant influence on growth.

The ECOWAS committees can also bring about incentive for improvement in the quality of governance as we have it that some of the countries in bad governance will be exempted from some benefits.

Property right and contract laws that are well backed by an effective and impartial judiciary system of government should be put in place to drive growth.

International agencies that aid can also give some benchmark level of governance reform that must be achieved before a certain level of aid can be given to the countries in the region so as to improve the quality of governance.

Economic growth cannot bring improvement in the quality of governance, so while trying to work on improving quality of the dimension of governance, the actors should look beyond economic growth to achieve it.

5.4Limitation of the study
The limitation encountered by this study mainly has to do with the gathering of data of some countries in the region of ECOWAS. In addition to that, the study also face the constrain of time because the shortness of the time frame for the study did not give much room to dig deeper in considering other subject matter attached to the topic of discussion.
5.5Suggestions for the future research
For further research on this topic, the following areas are suggested for further research to contribute to the existing knowledge.

Further studies using the dimension of governance broken down into economic, political and institutional when considering causality between governance and growth to know if there is improvement to bring about changes in the significance on the variables.

Also, same study (The dimension of governance and growth nexus) can be conducted on other region of the African continent to confirm if the findings of the research hold there also.
Further studies can also examine the nexus of the dimension of governance and growth looking at the sectoral contribution of the primary, secondary and tertiary sector to the GDP of the nation.

Control variables can be added to the model in other to have more grips on the relationship between governance and growth.

The effect of the dimension of governance (economical, political and institutional) can be examined on the sustainable development of the region.

REFERENCES
Allafrica (2000). West Africa: ECOWAS to Isolate Members for Bad Governance.

Acemoglu, D. and Robinson, J. A., (2012), “Why Nations Fail: The Origins of Power, Prosperity, and Poverty”, New York: Crown Business.
Acemoglu, D. and Robinson, J. A., (2008), “The Role of Institutions in Growth and Development”. The International Bank for Reconstruction and Development, The World Bank, WP#10.
AlBassam, A. B., (2013). The Relationship Between Governance and Economic Growth During Times of Crisis. European Journal of Sustainable Development (2013), 2, 4, 1-18 ISSN: 2239-5938
Blaydes, L. and Kayser, M. A., (2011), “Counting Calories: Democracy and Distribution in the Developing World” International Studies Quarterly, 55 (4), pp. 887–908.
Bota-Avram C., Grosanu A., Rachisan P., and Gavriletea M. (2018). The Bidirectional Causality between Country-Level Governance, Economic Growth and Sustainable Development: A Cross-Country Data Analysis. MDPI. Sustainability 2018, 10, 502
Campos, N.F., and Nugent, J.B. (1999), “Development Performance and the Institutions of Governance: Evidence from East Asia and Latin America,” World Development, Vol. 27, No. 3:439-452.
Chauvet, Lisa and Paul Collier (2004), “Development Effectiveness in Fragile States: Spillovers and Turnarounds”, Centre for the Study of African Economies, Department of Economics, Oxford University (Mimeo).

Dixit, A. K. (2004). Lawlessness and Economics: Alternative modes of Governance. Princeton, NJ: Princeton University Press.

Drewry, G., ; Chan, C.-P. (2001). Civil Service Reform in the People’s Republic of China: Another Mirage of the New Global Paradigm of Public Administration. International Review of Administrative Sciences, 67, 461-478
ECOWAS (2016). ECOWAS Common External Tariff (CET): Achievements, Challenges and Prospects. Abuja
Emara, N., Jhonsa, E. (2014), “Governance and Economic Growth: The Case of Middle East and North African Countries”, Journal of Development and Economic Policies, 16(1): 47-71.
Fiyassa, B., and Nsiah, C., (2013). The Impact of Governance on Economic Growth in Africa. The Journal of Developing Areas. Research gate. DOI: 10.1353/jda.2013.0009
Gareth, J, Daniela, W., Trevor, H., and Tibshirani R., (2013), “An Introduction to Statistical Learning with Applications in R, Springer Science + Business Media, New York.
Gerring, J., P. Kingstone, M. Lange, and A. Sinha (2011), “Democracy, History and Economic Performance: A Case-Study Approach”, World Development, 39 (10), pp. 1735–1748.

Gray, H., (2007) Governance for Economic Growth and Poverty Reduction: Empirical Evidence and New Directions Reviewed. The London School f Economics and Political Science. Department for International Development/World Bank.
Han, X., Khan, H., and Zhuang, J. (2014), “Do Governance Indicators Explain Development Performance? A Cross-Country Analysis”, ADB Economics Working Paper Series, No.417.
Haq, R. and Zia, U. (2009). Does Governance Contribute to Pro-poor Growth? Evidence from Pakistan. PIDE Working Papers 2009:52
Johnson, E. (2012), “Corruption Trends in the Middle East and North Africa Region”, 2007-2011, U4 Expert Answer, Jan, No. 302.

Kaufmann, D. and Kraay, A. (2003). Governance and Growth: Causality which way? – Evidence for the World, in brief
Kaufmann, D., Kraay, A. and Mastruzzi, M. (2005), “Governance Matters IV: Governance Indicators for 1996-2004”, World Bank Policy Research Working, Paper No., Washington DC: World Bank.
Kaufmann, D., Kraay, A., and Zoido-Lobatón, P., (2002). “Governance Matters II –Updated Indicators for 2000/01”. World Bank Policy Research Department Working Paper No. 2772, Washington, D.C.
Keefer, P., (2007). Governance and Economic Growth in China and India: Dancing with Giants: China, India, and the Global Economy (pp.191-215)
Khan, M. H., (2008). Governance and Development: The Perspective of Growth-enhancing Governance. In Diversity and Complimentarity in Development Aid: East African Lessons for African Growth. (pp.107-152) Tokyo: GRIPS
Knack, S. 2002. “Governance and growth: Measurement and evidence”. IRIS Discussion Papers on Institutions and Development Paper No. 02/15. IRIS Centre. Washington, D.C
Lahouij H., (2016). Does Governance Matter to Economic Growth? Evidence from MENA Countries. ECN 5900
Mehanna, R., Yazbeck, Y., and Sarieddine, L. (2010), “Governance and Economic Development in MENA Countries: Does Oil Affect the Presence of a Virtuous Circle?”, Journal of Transnational Management, 15 (2), p. 117-150.

Mira, R. and Hammadache A., (2017). Relationship Between Good Governance and Economic Growth: A Contribution to the Institutional Debate about State Failure in Developing Countries. CEPN, UMR-CNRS 723
Musah, A. (2009). West Africa: Governance and Security in a Changing Region. Africa Program Working Paper Series. International Peace Institute.

OECD (2009), “Better Polices for Better Lives, Business Climate Development Strategy the Authority of the Steering Groups of the MENA-OECD Initiative” http://www.oecd.org/globalrelations/46341460.pdf.
Oster, E. (2009), “Does Increased Access Increase Equality? Gender and Child Health Investments in India”, Journal of Development Economics.,89 (1), pp. 62-76.
Paul Collier (2007). The Bottom Billion: Why the Poorest Counties are Failing and What can be Done About it. New York, NY: Oxford University Press.

Pintak, L.(2011), “Breathing Room Toward a New Arab Media”, Columbia Journalism Review, June. http://www.cjr.org/cover_story/breathing_room.php.
Sachs, J. D., McArthur J. W., Guido S., Margaret K., Chandrika B., Michael F. and Gordon M., (2004). Ending Africa’s Poverty Trap, Brookings Papers on Economic Activity 1: 117-240.
Siddiqui, D. A. and Ahmed Q. M. (2009). Institutions and Economic Growth: A Cross country Evidence. Munich Personal RePEc Archive. MPRA Paper No. 19747
Stojanovic, I., Ateljevic, J., and Stevic, R.S., (2016). Good Governanve as a Tool of Sustainable Development. European Journal of Susutainable Development 2016,5,558-573
UNDP (2010), “Voice and Accountability for Improved Service Delivery”, UNDP Regional Training Event for the Arab States, October, Cairo, Egypt. http://www.undp.org/content/dam/undp/documents/partners/civil_society/facilitating_citizen_action/VA_background_paper_final.pdf.
Wilson, R. (2015). Does Governance cause Growth? Evidence from China. World Dev. 2016,79, 138–151.

World Bank, (2003) “Better Governance for Development in the Middle East and North Africa: Enhancing Inclusiveness and Accountability”, http://www-wds.worldbank.org/servlet/WDSContentServer/WDSP/IB/2003/11/06/000090341_20031106135835/Rendered/PDF/271460PAPER0Be1ance0for0development.pdf
Zhuang, J., E. de Dios, and A. Lagman-Martin (2010), “Governance and Institutional Quality and the Links with Growth and Inequality: How Asia Fares”. In J. Zhuang, ed. Poverty, Inequality, and Inclusive Growth in Asia: Measurement, Policy Issues, and Country Studies. London: Anthem Press and Manila: Asian Development Bank.
APPENDIX A
Data on Quality of Governance and Economic Growth
COUNTRIES CID YEAR ECO POL GDP
BENIN 1 1996 -0.408391714 0.537184749 3888602421
  1 1997 -0.33781229 0.518630462 4111601652
  1 1998 -0.267232865 0.500076175 4274462693
  1 1999 -0.33781229 0.54652314 4502780954
  1 2000 -0.266060919 0.539256498 4766643562
  1 2001 -0.290368691 0.528618604 5020725248
  1 2002 -0.383612484 0.366424511 5253911327
  1 2003 -0.331409842 0.450436391 5434858389
  1 2004 -0.361421257 0.178732369 5675602454
  1 2005 -0.622927666 0.230774139 5772744796
  1 2006 -0.569267273 0.42264393 6000595826
  1 2007 -0.517233074 0.33364135 6359822427
  1 2008 -0.449205786 0.317059577 6671033591
  1 2009 -0.55637753 0.350929961 6825749998
  1 2010 -0.577983975 0.254887655 6970240895
  1 2011 -0.523067057 0.214626223 7176751732
  1 2012 -0.509002984 0.244426537 7521973830
  1 2013 -0.487232715 0.228342883 8063036668
  1 2014 -0.463309467 0.146721025 8575645914
  1 2015 -0.617992043 0.142651406 8755259214
  1 2016 -0.522844742 0.27533645 9103831278
BURKINA FASO 2 1996 -1.02304256 -0.573134005 4045012129
  2 1997 -0.886284292 -0.480371088 4300528860
  2 1998 -0.749526024 -0.387608171 4614799452
  2 1999 -0.685630769 -0.286958934 4956487462
  2 2000 -0.621735513 -0.186309698 5046707524
  2 2001 -0.67037046 -0.304786952 5380466456
  2 2002 -0.719005406 -0.423264205 5614678394
  2 2003 -0.620383799 -0.148030715 6052760301
  2 2004 -0.525326073 -0.280374724 6323831618
  2 2005 -0.593393922 -0.267703291 6871593161
  2 2006 -0.748452723 -0.115329884 7301284780
  2 2007 -0.735306442 -0.028986409 7714172808
  2 2008 -0.470285952 -0.111680504 8276883001
  2 2009 -0.582239509 -0.150224819 8522045465
  2 2010 -0.560449839 -0.212789766 8979966766
  2 2011 -0.544241309 -0.443794072 9575057058
  2 2012 -0.622562766 -0.439050034 10192906646
  2 2013 -0.663503289 -0.509408087 10783339867
  2 2014 -0.578332722 -0.569490224 11249917445
  2 2015 -0.586434007 -0.362033363 11688050576
  2 2016 -0.609423339 -0.453631684 12379492024
CABO VERDE 3 1996 0.332665821 0.926461667 596807529.5
  3 1997 0.325061649 0.946910352 663163133.2
  3 1998 0.347874165 0.967359036 746173157.1
  3 1999 0.325061649 0.947123305 829913547.9
  3 2000 0.302249134 0.926887572 948465609.7
  3 2001 0.093311362 0.790674941 969632085
  3 2002 -0.11562641 0.65446201 1020546411
  3 2003 -0.099812038 0.784985572 1063167076
  3 2004 -0.069832303 0.882092744 1171578994
  3 2005 -0.225601956 0.633207947 1252563900
  3 2006 0.119891539 0.859749764 1352565476
  3 2007 0.351919144 0.869962305 1557758965
  3 2008 0.057764538 0.878011525 1661358044
  3 2009 0.03621082 0.863054723 1640251720
  3 2010 -0.019361438 0.870048732 1664310770
  3 2011 0.156334966 0.845762253 1730365372
  3 2012 0.120101236 0.873903006 1749086512
  3 2013 0.095790669 0.847012162 1763128136
  3 2014 0.075785123 0.653875932 1773904599
  3 2015 0.150738552 0.912874758 1792815240
  3 2016 0.107438115 0.948937744 1863071658
COTE D’IVOIRE 4 1996 -0.059179474 -0.316529164 20585330896
  4 1997 -0.133297013 -0.407037539 21355953698
  4 1998 -0.207414553 -0.497545913 22408947356
  4 1999 -0.542713039 -0.855155923 22771418229
  4 2000 -0.878011525 -1.212765932 22300414202
  4 2001 -0.907997845 -1.402387321 22327480641
  4 2002 -0.937945366 -1.59200871 21955138166
  4 2003 -1.022945166 -1.548320293 21656650171
  4 2004 -1.266809225 -1.729726374 21923410906
  4 2005 -1.355715871 -1.830492914 22300767039
  4 2006 -1.202372909 -1.623702228 22638811513
  4 2007 -1.211128116 -1.584774137 23038394864
  4 2008 -1.212171555 -1.532402098 23624224724
  4 2009 -1.081923842 -1.198014021 24392355457
  4 2010 -1.26095736 -1.32996881 24884505035
  4 2011 -1.127276301 -1.265674949 23792758396
  4 2012 -1.079614282 -1.020605534 26340131050
  4 2013 -0.908289254 -0.899957627 28681616270
  4 2014 -0.806210995 -0.772424012 31203899802
  4 2015 -0.650963187 -0.618869811 33963218674
  4 2016 -0.788487812 -0.593027055 36794322093
GAMBIA 5 1996 -0.614799261 -0.427071825 535802330.3
  5 1997 -0.548329562 -0.312436589 562056639.6
  5 1998 -0.481859863 -0.197801352 581728613.6
  5 1999 -0.489192158 -0.271485374 618959240.9
  5 2000 -0.496524453 -0.345169395 653002000.2
  5 2001 -0.584224761 -0.153013111 690876117.8
  5 2002 -0.671925068 0.039143175 668422641.8
  5 2003 -0.465315849 -0.032260478 714343276
  5 2004 -0.498279899 -0.193265498 764704476.5
  5 2005 -0.668266118 -0.361149438 757503438.6
  5 2006 -0.687482297 -0.446272422 766018532.5
  5 2007 -0.573611617 -0.410950389 793832861
  5 2008 -0.718342781 -0.386896852 839356333.2
  5 2009 -0.628137469 -0.434498303 893492263.9
  5 2010 -0.655854702 -0.503425833 951804226.8
  5 2011 -0.606662214 -0.608936246 910923076.2
  5 2012 -0.500229001 -0.644128464 961932593.4
  5 2013 -0.696473956 -0.652995214 1007998795
  5 2014 -0.640547931 -0.709574714 1017070785
  5 2015 -0.89269346 -0.720201801 1060804826
  5 2016 -0.743238449 -0.898026496 1084315801
GHANA 6 1996 -0.114796564 -0.332535505 15541590590
  6 1997 -0.127852172 -0.317220308 16193771305
  6 1998 -0.140907779 -0.30190511 16954941838
  6 1999 -0.059967841 -0.280476654 17700958746
  6 2000 0.020972097 -0.259048197 18355894240
  6 2001 -0.052776146 -0.204839372 19090130010
  6 2002 -0.126524389 -0.150630547 19949185803
  6 2003 -0.185716361 0.128575098 20986543461
  6 2004 -0.161034778 0.078826877 22161789893
  6 2005 -0.155438989 0.209457219 23469336373
  6 2006 0.129631728 0.192301806 24971353346
  6 2007 0.108944893 0.202455267 26056812916
  6 2008 0.036415271 0.183427623 28440958948
  6 2009 -0.038271036 0.261412477 29819138460
  6 2010 -0.038427439 0.257792552 32174772956
  6 2011 -0.050459608 0.307502717 36694042412
  6 2012 -0.044368561 0.281023592 40103840659
  6 2013 -0.095546782 0.24973033 43036444041
  6 2014 -0.291979074 0.178221799 44751818870
  6 2015 -0.255614817 0.231685432 46504253396
  6 2016 -0.214380225 0.240180761 48167546935
GUINEA 7 1996 -1.244351387 -1.289356172 4352705760
  7 1997 -1.021657944 -1.115315512 4578245712
  7 1998 -0.7989645 -0.941274852 4745082536
  7 1999 -0.81296438 -1.271322593 4925965274
  7 2000 -0.826964259 -1.601370335 5049265168
  7 2001 -0.887644559 -1.524675674 5233984782
  7 2002 -0.948324859 -1.448418617 5504299656
  7 2003 -0.760615885 -1.028580815 5573026411
  7 2004 -0.884422004 -1.168546796 5703441766
  7 2005 -1.058594465 -1.157889962 5874389462
  7 2006 -1.324062347 -1.565402985 6021057021
  7 2007 -1.262849927 -1.884069562 6413123414
  7 2008 -1.185112119 -1.774216533 6678750578
  7 2009 -1.030955791 -1.693963706 6576062365
  7 2010 -1.131638885 -1.313105494 6861546076
  7 2011 -1.147789121 -1.164658159 7238226700
  7 2012 -1.263639927 -1.150450587 7666391143
  7 2013 -1.174185157 -1.126081586 7967959182
  7 2014 -1.238562822 -0.906656742 8263432291
  7 2015 -1.13575387 -0.622471929 8553142029
  7 2016 -1.18283395 -0.56972681 9119962270
GUINEA BISSAU 8 1996 -1.465660214 -1.440084219 811371237.5
  8 1997 -1.389237046 -1.492478878 864110367.9
  8 1998 -1.312813878 -1.544873536 621295529.7
  8 1999 -1.188088536 -1.088311524 627667147.8
  8 2000 -1.063363194 -0.631749511 661730564.6
  8 2001 -1.106966555 -0.721675709 676215230
  8 2002 -1.150569916 -0.811601907 669553309.5
  8 2003 -1.307752728 -0.833184928 673359788.2
  8 2004 -1.492736697 -0.621584594 691953958.8
  8 2005 -1.379492998 -0.46974282 721468737.3
  8 2006 -1.128514409 -0.590341344 738132862.4
  8 2007 -1.10763061 -0.610622048 761790372.3
  8 2008 -1.062255025 -0.744767934 786223863
  8 2009 -1.030809402 -0.733815044 812261063.1
  8 2010 -1.037178397 -0.776068509 850633309.8
  8 2011 -1.03810811 -0.830541342 919405516.2
  8 2012 -1.231732607 -1.145500869 903658701.4
  8 2013 -1.433684826 -1.121682286 933081095.1
  8 2014 -1.600133419 -0.778449804 942081094.9
  8 2015 -1.622959018 -0.633903116 999855133
  8 2016 -1.552259088 -0.602512121 1057415318
LIBERIA 9 1996 -1.873903513 -2.022718132 269199729.3
  9 1997 -1.927956462 -1.726908177 555304688
  9 1998 -1.982009411 -1.431098223 723051855.7
  9 1999 -1.912418425 -1.51655373 880253798.1
  9 2000 -1.842827439 -1.602009237 1132146638
  9 2001 -1.725794317 -1.715236545 1165190798
  9 2002 -1.608760834 -1.828463852 1209034190
  9 2003 -1.505164385 -1.872414768 844568142.7
  9 2004 -1.585014105 -1.356300235 866700939.8
  9 2005 -1.461011291 -0.86896944 912479873.1
  9 2006 -1.323827505 -0.770700708 985872107.8
  9 2007 -1.254359126 -0.725043334 1079873957
  9 2008 -1.309924364 -0.764822908 1157051893
  9 2009 -1.239752293 -0.639859721 1218376790
  9 2010 -1.270562053 -0.35863471 1292697100
  9 2011 -1.242133379 -0.365160435 1398698238
  9 2012 -1.147094846 -0.395546377 1510516487
  9 2013 -1.364991903 -0.431414887 1641989954
  9 2014 -1.340750694 -0.445641518 1653502630
  9 2015 -1.367290854 -0.519526303 1653502630
  9 2016 -1.357677817 -0.320883241 1627046588
MALI 10 1996 -1.209976912 0.056393802 5096827925
  10 1997 0 0 5342954986
  10 1998 -1.054973841 0.078686185 5747505780
  10 1999 0 0 6075167846
  10 2000 -0.872311294 0.014817432 6071472020
  10 2001 -0.736317128 0.144995868 7005036096
  10 2002 -0.600322962 0.275174305 7222634110
  10 2003 -0.598230779 0.303582266 7881269148
  10 2004 -0.619374573 0.386200383 8004216841
  10 2005 -0.686925948 0.219125859 8527273414
  10 2006 -0.68713218 0.323344797 8924830837
  10 2007 -0.734230101 0.195950747 9236630227
  10 2008 -0.764263988 0.167218879 9677529939
  10 2009 -0.789364934 -0.018085256 10130347923
  10 2010 -0.839989305 -0.036905102 10678749467
  10 2011 -0.787665427 -0.265855759 11024767958
  10 2012 -0.981540918 -1.264929414 10932581179
  10 2013 -0.884631813 -1.003335044 11184422458
  10 2014 -1.080137849 -0.937695779 11972181171
  10 2015 -0.911282539 -0.945022486 12686032241
  10 2016 -0.958684067 -0.864825547 13421822111
NIGER 11 1996 -1.244351387 -0.973269872 3288171553
  11 1997 0 0 3378726668
  11 1998 -1.081475139 -1.010386199 3730866431
  11 1999 0 0 3709646817
  11 2000 -1.084876657 -0.101569954 3657358998
  11 2001 -0.985078305 -0.183803687 3917190239
  11 2002 -0.885297954 -0.266037419 4034671555
  11 2003 -0.749958098 -0.079930381 4248509147
  11 2004 -0.686409354 -0.350693874 4252757656
  11 2005 -0.787552834 -0.387636378 4444131751
  11 2006 -0.787423372 -0.308639564 4701891392
  11 2007 -0.794261038 -0.442792475 4849839715
  11 2008 -0.716679811 -0.589022666 5314826498
  11 2009 -0.660525739 -0.970559895 5276949262
  11 2010 -0.665457964 -0.920998573 5718589799
  11 2011 -0.626749575 -0.580792978 5850769296
  11 2012 -0.687979817 -0.716379225 6541959480
  11 2013 -0.727196395 -0.81404452 6886614682
  11 2014 -0.704036534 -0.685488753 7405110305
  11 2015 -0.607292235 -0.639601663 7698208265
  11 2016 -0.679508388 -0.711487949 8085878853
NIGERIA 12 1996 -0.975662589 -1.417699277 1.40933E+11
  12 1997 0 0 1.44882E+11
  12 1998 -1.121873498 -0.951943278 1.48817E+11
  12 1999 0 0 1.49523E+11
  12 2000 -0.956948876 -1.050612062 1.57474E+11
  12 2001 -1.006692886 -1.126838044 1.64421E+11
  12 2002 -1.056436896 -1.203064024 1.70643E+11
  12 2003 -0.964787304 -1.14317888 1.88312E+11
  12 2004 -0.912588716 -1.24441272 2.51841E+11
  12 2005 -0.882583737 -1.243956655 2.60516E+11
  12 2006 -0.955720365 -1.337189376 2.81906E+11
  12 2007 -1.040885091 -1.398115337 3.01156E+11
  12 2008 -0.96831131 -1.309488237 3.20039E+11
  12 2009 -1.200623512 -1.41221717 3.42232E+11
  12 2010 -1.151616096 -1.495906651 3.69062E+11
  12 2011 -1.079214811 -1.345628083 3.871E+11
  12 2012 -0.989000261 -1.371146798 4.03665E+11
  12 2013 -0.985179424 -1.390753269 4.2544E+11
  12 2014 -1.182586908 -1.358516186 4.52285E+11
  12 2015 -0.951978087 -1.14602007 4.64282E+11
  12 2016 -1.039914807 -1.074467003 4.56775E+11
SENEGAL 13 1996 0.021935342 -0.357230339 7232428842
  13 1997 0 0 7458372326
  13 1998 -0.122764654 -0.59228681 7898317288
  13 1999 0 0 8399632633
  13 2000 -0.123349093 -0.306542568 8668335875
  13 2001 -0.051337847 -0.155656379 9065425099
  13 2002 0.020673398 -0.00477019 9124785833
  13 2003 -0.258970529 -0.012125865 9734616091
  13 2004 -0.178455621 0.083088933 10306113228
  13 2005 -0.245526865 -0.095609765 10885585457
  13 2006 -0.316319406 -0.138003892 11153541790
  13 2007 -0.460646659 -0.248329006 11704357788
  13 2008 -0.127283797 -0.21464248 12135373619
  13 2009 -0.495832086 -0.263440311 12429435070
  13 2010 -0.557307243 -0.372074991 12948906289
  13 2011 -0.468840003 -0.280242398 13176951814
  13 2012 -0.458966106 -0.056255583 13758210915
  13 2013 -0.408800125 -0.000332717 14233621993
  13 2014 -0.38972944 0.047028355 14813653949
  13 2015 -0.43514502 0.08130255 15770157927
  13 2016 -0.411224862 0.053708926 16833353304
SIERRA LEONE 14 1996 -1.465660214 -1.251475394 1384449201
  14 1997 0 0 1303083986
  14 1998 -1.456789494 -1.824424267 1326344221
  14 1999 0 0 1300092081
  14 2000 -1.460106969 -1.770475507 1386583669
  14 2001 -1.487207115 -1.268824399 1287577162
  14 2002 -1.514307261 -0.76717329 1625805355
  14 2003 -1.231035948 -0.743291512 1778344672
  14 2004 -1.101667523 -0.380342215 1892470594
  14 2005 -1.356493354 -0.473788589 1974461587
  14 2006 -1.138890028 -0.317185983 2083567040
  14 2007 -1.180179238 -0.137632102 2251467572
  14 2008 -1.176205754 -0.191409938 2373039542
  14 2009 -1.214610219 -0.28023085 2483374322
  14 2010 -1.210600734 -0.212597959 2616610911
  14 2011 -1.187075853 -0.202412836 2742474775
  14 2012 -1.205381036 -0.300298914 3158830962
  14 2013 -1.20441699 -0.25865493 3813207065
  14 2014 -1.222037792 -0.195366416 3986803557
  14 2015 -1.257693648 -0.183698129 3169794532
  14 2016 -1.228049477 -0.14390333 3369621899
TOGO 15 1996 -0.801733673 -0.78837043 2255767596
  15 1997 0 0 2580087820
  15 1998 -0.977157712 -0.978331655 2520748985
  15 1999 0 0 2583307709
  15 2000 -1.240854979 -0.811560109 2563068020
  15 2001 -1.305792272 -0.785557112 2521371869
  15 2002 -1.370729566 -0.759554114 2498121012
  15 2003 -1.54559958 -0.744134091 2621887861
  15 2004 -1.604685187 -0.798611656 2677447378
  15 2005 -1.49569881 -1.370217681 2709052153
  15 2006 -1.562157035 -0.914766639 2818834122
  15 2007 -1.553733826 -0.761378884 2883398218
  15 2008 -1.510351062 -0.626930661 2947567675
  15 2009 -1.390850186 -0.611330673 3051036045
  15 2010 -1.385352254 -0.601193994 3172945645
  15 2011 -1.362831712 -0.564725168 3327904820
  15 2012 -1.311035395 -0.691133603 3488321264
  15 2013 -1.291032672 -0.629091322 3626730618
  15 2014 -1.248044133 -0.466807738 3839682315
  15 2015 -1.179098129 -0.464938201 4041404331
  15 2016 -1.239391645 -0.353067912 4245259745
APPENDIX B
Panel Unit Root Test for GOV
Panel unit root test: Summary  Series: GOV Date: 08/11/18 Time: 16:00 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 1
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -1.13175  0.1289  15  297
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat  -0.24576  0.4029  15  297
ADF – Fisher Chi-square  29.2187  0.5061  15  297
PP – Fisher Chi-square  27.8819  0.5767  15  300
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Panel unit root test: Summary  Series: D(GOV) Date: 08/11/18 Time: 16:07 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 1
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -11.3861  0.0000  15  283
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat  -8.92861  0.0000  15  283
ADF – Fisher Chi-square  129.803  0.0000  15  283
PP – Fisher Chi-square  131.371  0.0000  15  285
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Panel Unit Root test for LGDP
Panel unit root test: Summary  Series: LGDP Date: 08/11/18 Time: 16:10 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 4
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -1.57985  0.0571  15  289
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat   3.18975  0.9993  15  289
ADF – Fisher Chi-square  40.2207  0.1007  15  289
PP – Fisher Chi-square  28.4673  0.5457  15  300
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Panel unit root test: Summary  Series: D(LGDP) Date: 08/11/18 Time: 16:14 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 4
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -8.29686  0.0000  15  277
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat  -10.1800  0.0000  15  277
ADF – Fisher Chi-square  151.787  0.0000  15  277
PP – Fisher Chi-square  182.115  0.0000  15  285
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Panel unit root test: Summary  Series: ECO Date: 08/22/18 Time: 03:25 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 4
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -2.27893  0.0113  15  293
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat  -5.28602  0.0000  15  293
ADF – Fisher Chi-square  87.0778  0.0000  15  293
PP – Fisher Chi-square  107.403  0.0000  15  300
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Panel unit root test: Summary  Series: POL Date: 08/22/18 Time: 03:29 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 4
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -3.11586  0.0009  15  287
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat  -3.39338  0.0003  15  287
ADF – Fisher Chi-square  70.4734  0.0000  15  287
PP – Fisher Chi-square  73.0552  0.0000  15  300
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Panel unit root test: Summary  Series: INS Date: 08/22/18 Time: 03:33 Sample: 1996 2016 Exogenous variables: Individual effects
Automatic selection of maximum lags Automatic lag length selection based on SIC: 0 to 4
Newey-West automatic bandwidth selection and Bartlett kernel
Cross- Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process) 
Levin, Lin & Chu t* -1.93248  0.0267  15  295
Null: Unit root (assumes individual unit root process) 
Im, Pesaran and Shin W-stat  -4.42176  0.0000  15  295
ADF – Fisher Chi-square  76.6938  0.0000  15  295
PP – Fisher Chi-square  77.1543  0.0000  15  300
** Probabilities for Fisher tests are computed using an asymptotic Chi
        -square distribution. All other tests assume asymptotic normality.

Co-integration test
Pedroni Residual Cointegration Test Series: LGDP POL INS GOV ECO  Date: 08/22/18 Time: 03:42 Sample: 1996 2016 Included observations: 315 Cross-sections included: 15 Null Hypothesis: No cointegration Trend assumption: No deterministic trend Automatic lag length selection based on SIC with a max lag of 3
Newey-West automatic bandwidth selection and Bartlett kernel
Alternative hypothesis: common AR coefs. (within-dimension)
Weighted Statistic Prob. Statistic Prob.

Panel v-Statistic -2.948185  0.9984 -2.612984  0.9955
Panel rho-Statistic  2.437032  0.9926  2.188641  0.9857
Panel PP-Statistic  1.111001  0.8667  0.583068  0.7201
Panel ADF-Statistic  1.792682  0.9635  1.173057  0.8796
Alternative hypothesis: individual AR coefs. (between-dimension)
Statistic Prob. Group rho-Statistic  3.618577  0.9999 Group PP-Statistic  1.079420  0.8598 Group ADF-Statistic  1.459869  0.9278 Cross section specific results Phillips-Peron results (non-parametric) Cross ID AR(1) Variance HAC   Bandwidth Obs
 1 0.633 0.009976 0.009565 2.00 20
 2 0.667 0.022233 0.022233 0.00 20
 3 -0.081 0.001767 0.002000 1.00 20
 4 0.061 0.004077 0.002386 4.00 20
 5 0.340 0.004857 0.004857 0.00 20
 6 0.788 0.014399 0.012644 1.00 20
 7 0.205 0.006880 0.006880 0.00 20
 8 0.247 0.006866 0.006275 2.00 20
 9 0.387 0.011687 0.015154 1.00 20
 10 0.458 0.016854 0.017684 2.00 20
 11 0.357 0.021632 0.024943 2.00 20
 12 0.563 0.046360 0.062550 2.00 20
 13 0.564 0.009897 0.009463 1.00 20
 14 0.502 0.015997 0.015997 0.00 20
 15 0.531 0.009528 0.009528 0.00 20
Augmented Dickey-Fuller results (parametric) Cross ID AR(1) Variance Lag Max lag Obs
 1 0.633 0.009976 0 3 20
 2 0.667 0.022233 0 3 20
 3 -0.081 0.001767 0 3 20
 4 -0.438 0.003495 1 3 19
 5 0.340 0.004857 0 3 20
 6 0.788 0.014399 0 3 20
 7 0.205 0.006880 0 3 20
 8 0.247 0.006866 0 3 20
 9 0.402 0.007073 1 3 19
 10 0.458 0.016854 0 3 20
 11 0.391 0.016381 2 3 18
 12 0.563 0.046360 0 3 20
 13 0.564 0.009897 0 3 20
 14 0.502 0.015997 0 3 20
 15 0.531 0.009528 0 3 20

Test for selecting the number of lags
VAR Lag Order Selection Criteria Endogenous variables: D(LGDP) D(ECO) D(POL) D(INS)  Exogenous variables: C  Date: 08/22/18 Time: 04:12 Sample: 1996 2016 Included observations: 180  Lag LogL LR FPE AIC SC HQ
0  798.1645 NA   1.73e-09 -8.824050  -8.753095* -8.795281
1  824.8160  51.82233   1.54e-09*  -8.942400* -8.587627  -8.798555*
2  837.1805  23.49262  1.60e-09 -8.902006 -8.263414 -8.643084
3  853.2954  29.90210  1.60e-09 -8.903282 -7.980873 -8.529285
4  858.0016  8.523398  1.82e-09 -8.777795 -7.571567 -8.288722
5  868.2349  18.07876  1.94e-09 -8.713721 -7.223674 -8.109571
6  885.4517   29.65119*  1.92e-09 -8.727241 -6.953376 -8.008015
7  897.7014  20.55234  2.01e-09 -8.685571 -6.627888 -7.851269
8  905.4571  12.66767  2.21e-09 -8.593968 -6.252466 -7.644590
 * indicates lag order selected by the criterion  LR: sequential modified LR test statistic (each test at 5% level)  FPE: Final prediction error  AIC: Akaike information criterion  SC: Schwarz information criterion  HQ: Hannan-Quinn information criterion
Appendix C
VAR Result
 Vector Autoregression Estimates  Date: 08/22/18 Time: 03:57  Sample (adjusted): 1999 2016  Included observations: 270 after adjustments  Standard errors in ( ) & t-statistics in D(LGDP) D(ECO) D(POL) D(INS)
D(LGDP(-1))  0.130641  0.341198 -0.399353  0.120975
 (0.05544)  (0.16707)  (0.24387)  (0.17631)
2.35654 2.04227 -1.63757 0.68615
D(LGDP(-2))  0.123511  0.042411 -0.230054  0.181352
 (0.04543)  (0.13690)  (0.19984)  (0.14447)
2.71883 0.30979 -1.15121 1.25525
D(ECO(-1)) -0.019452 -0.505749 -0.637204 -0.364232
 (0.03215)  (0.09687)  (0.14141)  (0.10223)
-0.60512 -5.22076 -4.50621 -3.56284
D(ECO(-2)) -0.024710 -0.008973 -0.212251  0.044411
 (0.02953)  (0.08898)  (0.12989)  (0.09390)
-0.83685 -0.10084 -1.63409 0.47293
D(POL(-1))  0.042174 -0.025827 -0.199231  0.045972
 (0.01570)  (0.04731)  (0.06906)  (0.04993)
2.68628 -0.54587 -2.88477 0.92074
D(POL(-2))  0.006540  0.093953  0.046732  0.047799
 (0.01507)  (0.04542)  (0.06630)  (0.04793)
0.43399 2.06869 0.70491 0.99729
D(INS(-1)) -0.012252  0.106503  0.344147 -0.088730
 (0.03048)  (0.09186)  (0.13409)  (0.09694)
-0.40194 1.15940 2.56654 -0.91529
D(INS(-2))  0.006576  0.080994  0.303153  0.080874
 (0.02891)  (0.08714)  (0.12720)  (0.09196)
0.22741 0.92949 2.38336 0.87947
C  0.031770 -0.026807  0.032925 -0.017050
 (0.00409)  (0.01231)  (0.01797)  (0.01299)
7.77638 -2.17728 1.83201 -1.31226
 R-squared  0.103450  0.510710  0.386900  0.448572
 Adj. R-squared  0.075969  0.495713  0.368108  0.431670
 Sum sq. resids  0.608929  5.530213  11.78345  6.158880
 S.E. equation  0.048302  0.145563  0.212479  0.153614
 F-statistic  3.764488  34.05328  20.58819  26.53954
 Log likelihood  439.6409  141.7930  39.66957  127.2578
 Akaike AIC -3.189932 -0.983652 -0.227182 -0.875983
 Schwarz SC -3.069985 -0.863704 -0.107235 -0.756036
 Mean dependent  0.043593 -0.000664  0.014985  0.004802
 S.D. dependent  0.050248  0.204980  0.267297  0.203765
 Determinant resid covariance (dof adj.)  1.35E-08  Determinant resid covariance  1.18E-08  Log likelihood  932.3588  Akaike information criterion -6.639695  Schwarz criterion -6.159905
Dependent Variable: D(LGDP) Method: Panel Least Squares Date: 08/22/18 Time: 04:07 Sample (adjusted): 1999 2016 Periods included: 18 Cross-sections included: 15 Total panel (balanced) observations: 270 D(LGDP) = C(1)*D(LGDP(-1)) + C(2)*D(LGDP(-2)) + C(3)*D(ECO(-1)) +
        C(4)*D(ECO(-2)) + C(5)*D(POL(-1)) + C(6)*D(POL(-2)) + C(7)*D(INS(
        -1)) + C(8)*D(INS(-2)) + C(9) Coefficient Std. Error t-Statistic Prob.  
C(1) 0.130641 0.055438 2.356544 0.0192
C(2) 0.123511 0.045428 2.718829 0.0070
C(3) -0.019452 0.032145 -0.605124 0.5456
C(4) -0.024710 0.029527 -0.836845 0.4034
C(5) 0.042174 0.015700 2.686279 0.0077
C(6) 0.006540 0.015071 0.433985 0.6647
C(7) -0.012252 0.030482 -0.401936 0.6881
C(8) 0.006576 0.028915 0.227412 0.8203
C(9) 0.031770 0.004085 7.776377 0.0000
R-squared 0.103450     Mean dependent var 0.043593
Adjusted R-squared 0.075969     S.D. dependent var 0.050248
S.E. of regression 0.048302     Akaike info criterion -3.189932
Sum squared resid 0.608929     Schwarz criterion -3.069985
Log likelihood 439.6409     Hannan-Quinn criter. -3.141767
F-statistic 3.764488     Durbin-Watson stat 2.108795
Prob(F-statistic) 0.000340
Wald Test: Equation: Untitled Test Statistic Value df Probability
F-statistic  0.453208 (2, 261)  0.6361
Chi-square  0.906416  2  0.6356
Null Hypothesis: C(3)=C(4)=0 Null Hypothesis Summary: Normalized Restriction (= 0) Value Std. Err.

C(3) -0.019452  0.032145
C(4) -0.024710  0.029527
Restrictions are linear in coefficients.

Wald Test: Equation: Untitled Test Statistic Value df Probability
F-statistic  3.618707 (2, 261)  0.0282
Chi-square  7.237414  2  0.0268
Null Hypothesis: C(5)=C(6)=0 Null Hypothesis Summary: Normalized Restriction (= 0) Value Std. Err.

C(5)  0.042174  0.015700
C(6)  0.006540  0.015071
Restrictions are linear in coefficients.

Wald Test: Equation: Untitled Test Statistic Value df Probability
F-statistic  0.130563 (2, 261)  0.8777
Chi-square  0.261126  2  0.8776
Null Hypothesis: C(7)=C(8)=0 Null Hypothesis Summary: Normalized Restriction (= 0) Value Std. Err.

C(7) -0.012252  0.030482
C(8)  0.006576  0.028915
Restrictions are linear in coefficients.

Dependent Variable: D(ECO) Method: Panel Least Squares Date: 08/22/18 Time: 05:15 Sample (adjusted): 1999 2016 Periods included: 18 Cross-sections included: 15 Total panel (balanced) observations: 270 D(ECO) = C(10)*D(LGDP(-1)) + C(11)*D(LGDP(-2)) + C(12)*D(ECO(-1)) +
        C(13)*D(ECO(-2)) + C(14)*D(POL(-1)) + C(15)*D(POL(-2)) + C(16)
        *D(INS(-1)) + C(17)*D(INS(-2)) + C(18)
Coefficient Std. Error t-Statistic Prob.  
C(10) 0.341198 0.167068 2.042272 0.0421
C(11) 0.042411 0.136903 0.309788 0.7570
C(12) -0.505749 0.096873 -5.220760 0.0000
C(13) -0.008973 0.088983 -0.100839 0.9198
C(14) -0.025827 0.047313 -0.545873 0.5856
C(15) 0.093953 0.045417 2.068687 0.0396
C(16) 0.106503 0.091861 1.159399 0.2474
C(17) 0.080994 0.087138 0.929487 0.3535
C(18) -0.026807 0.012312 -2.177282 0.0304
R-squared 0.510710     Mean dependent var -0.000664
Adjusted R-squared 0.495713     S.D. dependent var 0.204980
S.E. of regression 0.145563     Akaike info criterion -0.983652
Sum squared resid 5.530213     Schwarz criterion -0.863704
Log likelihood 141.7930     Hannan-Quinn criter. -0.935486
F-statistic 34.05328     Durbin-Watson stat 1.694825
Prob(F-statistic) 0.000000
Wald Test: Equation: Untitled Test Statistic Value df Probability
F-statistic  2.478145 (2, 261)  0.0859
Chi-square  4.956290  2  0.0839
Null Hypothesis: C(10)=C(11)=0 Null Hypothesis Summary: Normalized Restriction (= 0) Value Std. Err.

C(10)  0.341198  0.167068
C(11)  0.042411  0.136903
Restrictions are linear in coefficients.

Dependent Variable: D(POL) Method: Panel Least Squares Date: 08/22/18 Time: 05:43 Sample (adjusted): 1999 2016 Periods included: 18 Cross-sections included: 15 Total panel (balanced) observations: 270 D(POL) = C(19)*D(LGDP(-1)) + C(20)*D(LGDP(-2)) + C(21)*D(ECO(-1)) +
        C(22)*D(ECO(-2)) + C(23)*D(POL(-1)) + C(24)*D(POL(-2)) + C(25)
        *D(INS(-1)) + C(26)*D(INS(-2)) + C(27)
Coefficient Std. Error t-Statistic Prob.  
C(19) -0.399353 0.243870 -1.637566 0.1027
C(20) -0.230054 0.199838 -1.151206 0.2507
C(21) -0.637204 0.141406 -4.506211 0.0000
C(22) -0.212251 0.129889 -1.634088 0.1034
C(23) -0.199231 0.069063 -2.884775 0.0042
C(24) 0.046732 0.066295 0.704910 0.4815
C(25) 0.344147 0.134090 2.566539 0.0108
C(26) 0.303153 0.127196 2.383357 0.0179
C(27) 0.032925 0.017972 1.832010 0.0681
R-squared 0.386900     Mean dependent var 0.014985
Adjusted R-squared 0.368108     S.D. dependent var 0.267297
S.E. of regression 0.212479     Akaike info criterion -0.227182
Sum squared resid 11.78345     Schwarz criterion -0.107235
Log likelihood 39.66957     Hannan-Quinn criter. -0.179016
F-statistic 20.58819     Durbin-Watson stat 1.728673
Prob(F-statistic) 0.000000
Wald Test: Equation: Untitled Test Statistic Value df Probability
F-statistic  2.697632 (2, 261)  0.0692
Chi-square  5.395264  2  0.0674
Null Hypothesis: C(19)=C(20)=0 Null Hypothesis Summary: Normalized Restriction (= 0) Value Std. Err.

C(19) -0.399353  0.243870
C(20) -0.230054  0.199838
Restrictions are linear in coefficients.

Dependent Variable: D(INS) Method: Panel Least Squares Date: 08/22/18 Time: 06:02 Sample (adjusted): 1999 2016 Periods included: 18 Cross-sections included: 15 Total panel (balanced) observations: 270 D(INS) = C(28)*D(LGDP(-1)) + C(29)*D(LGDP(-2)) + C(30)*D(ECO(-1)) +
        C(31)*D(ECO(-2)) + C(32)*D(POL(-1)) + C(33)*D(POL(-2)) + C(34)
        *D(INS(-1)) + C(35)*D(INS(-2)) + C(36)
Coefficient Std. Error t-Statistic Prob.  
C(28) 0.120975 0.176308 0.686154 0.4932
C(29) 0.181352 0.144475 1.255247 0.2105
C(30) -0.364232 0.102231 -3.562840 0.0004
C(31) 0.044411 0.093905 0.472933 0.6367
C(32) 0.045972 0.049930 0.920740 0.3580
C(33) 0.047799 0.047929 0.997285 0.3195
C(34) -0.088730 0.096942 -0.915292 0.3609
C(35) 0.080874 0.091958 0.879467 0.3800
C(36) -0.017050 0.012993 -1.312262 0.1906
R-squared 0.448572     Mean dependent var 0.004802
Adjusted R-squared 0.431670     S.D. dependent var 0.203765
S.E. of regression 0.153614     Akaike info criterion -0.875983
Sum squared resid 6.158880     Schwarz criterion -0.756036
Log likelihood 127.2578     Hannan-Quinn criter. -0.827818
F-statistic 26.53954     Durbin-Watson stat 1.755360
Prob(F-statistic) 0.000000
Wald Test: System: sys01 Test Statistic Value df Probability
Chi-square  11.27789  2  0.0036
Null Hypothesis: C(3)=C(4)=0 Null Hypothesis Summary: Normalized Restriction (= 0) Value Std. Err.

C(3)  0.116578  0.036555
C(4) -0.106533  0.035869
Restrictions are linear in coefficients.

Causality Test
VAR Granger Causality/Block Exogeneity Wald Tests
Date: 08/23/18 Time: 10:16 Sample: 1996 2016 Included observations: 270 Dependent variable: D(LGDP) Excluded Chi-sq df Prob.

D(ECO)  0.906416 2  0.6356
D(POL)  7.237414 2  0.0268
D(INS)  0.261126 2  0.8776
All  12.66152 6  0.0487
Dependent variable: D(ECO) Excluded Chi-sq df Prob.

D(LGDP)  4.956290 2  0.0839
D(POL)  5.306689 2  0.0704
D(INS)  1.846800 2  0.3972
All  15.53345 6  0.0165
Dependent variable: D(POL) Excluded Chi-sq df Prob.

D(LGDP)  5.395264 2  0.0674
D(ECO)  20.92996 2  0.0000
D(INS)  10.20494 2  0.0061
All  39.14364 6  0.0000
Dependent variable: D(INS) Excluded Chi-sq df Prob.

D(LGDP)  2.697270 2  0.2596
D(ECO)  14.07043 2  0.0009
D(POL)  1.518095 2  0.4681
All  19.15403 6  0.0039
Pairwise Dumitrescu Hurlin Panel Causality Tests
Date: 08/23/18 Time: 11:56
Sample: 1996 2016 Lags: 2  Null Hypothesis: W-Stat. Zbar-Stat. Prob. 
 POL does not homogeneously cause LGDP  5.47220  4.40329 1.E-05
 LGDP does not homogeneously cause POL  3.06879  1.03172 0.3022
 INS does not homogeneously cause LGDP  4.04943  2.40739 0.0161
 LGDP does not homogeneously cause INS  4.46812  2.99474 0.0027
 ECO does not homogeneously cause LGDP  4.03327  2.38473 0.0171
 LGDP does not homogeneously cause ECO  2.42999  0.13559 0.8921
 INS does not homogeneously cause POL  3.73020  1.95957 0.0500
 POL does not homogeneously cause INS  5.74133  4.78084 2.E-06
 ECO does not homogeneously cause POL  5.86761  4.95799 7.E-07
 POL does not homogeneously cause ECO  4.11351  2.49728 0.0125
 ECO does not homogeneously cause INS  3.11908  1.10228 0.2703
 INS does not homogeneously cause ECO  2.00084 -0.46644 0.6409