& D e ve l o p m e n t

Transcription

1 Journal of Economic Cooperation and Development June 2016 Journal of Economic Cooperation & D e ve l o p m e n t Atif Awad and Ishak Yussof Democracy and Human Development Nexus: The African Experience Dr. Abdel Hakeem Ahmad Eltalla The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model Reza Ghazal and Muhamed Zulkhibri Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Ebaidalla Mahjoub Ebaidalla Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Mansour Zarra-Nezhad, Majid Sheikh Ansari and Mahvash Moradi De ter minants of Tax Re venue: Does Liberalization Boost or Decline it? EL Mostafa Bentour On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco V O LU M E 37, No. 2 Kudüs Cad. No:9 Diplomatik Site ORAN-Ankara, Turkey Tel: (90-312) Fax: (90-312) oicankara@sesric.org Web: Volume 37, No. 2 S t a t i s t i c a l E c o n o m i c a n d S o c i a l R e s e a rch and Training Centre for Islamic Countries (SESRIC) CMYK J u n e 2016 CMYK STATISTICAL, ECONOMIC AND SOCIAL RESEARCH AND TRAINING CENTRE FOR ISLAMIC COUNTRIES ISSN

2 ISSN Volume 37, No.2, June 2016 J o u r n a l o f E c o n o m i c C o o p e r a t i o n & D e v e l o p m e n t Statistical Economic and Social Research and Training Centre for Islamic Countries (SESRIC)

3

4 Contents Atif Awad and Ishak Yussof Democracy and Human Development Nexus: The African Experience... 1 Dr. Abdel Hakeem Ahmad Eltalla The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model Reza Ghazal and Muhamed Zulkhibri Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Ebaidalla Mahjoub Ebaidalla Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Mansour Zarra-Nezhad, Majid Sheikh Ansari and Mahvash Moradi Determinants of Tax Revenue: Does Liberalization Boost or Decline it? EL Mostafa Bentour On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco

5

6 EDITORIAL NOTE I am pleased to inform our readers that we have selected six remarkable articles tackling issues of great importance to the OIC member countries in this June Issue of the Journal of Economic Cooperation and Development. The development of modern and sustainable economy has recently gained greater importance and attention in many developing countries as a major driver of social-economic progress. In this connection, this June issue sheds light on themes related to impacts on prices and policy implication, liberalization, youth unemployment, Islamic Inclusive Growth Index, economic impact of constructing a Gaza Seaport and democracy and human development nexus. The first article focuses on the use of data from 46 African countries over the period from 1990 to 2012 by examining three principal issues: 1) whether human development is affected by the level or the stock of democracy in these countries; and whether the affect varies over time. 2) whether a country s level of development and education level foster or impede the impact of democracy on human development. 3) whether a democratic regime helps to further improve the health of its population via redistribution mechanisms. The results seem to contain good news for African countries that inherited bad political institutions or systems from the earlier or colonial regimes. The second article focuses on the role of infrastructure as a catalyst of economic development. Among the different types of transport infrastructure, seaports are considered as especially strategic due to the increasing importance of maritime transport in deciding transport costs and international trade in the age of globalization. The construction of a Gaza Seaport is a computable general equilibrium model of the Palestinian economy to simulate the effects on the Palestinian economy. The third article develops an Islamic Inclusive Growth Index (i-igi) - a composite index of growth inclusiveness - for the Organization of Islamic Cooperation (OIC) of the member countries. It comprises of three subpillars to measure the countries performance based on three building blocks of inclusiveness. Despite having a good performance on the three pillars of inclusiveness, the OIC member countries in the Central Asia are lagging behind in terms of their conformity with Maqasid al-shari ah principles.

7 The fourth article paper examines the determinants of youth unemployment in OIC countries during the period The study used a dynamic panel data method for a sample of 32 OIC countries, focusing on the impact of economic, demographic and institutional factors. The empirical results show that youth unemployment in OIC countries is influenced by economic environment measured by GDP growth, inflation and domestic investment. Fertility rate and bureaucracy quality are found to be one of the significant factors behind the high rate of youth unemployment in OIC member countries. The fifth highlights that trade liberalization due to its improving role has been at the center of economies in recent years. To test the hypothesis whether trade liberalization boosts tax revenue, a panel of 83 countries, over , has been applied to perform Generalized Method of Moment regression. The results of dynamic panel estimation show more trade liberalization is accompanied by more tax revenue. Therefore, there is the need for appropriate macroeconomic policy to enhance the trade liberalization in order to accelerate government revenue. The sixth and last examines the input-output model by analysing the effect of removing energy subsidies on prices in Morocco. Three scenarios of increasing oil prices by 25%, 50% and 75% have set to show that the effects on inputs prices are high in intensive oil products sectors such as transports, manufacturing industries, fishing and aquaculture, and, electricity and water sectors. Amb. Musa KULAKLIKAYA Editor-in-chief

8 Journal of Economic Cooperation and Development, 37, 2 (2016), 1-34 Democracy and Human Development Nexus: The African Experience Atif Awad 1 and Ishak Yussof 2 Using data from 46 African countries over the period from 1990 to 2012, the present study examines three principal issues. First, the study examines whether human development is affected by the level or the stock of democracy in these countries; and whether the affect varies over time. Second, the study investigates whether a country s level of development and education level foster or impede the impact of democracy on human development. Third, the study examines whether a democratic regime helps to further improve the health of its population via redistribution mechanisms. The results of the Arellano-Bond (A-B) GMM technique show that democracy, irrespective of the measurement employed, has a positive impact on human development in both the long run and the short run (i.e., infant mortality rate and life expectancy). The results also show that human development is independent of the country s level of development and the education level of its population. Additionally, democratic regimes tend to devote a considerable portion of government resources to the health sector, which is likely to be reflected in further improvements in the well-being of a population via redistribution mechanisms. The results seem to contain good news for African countries that inherited bad political institutions or systems from the earlier or colonial regimes. This is because the results tell us that African countries may still have the ability to improve their population's health, even with their contemporary status of political institutions. 1. Introduction Since the end of 1990s, most African countries have experienced a wave of democratization (Kudamatsu 2012). The discussion among political scientists pertaining to African countries focuses upon the causes of 1 Department of Finance & Economics, College of Business Administration University of Sharjah, UAE. 2 School of Economics, Faculty of Economics and Management, Universiti Kebangsaan Malaysia.

9 2 Democracy and Human Development Nexus: The African Experience democratization; and whether new democracies in the region will be strengthened. Unfortunately, very few studies examine whether such political changes affect the lives of people on the continent. Improvements among political institutions do not always imply that similar improvements will occur in relation to the standard of living because no consensus exists in regards to whether democracy enhances economic development (Przeworski et al. 2000; Kurzman et al. 2002). In contrast, consensus does exist for the notion that democracy improves human development (Lipset 1959; Boix 2001; Lenski 1966; Muller 1988; Dreze and Sen 1989; Lake and Baum 2001; Ghobarah et al. 2004; Brown and Hunter 2004; Ghobarah et al. 2004; Kudamatsu 2006; Brown and Mobarak 2009). The logic of the argument is based upon the idea that widespread participation in government empowers ordinary citizens, including the very poor, and should therefore lead governments to be more accountable to the population governed (Vollmer and Ziegler 2009; Gerring et al. 2012). For a long period, most political economists believed that democracy was one of the efficient channels to improve human development (UNDP ; Gerring et al. 2012). Nevertheless, the view has been strongly challenged in recent years. Several studies maintain that no positive correlation exists between regime type and different measures of human development; or that these relationships are not particularly robust (Gauri and Khaleghian 2002; McGuire 2004; Shandra et al. 2004; Ross 2006). Surprisingly, such studies claim that even under the auspices of authoritarian rule, the level of human development remained high in some countries (i.e., East Asian countries). In contrast, some democratic societies in developing countries witnessed widespread of poverty and income inequality (i.e., India). Gerring et al. (2012) criticize the extant studies concerning the democracy human development nexus, claiming that such studies are based upon the fundamental assumption that a proximal relationship exists between the two variables. Subsequently, Gerring et al. (2012) introduce the possibility that the developmental effects of democracy might be long-term and characterized by a distal rather than proximal causal relationship due to the fact that new democracies and old democracies vary. While new democracies are prone to a host of problems associated with regime transition, older democracies are more institutionalized and generally enjoy higher-quality governance

10 Journal of Economic Cooperation and Development 3 (Kapstein and Converse 2008; Keefer 2006). To support the argument, Gerring et al. (2012) examine samples from 159 countries (developed and developing countries) over the period of 1960 to The results demonstrate that a country s contemporary level of democracy has only a weak association with the improvement of human development, while a country s historical experience with democracy has a strong and robust influence on human development. Using data from 46 African countries for the period , the present study examines three principal issues. First, the study examines whether human development is affected by the level or the stock of democracy in these countries; and whether the affect varies over time. Second, the study investigates whether a country s level of development; and whether education level fosters or impedes the impact of democracy on human development. Third, the study examines whether a democratic regime leads to further improvements in the health of its population via redistribution mechanisms. The results of the Arellano-Bond GMM (A-B GMM) technique show that democracy, irrespective of the measurement employed, has a positive impact on human development in both the long run and the short run (i.e., infant mortality rate, and life expectancy). The results also show that the positive impact of democracy, irrespective of the measurement employed, on human development is independent of a country s level of development and the education level of its population. Additionally, democratic regimes through tend to devote a considerable portion of a government resources to the health sector via redistribution mechanisms, which is likely to be reflected in further improvements in the well-being of a population. The remainder of the paper is organized as follows. Section 2 provides an overview of the causal pathways through which democracy can influence human development. Section 3 presents the data and methodology employed in the present study to perform the empirical estimation. Section 4 analyses the econometric results and discuss the findings. Section 5 summarizes the study and presents the conclusions of the present paper.

11 4 Democracy and Human Development Nexus: The African Experience 2. Democracy and Human Development Three principal theories highlight the importance of political institutions in increasing the living standards of citizens. Two of the theories are associated with Sen (1981, 1999), whose works concerning the causes of famine extend to addressing the causes of poverty more generally (Rose 2006). The first theory argues that democracies, through the electoral processes, allow the poor to penalize governments that allow famines to occur and result in political leaders actively attempting to avoid famines (Rose 2006: 23). The second argument put forward by Sen is democracies are better than non-democracies in facilitating the transmission of information from poor and distant areas to the central government due to the freedom of the press (Ross 2006). A third theory suggests that a democratic regime tends to help the poor more than nondemocracies by creating more public goods and improving the quality of the redistribution of income because the electoral processes present in democracies force politicians to spend government revenues on government services, while autocratic governments face no such constraints (Deacon 2003; Lake and Baum 2001; McGuire and Olson 1996; Niskanen 1997). In other words, democratic governments have a wider range of supporters to satisfy, which encourages democratic regimes to produce public goods instead of private ones (Bueno de Mesquita et al. 2003; Ghorbarah et al. 2004). The focus of the present study is to examine policies with redistributive characteristics that are intended to promote better health for the population as a whole. In accordance with the third theory, democratic regimes can affect human development through policies on the protection of property rights and redistribution (e.g., Acemoglu and Robinson 2005; Boix 2003; Bueno de Mesquita et al. 2003; Ghorbarah et al. 2004; McGuire and Olson 1996; Meltzer and Richard 1981; Niskanen 1997) 3. Nonetheless, matching societal and individual needs with a sufficient redistribution system and suitable public provision of goods and services provides a more direct link between political institutions and human development than property rights protection 3 Vollmer and Ziegler (2009) discuss the possibility of trade-offs between policies for the protection of property rights and/or policies for redistribution. Redistribution refers to the effect of political systems on income inequality or on the provision of public goods; and the size of the public sector (see Boix 2001; Gradstein and Milanovic 2004; Persson et al. 2002, Stasavage 2005a; Persson andtabellini. 2000).

12 Journal of Economic Cooperation and Development 5 (Vollmer and Ziegler 2009). Policies for redistribution have an equalizing influence on the distribution of wealth in a society. Such policies are particularly suitable in conjunction with broad-based programs and the provision of public goods and services, in which redistribution policies can offset market failures and normative social optimum levels can be achieved. The redistribution of wealth from the rich to the poor, and vice versa, exists in both autocratic and democratic systems. However, Meltzer and Richard (1981) argue that income redistribution from the rich to the poor is more pronounced and at a higher level in democracies (Avelino et al. 2005; Brown and Hunter 2004; Gerring et al. 2005; Kaufman and Segura-Ubiergo 2001; McGuire 2006; Stasavage 2005a; Tavares and Wacziarg 2001; Gradstein and Milanovic 2004). According to the model by Meltzer and Richard (1981), the median voter in a democratic government is the decisive voter. The more the income of the median voter falls short of the average income of all voters, the higher the tax rate becomes due to their pressure for redistribution. Consequently, if the majority of the voting public lives at the bottom of the income distribution and only a small part enjoys or controls the wealth, government spending will increase and social services will become more extensive in democratic regimes (Keefer and Khemani 2005). In contrast, the distribution of wealth does not play a decisive role in authoritarian systems. All or a considerable part of the electorate is left out of the decision-making process, so as to avoid the redistributive consequences of democracy. As a result, the magnitude of the public sector on average remains small (Boix 2001). 4 Autocrats do not implement redistributive policies because of institutional structures, but due to ideological causes; or only to levels that assist them to remain in power and to increase their personal wealth (Vollmer and Ziegler 2009). However, even in democratic governments, voting alone does not help to overcome aggregation problems stemming from the vast differences in individual preferences. The existence of vast differences in individual preferences raises the issue of whether democratic regimes are more responsive to the needs of citizens when compared against citizens of 4 Examples of autocracies with relatively large public sectors include Cuba and Venezuela (Gerring et al. 2012).

13 6 Democracy and Human Development Nexus: The African Experience autocratic regimes. According to Sen (1999a, 1999b), democracy, alongside its intrinsic value, is of distinguished importance for the development process because of the constructive and instrumental role it plays in the construction and aggregation of values, needs and preferences, and their translation into well-planned policies to benefit the society. For instance, the expansion of civil liberties that constitute parts of a democratic regime is likely to facilitate the formation of preferences and values, as well as access to appropriate information. Accordingly, democratic mechanisms will then enable the transmission of the needs of the citizens into the political arena where decisionmaking power is distributed among the legitimate representatives of the society as a whole (Vollmer and Ziegler 2009). Nevertheless, the existence of a democratic regime does not indicate an efficient redistribution or allocation of public goods and services to satisfy the societal needs (Vollmer and Ziegler 2009). Some factors can either hamper or foster the performance of democracy in relation to the satisfaction of societal needs. 5 For instance, the term 'redistribution' implies the presence of resources to be distributed in the form of public goods. The interaction between such factors and democracy at one point of time influences the output (i.e., policies through the provision of public goods) and the outcome (i.e., the level of human development). The positive effect of democracies on the provision of public goods and, thereby, human development is likely to strengthened by the level of economic development, as argued by the third theory presented above. In addition, the inadequacy of the information available to the voters may lead to insufficient participation, which is essential for the expression of public opinion and social needs. Consequently, the quality of responsiveness of a government results in a decrease in the unaddressed needs and demands of the society. Moreover, since accountability suffers due to information restrictions, voters are unable to control the behavior of politicians (Vollmer and Ziegler 2009). Education is one of the key factors with the potential to eliminate or minimize problems associated with restricted information. Education, in 5 Many factors (e.g., cultural; ethnic fragmentation; media; the degree of income distribution) are likely to hamper or foster the performance of democracy concerning human development (Vollmer and Ziegler 2009). However, the present study focuses on factors that represent the level of development and education because data or proxies for others factors are not available for many countries in Africa and/or over a long period of historical time.

14 Journal of Economic Cooperation and Development 7 the present context, is not an intrinsic component of human development, but a means to human development (Vollmer and Ziegler 2009). Although, education alone may influence the quality of the democracy, the former may also influence the latter in terms of a more efficient and effective provision of public goods (Lipset 1959; Glaeser et al. 2007; Keefer and Khemani 2005). Few studies empirically investigate the links between political systems and measures for non-income dimensions of human development. 6 While some studies find a positive relationship between democracy and human development (e.g., Besley and Kudamatsu 2006; Franco et al. 2004; Tsai 2006; Vollmer and Ziegler 2009), others studies find less evidence supporting the existence of a positive relationship between democracy and human development (Gauri and Khaleghian 2002; McGuire 2004; Shandra et al. 2004; Ross 2006). The existence of the divergent results regarding the relationship between democracy and human development has been attributed, for instance, to the fact that while democracies spend more money on education and health than nondemocracies, the benefits of such spending are primarily enjoyed by middle and upper income groups (Ross 2006). The research efforts referred to thus far are either limited to the sub-sample of developing countries (e.g., Tsai 2006; Vollmer and Ziegler 2009); to only one of the non-income dimensions of human development (e.g., Besley and Kudamatsu 2006; Franco et al. 2004; Ross 2006; Gerring et al. 2012); or to a cross-sectional focus leaving out developments over time (e.g., Tsai 2006; Franco et al. 2004). Moreover, the investigations, while having in mind potential conditions that influence the performance of a democracy, only include selected factors as control variables in the regression models employed in the respective models. In relation to the African continent, and to the best of the knowledge of the authors, no studies employ panel-time series data for a large number of countries and employ A-B GMM methods to evaluate the effect of democracy (not the outcomes of the democracy) at current level or stock on nonincome dimensions of human development. Nonetheless, with growing 6 Numerous studies examine human development and utilize proxies used to measure institutional qualities. Due to the number of such studies, the findings of all relevant studies cannot be summarized here. Therefore, the present study focuses upon political institutions (i.e., democratic and non-democratic) and their relationship with human development.

15 8 Democracy and Human Development Nexus: The African Experience importance of the human development issues (e.g., the United Nations Millennium Development Goals (MDGs) emphasize human development), few studies examine certain proxies that represent the quality of the governance and its effect on the human development. Several studies examine issues concerning democracy and human rights in Africa. For instance, Baliamoune-Lutz and Boko (2012) examine the possibility that trade openness (i.e., openness) serves as a conduit through which political institutions, such as political rights, civil liberties and the rule of law (i.e., the outcomes of democracy), affect human development (i.e., life expectancy and literacy rates) through trade (i.e., openness). The study examined of a large group of African countries during the period of 1975 to 2001 and finds that trade and institutions exert little influence on human development in the form of literacy. Additionally, the study finds that income is the primary determinant of human development when measured according to literacy and life expectancy rates. Similarly, using retrospective fertility surveys conducted in 28 African countries, Kudamatsu (2012) examine whether democracy helps babies survive in sub-saharan Africa. 7 The study compares the survival of infants born to the same mother before and after democratization to disentangle the effect of democracy from that of changes in population characteristics. The study finds that the infant mortality fell by 1.2 percentage points, which is roughly 12% of the sample mean, after democratization during the post-cold War period. Nevertheless, due to the lack of appropriate data, the author fails to provide conclusive evidence concerning the mechanism through which democratization reduces infant mortality. The present study seeks to fill the gap in extant empirical literature concerning the influence of democracy on non-income dimensions of human development, particularly in relation to the African continent, by incorporating the theoretical explanations of Vollmer and Ziegler (2009) and Gerring et al. (2012). First, although Vollmer and Ziegler (2009) 7 In the study, democracy refers to a country satisfying the following two conditions: (1) the chief executive of the government has been elected as a result of multiparty elections with universal suffrage, without subsequently banning opposition parties; and (2) a new chief executive assumes office by winning multiparty elections.

16 Journal of Economic Cooperation and Development 9 make reasonable theoretical justifications concerning the inclusion of several other factors in the analysis (i.e., education, distribution of income, ethnic fermentation), the study ignores the possibility that the relationship between democracy and human development may be a historical phenomenon, as suggested by Gerring et al.(2012). In contrast, while Gerring et al. (2012) examine the relationship between both the level and stock of democracy on human development, the incorporation of the selected variables (i.e., urbanization and instability) occurs without any clear theoretical justification. However, while Vollmer and Ziegler (2009) advocate the incorporation of variables that represent government expenditure on education and health in the process of investigating the impact of democracy on human development (i.e., to capture the redistribution mechanism of the democracy), the authors fail to capture such effect due to the lack of sufficient data on such expenditures. Second, Vollmer and Ziegler (2009) examine the possibility that the effect of democracy on human development may depend on a country s level of development or the education level of its population, but Gerring et al. (2012) ignore this possibility altogether. Most importantly, common problems exist in these two studies that may affect the outcomes or at least make the robustness of the analyses questionable. First, in both of the studies, the data is analyzed using a fixed effect approach, which is widely criticized due to its inability to resolve issues arising due to endogeneity and the omitted variables problems (Baliamoune-Lutza and Bokoc 2012). Second, both studies employ only one measure for human development and the results may not be sufficiently robust when an alternative measure for human development is introduced. The robustness of the results is of particular importance since some extant studies indicate that the relationship between democracy and human development is not robust (e.g., Gauri and Khaleghian 2002; McGuire 2004; Shandra et al. 2004; Ross 2006), as mentioned earlier. In the present study, the combination of the theoretical framework by Vollmer and Ziegler (2009) and Gerring et al. (2012), in conjunction with employment of a more appropriate estimation method (i.e., A-B GMM), are likely to be sufficiently superior to overcome the aforementioned shortcomings in previous studies. Most importantly, the comments of Vollmer and Ziegler (2009) are addressed in the present study and government expenditures on health variables are considered in

17 10 Democracy and Human Development Nexus: The African Experience the analysis of the present study, which enables the examination of the impact of democracy on human development through redistribution mechanisms. Additionally, to ensure robustness, two measures of human development are employed: life expectancy and infant mortality rates. In the present study, the selection of both African countries and the period after 1990 is justifiable and is expected to have greater policy implications. As mentioned previously, the countries of the continent experienced remarkable improvement in their political institutions after 1990 (i.e., after the cold war). Nevertheless, no existing study addresses whether the level of human development in countries on the continent is affected by the contemporary status of political institutions; or whether the countries of the continent require a longer period of time for the current political institutions to have a measurable effect on the wellbeing of their populations. Extant studies propose that democracy and authoritarianism construct deep legacies, extending back several decades, perhaps even centuries (Collier and Collier 1991; Hite and Cesarini 2004). Further, in the framework of democratic regimes, whether the strength of the impact of such regimes on human development depends on a country s level of development and the education level of its population is examined. In regards to policy implications, if the results show that the contemporary status of democracy exerts a significant influence over human development, the finding will imply that the current efforts to improve political institutions in the African region will manifest in similar improvements in human development. Additionally, the results also imply that new emerging countries in Africa, such as Southern Sudan, are able to improve the well-being of their populations by improving their political institutions. However, if the results show that human development on the African continent is more greatly affected by the accumulation effect of democracy, the finding will imply that countries with a bad historical record regarding the quality of their political institutions will suffer more as a result of improving the welfare of the population. Furthermore, if the results show that the impact of democracy on human development is independent of a country s level of development, the finding will imply that it is possible for poor and democratic governments on the African continent to continue to enhance the well-being of their populations. Otherwise, the current efforts to improve political institutions on the African continent must be

18 Journal of Economic Cooperation and Development 11 undertaken in conjunction with economic reform to ensure better human development for the populations affected. 3. Variables, Data and Methodology 3.1 Variables and Data Following Vollmer and Ziegler (2009) and Gerring et al. (2012), the present study employs Polity 2 as indicators for the degree of democratization of the institutions in each country. Polity 2 measures the extent to which democratic or authoritarian political systems are institutionalized in a given country by taking into account how the executive is selected, the degree of checks on executive power, and the form of political competition. The Polity 2 score ranges from 10 (highly democratic) to minus 10 (highly autocratic), while a zero score indicates a state between autocracy and democracy. The data are gathered from the Polity IV data set (Marshall and Jaggers 2013). For the democracy level, the score a country receives on the Polity 2 for a given year is utilized. For the stock of democracy, the approach of Gerring et al. (2012) is followed and the sum of each country s score from 1990 to the 2012 is computed while applying a 1% annual depreciation rate. The manner in which the stock of democracy is computed allows for the years that are more distant to be weighted less than recent years while allowing for a country s regime stock to be analyzed over a period of two decades. The expectation is that the causal effect of democracy, like other capital stocks, depreciates over time (Gerring et al. 2012). For non-income dimensions of human development, infant mortality rate (per 1000) and life expectancy at birth (in years) are employed as common and conventional measure. Other variables that are expected to influence human development and to describe the possible conditions under which democracy affects human development include GDP per capita (measure in US$ 2005) as the proxy for level of economic development. Since education is also a factor that influences the performance of democracy, primary education enrolments are also used as an explanatory variable in the panel analysis (Vollmer and Ziegler 2009). To examine the possibility that democratic regimes can affect human development through the provision of public goods (i.e., the

19 12 Democracy and Human Development Nexus: The African Experience redistribution mechanism), government expenditure on health is incorporated in the analysis. Ranis et al. (2000) states that three important ratios should be considered when linking government expenditure to human development: the proportion of GDP spending by various levels of government; the proportion of government expenditure devoted to human development investments; and the proportion of spending on human development that is allocated according to priorities. The 1991 Human Development Report (HDR) suggests that good human development can be achieved when government expenditure accounts for approximately 25% of the GDP, of which 40% (or more) is allocated to social spending and more than 50% of this is spent on social priorities (UNDP 1991). Generally, increases in public expenditures for health can be decomposed into two components: an increase due to higher total expenditures; and an increase due to different priorities in government spending. While the first source is primarily driven by economic growth, democratic regimes are expected to be the main driver of the second source (Vollmer and Ziegler 2009). Thus, in the present study, such expenditures are measured as the proportion of total government expenditure (see Table 1 for more details). Data for the present study are gathered from 46 African countries over the period of 1990 to 2012 (the list of countries is reported in Table A1 in the Appendix). Data concerning human development measurements, education, health expenditure and per capita GDP are obtained from the World Bank Development Indicator Database (WBDI). According to Gerring et al. (2012), data from the WBDI is more reliable because it has broader country coverage and is less vulnerable to sample biases. A few missing data for some right-hand side variables are estimated using both straight-line interpolation and extrapolation methods. Table 1 shows the main descriptive statistics of the data. Meanwhile, Table 2 shows the statistically significant, but relatively weak, correlations between the democracy level and the stock of democracy; and the remaining variables. Per capita GDP is the only factor that is relatively significant and has a high correlation with the human development variables.

20 Journal of Economic Cooperation and Development 13 Table 1: Descriptive Statistics Variable Obs Mean Std. Dev. Min Max Ln GDP Infant mortality rate (ln) Life expectancy at birth(ln) Democracy, Level Democracy, Stock Primary education Government expenditure on Health Source: Author s calculation Table 2: Pairwise Correlations Variable Democracy, Level Democracy, Stock Infant mortality rate (ln) Life expectancy at birth(ln) Ln GDP Government expenditure on Health Primary education Democracy, Level [0.000] [0.000] 0.05 [0.14] 0.13 [0.000] 0.07 [0.03] Democracy, Stock [0.000] 0.16 [0.000] 0.18 [0.000] 0.03 [0.36] [0.000] [0.000] Note: P-values in parentheses Source: Author s calculation Infant mortality rate(ln) [0.000] [0.000] [0.47] [0.000] Life expectancy at birth(ln) [0.000] 0.02 [0.37] 0.34 [0.000] Ln GDP [0.035] 0.39 [0.000] Government expenditure on Health [0.12] Primary education Methodology The panel-data models are estimated using the A-B GMM technique developed by Arellano and Bond (1991). The A-B GMM methodology is used to address the possible endogeneity of the right-hand side

21 14 Democracy and Human Development Nexus: The African Experience variables (i.e., per capita income, education, health expenditure and democracy), whilst life expectancy and infant mortality rate are used as separate dependent variables (Baliamoune-Lutza and Bokoc 2012). The model is written as follows: y it = αy i,t 1 + βx i,t 1 + γz it + v i + ε it (1) where, y it represents human development variables in country i at time t; X it is a vector of predetermined and endogenous variables; z it is a vector of exogenous variables; and α, β and γ are parameters to be estimated. The term v i, which represents country specific random effects, is assumed to be independent and identically distributed (iid) over the countries; and ɛ it represents iid disturbances. Terms v i and are assumed to be independent over all time periods and for each country i. The A-B GMM method is derived from instrumental variables principles and provides convergent estimators. The A-B GMM also resolves the problem of correlation between the lagged dependent variable y t-1 and the error term ε it, as well as between explanatory variables X it and/or z it ; and the unobserved country specific term v i. The GMM procedure is based upon a set of orthogonality conditions, which may arise between the error terms and a set of instrumental variables. According to this principle, the GMM estimator must be able to minimize the empirical counterpart of these conditions to zero. The most efficient estimator is obtained when the model in Equation 1 is transformed into a difference equation as follows (Hansen and Tarp 2001; Naceur and Ghazouani 2007): (y it y it 1 ) = (βx it βx i,t 1 ) + (γz it γz it 1 ) + (ε it ε it 1 ) (2) In this specification, the country specific effect is excluded, but a new kind of bias arises since (y it 1 y it 2 ) is correlated with the transformed error term (ε it ε it 1 ). Recall that the vector (βx it βx i,t 1 ) contains the components(y it 1 y it 2 ). Hence, Arellano and Bond (1991) propose the following moment conditions: E(X it s (ε it ε it 1 )) = 0, for s 2; t = 3, T i (3) E(Z it s (ε it ε it 1 )) = 0, for s 2; t = 3, T i (4)

22 Journal of Economic Cooperation and Development 15 With these conditions in mind, the so-called difference estimator is provided after running two steps. In the first run, the error terms are assumed to be independent and homoscedastic across countries and over time. In the second run, the residuals retained at this step serve to construct a consistent estimate for the variance covariance matrix. Thus, the difference estimator is asymptotically more efficient than the first step estimator. Alongside the estimation procedure, Arellano and Bond (1991) construct a test to determine whether second-order correlations exist among the error terms of the first-difference equation provided by Equation 2. The importance of the test is due to the fact that the consistency of the GMM estimator depends on the assumption that E(ε it ε it-2 ) = 0. The appropriate statistic of the test is asymptotically standard normal under the null hypothesis and is defined as follows: N= ε 2 ε ε (5) where, ε 2 is the vector of residuals lagged twice; and ε is a vector of trimmed ε to match ε 2. A Hansan specification test, which is a test of over-identifying restrictions, is also conducted. Under the null hypothesis, the Hansan statistic is asymptotically distributed as χ 2 with p k degrees of freedom and is written as follows: n ε W( i=1 w i ε i ε iw i ) 1 w ε (6) where, W is the chosen matrix of instruments; p indicates the number of columns in W; and k the number of parameters to be estimated. The Hansen test is used to verify independence between the instruments and the error term. The null hypothesis, in this case, is that the instruments and the error term are independent. The Difference-Hansen test is used to verify that the error term is not serially correlated as assumed. Under the null hypothesis, there is no second-order serial correlation. Thus, a failure to reject the null hypothesis for both tests indicates that the instruments are valid. Both the Hansen and Difference-Hansen tests are distributed as χ 2 under the null hypothesis (Naceur and Ghazouani 2007). Firstly, separate examinations of the impacts of democracy level and stock on each of the human development variables are examined.

23 16 Democracy and Human Development Nexus: The African Experience Thereafter, the impacts of the democracy level and the stock of democracy on each of the human development variables are examined to determine whether either democracy level or stock of democracy are dependent or affected by a given country s level of development; the education level of its population; and the provision of health services. To do so, an interaction term is created between the democracy level and the stock of democracy; and the right hand side variables (i.e., per capita GDP, primary education and government expenditure on health). The introduction of the interaction term may lead to multi-collinearity, as it is likely to be strongly correlated with the original variables used to construct the interaction terms (Darlington 1990; Azman et al. 2010). In order to resolve this problem, the interaction term is orthogonalized using the two-step procedures suggested by Burill (2007). First, the interaction term between each pair of variables, (e.g., democratic level and per capita GDP) are regressed on the democratic level and per capita GDP. Second, the residuals from each regression in the first step are used to represent the interaction term (Azman et al. 2010). 4. Estimation Results and Discussion Before the results are interpreted, it is important to note that the results of both the Hansen test for over-identifying restrictions and the test for serial correlation of the residuals (i.e. AR [1] and AR [2]) result in the rejection of the assumption of inconsistency of the GMM estimator. In addition, the difference-in-hansen test of exogeneity results indicate that any correlation between the endogenous variables and the unobserved (fixed) effect is constant over time, which implies that the hypothesis that the additional subset of instruments used in the GMM estimates is exogenous cannot be rejected. Thus, the conclusion is drawn that the results are safe from any statistical problem that may influence the outcomes of the study. Additionally, the study employs data for a large number of African countries that differ in terms of economic structure and level of development. Thus, it is likely that outlier values exist in the data. The data are checked for the presence of outlier values and the model is re-analyzed. Since the quantity of the outlier value in the data is very limited, the results, with and without these values, are identical. Table 3 reports the results of the impacts of democracy level and the stock of democracy on the human development variables without an interaction term. For the variables of interest, the results show that the

24 Journal of Economic Cooperation and Development 17 short run effect, together with the autoregressive coefficient, implies that a ten percent increase in a country s democracy level will lower the infant mortality rate by approximately percent in the short run and by approximately 0.1 percent in the long run. 8 Similarly, a 10 percent increase in a country s stock of democracy will lower the infant mortality rate by approximately percent in the short run and by approximately 0.13 percent in the long run. Meanwhile, for the second human development measurement (i.e., life expectancy), the results shows that a ten percent increase in a country s level of democracy will increase the life expectancy at birth by approximately percent in the short run and by approximately 0.03 percent in the long run. Clearly, the magnitude of the impact of democracy on human development is relatively large when democracy is measured according to democracy level and human development is measured by infant mortality rate. In addition, the magnitude of the impact of the democracy, irrespective of the measurement employed, on human development is relatively large in the long run compared to short run. The finding implies that, over time, political reform in African countries will likely result in improvements in health of their populations. Interestingly, the magnitude of the democracy coefficients in the short run is, to a certain extent, identical to that obtained by Gerring et al. (2012), but a slight difference exists. Gerring et al. (2012) detect an insignificant impact for democracy level on infant mortality rate, but the finding is statistically significant in the case of stock of democracy. This may be due to the difference in the period; the difference in the other variables that are included in the analysis; or due to the difference in the method of estimation. Nevertheless, the results appear to contain good news for African countries that inherited bad institutional systems from earlier regimes and for newly independent countries (e.g., South Sudan). The results indicate that African countries have the ability to improve the well-being of their populations, even with the contemporary status of their respective political institutions. This finding is consistent with recent empirical studies generally (e.g., Gerring et al. 2012; Vollmer and 8 The long run coefficient is given by the coefficient of the democracy /(1- the coefficient of the lag dependent variable) (see Felbermayr et al. 2011).

25 18 Democracy and Human Development Nexus: The African Experience Ziegler 2009), particularly in relation to African countries (Baliamoune- Lutz and Boko 2012; Kudamatsu (2012). Table 3: The impact of democracy on human development (without interaction terms) Explanatory variables Lag Dependent variable Ln GDP Primary education Government expenditure on Health Democracy, Level Dependent variable Ln/ Infant mortality rate 0.97*** 0.97*** [01375] [0.03] [.041] [0.041] *** *** [ ] [0.0001] *** [0.001] *** [0.0004] *** [0.0003] - Dependent variable Ln/ Life expectancy at birth 0.83*** 0.79*** [0.05] [0.05] [0.007] [0.0004] *** *** [0.0001] [0.0001] [0.0004] ** [0.0002] [0.0015] Democracy / Stock *** [0.0002] [0.0002] AR(1) AR(2) Hansen test Hansen test excluding group Difference-in-Hansen tests of exogeneity of instrument Wald test, chi 2 (probability) Number of observation Note: Robust Standard errors in parentheses. *** denotes significance at the 1% level. The results also indicate that education has a statistically significant positive impact on population health, but the impact is marginal. Several empirical studies document the importance of education in relation to health performance; and identify various channels through which education can affect health (e.g., Arendt 2005; Grossman 2005; Cutler and Lleras-Muney 2010). For instance, parental education, particularly the education level of the mother, affects the health of the offspring. The education level of the mother is more significant than that of the father in regards to the health of the child, but both have a positive and -

26 Journal of Economic Cooperation and Development 19 significant effect. The difference results from the fact that mothers are more involved with children s health than fathers (Grossman 2006). Nevertheless, compared to other regions, the African continent still has the lowest literacy rates in the world. For instance, the WBDI database shows that in 2011, the rate of the literacy among adult people in Africa was about 59% compared to 98%, 94% and 75% for Europe; East Asia and Pacific; and the world, respectively. Since the expansion of education is necessary for health improvement, further efforts should be devoted to the improving education levels of populations on the continent. The results also detect the positive influence of government expenditure on health and infant mortality rates. Because the data on this variable (health expenditure) include information on aid and assistance from abroad that is mainly focused on child and maternal health, its effect on the overall mortality rate (i.e., life expectancy) appears to be statistically insignificant. 9 However, generally, the results concerning the positive influence of government expenditure on health and the outcome of such expenditures is consistent with the human development approach, which advocates the significant role of a government in advancing the health of its population through expenditure mechanisms (Haq 2000). The finding is also consistent with others empirical studies that determine the existence of a positive relationship (Anand and Ravallion 1993; Hojman 1996; Bidani and Ravallion1997). However, a number of studies find that health expenditure by the government contributes very little or is statistically insignificant to health status (Mingat and Tan 1992, 1998; Filmer and Pritchett 1999; McGuire 2004). Table 1 demonstrates that African countries, on average, devote 8% of total government expenditures to the health sector, but this ratio is very small compared to other regions. For instance, each of the European and OECD regions devoted more than 16% of the total government expenditure to the health sector in 2011 (i.e., double that of African countries). Since the results indicate the importance of such expenditures on the health of the population, devoting more resources to the health sectors of African countries will likely result in further improvements in the health of the populations of African countries. 9 See the definition of the health expenditure in the World Bank Development Indicator.

27 20 Democracy and Human Development Nexus: The African Experience Columns 1, 2, 4 and 5 in Table 4 show the results concerning the possibility that a country s level of development and the education level of its population are likely to foster or distract the impact of the contemporary status of democracy on each of the human development variables. Columns 3 and 6 in Table 4 represent the impact of the contemporary status of democracy on each of the human development variables through the redistribution mechanism (i.e., provision of more health services). The results show that the impact of the contemporary status of democracy on human development variables is independent of the country s level of development and the education level of its population. The conclusion drawn is despite the fact that GDP per capita and education are essential for human development, they do not distract or foster the ability of democracy to promote the health of a population. Hence, the results prove that it is democracy itself that is imperative for the human well-being. This finding is indistinguishable from the findings of Vollmer and Ziegler (2009), who conclude that the impact of democracy on human development (i.e., life expectancy) is independent of the country s level of development (i.e. per capita GDP) and the education level of the population (i.e., literacy rate). In contrast, as expected, democratic regimes tend to devote more resources to the health sector, which is reflected in further improvements in the health of the population, particularly the health of children. This finding is consistent with the theoretical prediction (i.e., the third theory) concerning the impact of democracy on human development. According to the theory, democracies are forced by electoral processes to spend their revenues on government services, while autocratic governments face no such constraints (Deacon 2003; Lake and Baum 2001; McGuire and Olson 1996; Niskanen 1997).

28 Journal of Economic Cooperation and Development 21 Explanatory variables Lag Dependent variable, Ln GDP Table 4: The impact of contemporary status of democracy on human development (with interaction terms) Dependent variable Ln/ Infant mortality rate Dependent variable Ln/ Life expectancy at birth *** [0.03] [0.04] *** [0.0003] 0.96*** [0.03] [0.04] *** [0.0003] 0.95*** [0.03] [0.04] *** [0.0002] 0.75 [0.07] 0.02 [0.02] *** [0.0001] 0.75 [0.07] 0.02 [0.014] *** [0.0001] 0.75 [0.07] 0.02 [0.01] *** [0.0001] Primary education Government *** *** *** expenditure [0.0001] [0.0001] [0.0001] [0.0004] [0.004] [0.0004] on Health Democracy, * ** *** ** ** 0.001** Level [0.0002] [0.0003] [0.0003] [0.001] [0.0004] [0.0003] Democracy, Level Ln - - [0.0003] [0.0005] GDP - - Democracy, Level Primary [0.0006] [ ] - education Democracy, Level Government *** - - [ ] expenditure [ ] on Health AR(1) AR(2) Hansen test Hansen test excluding group Differencein-Hansen tests of exogeneity of instrument Wald test, chi 2 (probability) Number of observation Note: Robust Standard errors in parentheses. *** denotes significance at the 1% level.

29 22 Democracy and Human Development Nexus: The African Experience Alternatively, democratic governments have a wider range of supporters to appease, which encourages them to produce public goods instead of private ones (Bueno de Mesquita et al. 2003; Ghorbarah et al. 2004). Stasavage (2005a) analysis data for 44 African states and detects strong evidence that democracy has increased government spending on education, and a series of studies of Latin America finds that democracy is robustly linked to higher spending on health, education, and social security (Avelino, Brown, and Hunter 2005; Brown and Hunter 2004; Kaufman and Segura-Ubiergo 2001). The results of Table 4 are identical to the results obtained when democracy level is substituted with the stock of democracy, as shown in Table 5. The results imply that even if a country lacks an appropriate stock of political institutions, any attempt to improve the contemporary status of such institutions will be reflected in a significant improvement in the population's health, similar to a country that inherited a good stock of political institutions. Thus, the claim that the relative failure of most African countries to achieve the MDGs due to bad political institutions inherited from former regimes is inaccurate (UNDP, ). This is because the contemporary status of a country, and/or even its stock of democracy, is able to generate improvements for the population of that country. In order to check for robustness, the relationship between democracy and human development is re-examined using different proxies, 10 as discussed above. Literacy rate is utilized instead of primary education and the overall results remain the same. Nevertheless, when the model is estimated using fixed and random effect approaches, the results are slightly different because the fixed and random effect techniques are not able to capture endogeneity and omitted variables problems (Baliamoune-Lutza and Bokoc 2012). 10 The results are not reported, but available upon request.

30 Journal of Economic Cooperation and Development 23 Explanatory variables Lag Dependent variable Table 5: The impact of stock of democracy on human development (with interaction terms) Dependent variable Ln/ Infant mortality rate Dependent variable Ln/ Life expectancy at birth [0.03] 0.95 [0.03] 0.95 [0.03] 0.84 [0.04] 0.81 [0.05] 0.79 [0.06] Ln GDP [0.04] [0.04] [0.04] [0.009] [0.009] [0.009] Primary education *** [0.0002] *** [0.0003] ** [0.0002] *** [0.0001] *** [0.0001] 0.004*** [0.0001] Government *** *** *** expenditure on [0.0001] [0.0001] [0.0001] [0.001] [0.001] [0.002] Health Democracy, Stock [0.0001] ** [0.0002] ** [0.0002] [0.0003] [0.0003] [0.0003] Democracy, stock Ln [0.0002] GDP Democracy, stock Primary [0.004] [0.0003] - education Democracy, stock Government *** - - [0.0001] expenditure on [ ] Health AR(1) AR(2) Hansen test Hansen test excluding group Difference-in- Hansen tests of exogeneity of instrument Wald test, chi 2 (probability) Number of observation Note: Robust Standard errors in parentheses. *** denotes significance at the 1% level.

31 24 Democracy and Human Development Nexus: The African Experience 5. Conclusion Using data from 46 African countries for the period of 1990 to 2012, the present study examines three principal issues. First, the study examines whether human development is affected by the contemporary status or the stock of the democracy in the countries; and whether the effect varies over time. Second, in the context of democratic regimes, the study investigates whether the strength of the impact of such regimes on human development depends on a country s level of development and the education level of its population. Third, the study determines whether democratic regimes, through their redistribution mechanisms, help to further improve the well-being of the population. The results of the A-B GMM technique show that democracy level or stock of democracy has a positive influence on human development in terms of infant mortality rate and life expectancy. The results also show that this impact is independent of the country s level of development and the education level of the population. In addition, democratic regimes tend to devote a considerable proportion of government resources to the health sector through redistribution mechanisms, which is likely to be reflected by further improvements in the well-being of the population. The findings are robust, since the results are similar even when an alternative proxy and different estimation methods are employed. The finding implies that, over time, political reform in African countries will likely result in further improvements in the well-being of the population. In general, the results seem to contain good news for African countries that inherited bad political institutions or systems. The results indicate that the African countries can improve the well-being of their populations irrespective of the current status of their political institutions. The result also implies that even poor countries and countries with low levels of literacy can achieve better human development if the country have higher levels of democracy or are governed by a democratic regime. Nevertheless, because of data limitations, the present study is unable to consider the impact of democracy, irrespective of how it is measured, on human development due to inequalities in the distribution of income, assets and human capital between populations (i.e., the median voter theory). This is imperative since the unequal distribution of physical

32 Journal of Economic Cooperation and Development 25 capital and human resources between populations, which are a common feature amongst the African countries, may affect the strength of the impact of democracy on human development (Vollmer and Ziegler 2009). Thus, future studies in this area should incorporate the degree of such distribution in attempt to provide more light on this issue. Moreover, the influence of the ethnic fragmentation, which is also another common feature in the African region, on the strength of the impact of democracy on human development should be investigated. According to Vollmer and Ziegler (2009), under specific circumstances, ethnic fragmentation and the heterogeneity of the societies in a country may influence the relationship between political institutions and human development.

33 26 Democracy and Human Development Nexus: The African Experience References Acemoglu, D. and Robinson, J.A. (2005) Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press. Anand, S. and Ravallion, M. (1993) Human Development in Poor Countries: On the Role of Private Incomes and Public Services, The Journal of Economic Perspectives, 7(1): Arellano, M. and Bond, S. (1991) Some Tests of Specification for Panel Data: Monte Carlo Evidence and an democracy and human development, Application to Employment Equations, Review of Economic Studies, 58(April): Arendt, J. (2005) Does education cause better health? A panel data analysis using school reforms for identification, Economics of Education Review, 24(2): Avelino, G., Brown, D. and Hunter, W. (2005) The Effects of Capital Mobility, Trade Openness, and Democracy on Social Spending in Latin America, , American Journal of Political Science, 49(3): Azman-Saini,W., Zubaidi, A. and Hook, S. (2010) Law, Foreign direct investment, economic freedom and economic growth: International evidence, Economic Modelling, 27: Baliamoune, M. L. and Bokoc, S. (2012) Trade, Institutions, Income and Human Development in African Countries, Journal of African Economies, 22(2): Besley, T. and Kudamatsu, M. (2006) Health and Democracy. AEA Papers and Proceedings, 96(2): Bidani, B. and Ravallion, M. (1997) Decomposing Social Indicators Using Distributional Data, Journal of Econometrics, 77(1): Boix, C. (2003) Democracy and Redistribution. New York: Cambridge University Press.

34 Journal of Economic Cooperation and Development 27 Boix, C. (2001) Democracy, Development and the Public Sector, American Journal of Political Science, 45:1 17. Brown, D.S. and Mobarak, A.M. (2009) The Transforming Power of Democracy: Regime Type and the Distribution of Electricity, American Political Science Review, 103(May): Brown, D.S. and Hunter, W. (2004) Democracy and Human Capital Formation: Education Spending in Latin America, , Comparative Political Studies, 37: Bueno de Mequita, B., Smith, A., Siverson, R.M. and Morrow, J.D. (2003) The Logic of Political Survival. Cambridge, MA: MIT Press. Burill, D. (2007) Modelling and interpreting interactions in multiple regression, Internet, available at Collier, R.B., and Collier, D. (1991) Shaping the Political Arena: Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Princeton, NJ: Princeton University Press. Cutler, D. and Lleras-Muney, A.(2010) Understanding Differences in Health Behaviours by Education, Journal of Health Economics, 29(1):1-28. Darlington, R. (1990) Regression and Linear Models. McGraw-Hill, New York. Deacon, R.T. (2003) Dictatorship, Democracy, and the Provision of Public Goods. Unpublished manuscript. Dreze, J. and Sen, A. (1989) Hunger and Public Action. Oxford: Clarendon Press. Felbermayr, G., Prat, J., and Schmerer, H. (2011) Trade and unemployment: What do the data say?, European Economics Review, 55:

35 28 Democracy and Human Development Nexus: The African Experience Filmer, D. and Pritchett, L. (1999) The Impact of Public Spending on Health: Does Money Matter?, Social Science and Medicine, 49 (November): Franco, A., Alvarez-Dardet, C. and Ruiz, M.T. (2004) Effect of democracy on health: ecological study, British Medical Journal, 329: Gauri, V. and Lieberman, E. (2006) Boundary Politics and Government Responses to HIV/AIDS in Brazil and South Africa Studies, Comparative International Development, 41(3): Gauri, V. and Khaleghian, P.( 2002) Immunization in Developing Countries: Its Political and Organizational Determinants, World Development, 30 (December): Gerring, J. Thacker, S. and Alfaro, R. (2012) Democracy and Human Development, The Journal of Politics, 74(1):1 17. Gerring, J., Bond, P., Barndt, W. and Moreno, C. (2005) Democracy and Growth: A Historical Perspective, World Politics, 57(April): Ghobarah, H.A., Huth, P. and Russett, B. (2004) Comparative Public Health: The Political Economy of Human Misery and Well-Being, International Studies Quarterly, 48(1): Glaeser, E. L., Ponzetto, G.A.M. and Shleifer, A. (2007) Why does democracy need education?, Journal of Economic Growth, 12: Gradstein, M. and Milanovic, B. (2004) Does Liberte =Egalite? A Survey of the Empirical Links between Democracy and Inequality with some Evidence on the Transition Economies, Journal of Economic Surveys, 18 :(4): Grossman, M. (2006) Education and Nonmarket Outcomes, Handbook of the Economics of Education, 1: Hansen, H. and Tarp, F. (2001) Aid and growth regressions, Journal of Development Economics, 64:

36 Journal of Economic Cooperation and Development 29 Haq, M. (2000) Reflections on human development. (Expanded edition) New Delhi: Oxford University Press. Hite, K. and Cesarini, P.. (2004) Authoritarian Legacies and Democracy in Latin America and Southern Europe. Notre Dame, IN: University of Notre Dame Press. Hojman, D.E. (1996) Economic and other Determinants of Infant and Child Mortality in SmallDeveloping Countries: The Case of Central America and the Caribbean, Applied Economics, 28: Kapstein, E.B. and Converse, N. (2008) The Fate of Young Democracies. Cambridge: Cambridge University Press. Kaufman, R. R. and Segura-Ubiergo, A. (2001) Globalization, Domestic Politics, and Social Spending in Latin America, World Politics, 53(4): Kaufmann, D., Kraay, A. and Mastruzzi, M. (2007) Governance Matters VI: Governance Indicators for , World Bank Research Working Paper 4280, World Bank, Washington, DC. Keefer, P. and Khemani, S. (2005) Democracy, Public Expenditure, and the Poor: Understanding Political Incentives for Providing Public Services, The World Bank Research Observer, 20(1):1-27. Keefer, P. (2006) Clientelism, Credibility and the Policy Choices of Young Democracies, American Journal of Political Science, 51(October): Kudamatsu, M. (2012) Has Democratization Reduced Infant Mortality In Sub-Saharan Africa? Evidence from Micro Data, Journal of the European Economic Association 10(6): Kurzman, C.K., Werum, R.W. and Burkhart, R.E. (2002) Democracy s Effect on Economic Growth: A Pooled Time-Series Analysis, , Studies in Comparative International Development, 37 (Spring):3 33.

37 30 Democracy and Human Development Nexus: The African Experience Lake, D.A. and Baum, M.A. (2001) The Invisible Hand of Democracy: Political Control and the Provision of Public Services, Comparative Political Studies, 34(August): Lenski, G. (1966) Power and Privilege: A Theory of Social Stratification. Chapel Hill: University of North Carolina Press. Lipset, S.M. (1959) Some Social Requisites of Democracy: Economic Development and Political Development, American Political Science Review, 53(March): Marshall, M.G., and Jaggers, K. (2013) Polity IV Project: Political Regime Characteristics and Transitions, McGuire, M.C. and Mancur, O. (1996) The Economics of Autocracy and Majority Rule: The Invisible Hand and the Use of Force, Journal of Economic Literature, 34:1: McGuire, J.W. (2006) Democracy, Basic Service Utilization, and Under-5 Mortality: A Cross- National Study of Developing States, World Development, 34(3): McGuire, J.W. (2004) Democracy, Health Care, and Under- 5 Mortality: A Cross-National Study of Developing Countries. Wesleyan University. Unpublished manuscript. Meltzer, A.H. and Richard, S.F. (1981) A Rational Theory of the Size of Government, Journal of Political Economy, 89(5): Mingat, A. and Tan, J.P. (1998) The mechanics of progress in education: evidence from cross-country data, Policy research working paper 2015, World Bank, Washington, DC. Mingat, A. and Tan, P. (1992) Education in Asia: A Comparative Study of Cost and Financing, Washington, DC: World Bank. Muller, E.N. (1988) Democracy, Economic Development, and Income Inequality, American Sociological Review, 53(February):50 68.

38 Journal of Economic Cooperation and Development 31 Munck, G.L. and Verkuilen, J. (2002) Measuring Democracy: Evaluating Alternative Indices, Comparative Political Studies, 35(1):5 34. Samy, N. and Ghazouani, S. (2007) Stock markets, banks, and economic growth: Empirical evidence from the MENA Region, Research in International Business and Finance, 21: Niskanen, W.A. (1997) Autocratic, Democratic, and Optimal Government, Economic Inquiry, 35(3): Persson, T., G. Roland, and G. Tabellini (2000) Comparative Politics and Public Finance, The Journal of Political Economy, 108,( 6): Persson, T. and G. Tabellini (2000) Political Economics. Explaining Economic Policy. Cambridge, Massachusetts, London: MIT Press. Peters, G. B. (2005): Institutional Theory in Political Science. London, New York: Continuum. Przeworski, A., Alvarez, M.E., Cheibub, J.A. and Limongi, F. (2000) Democracy and Development: Political Institutions and Well-Being in the World, New York: Cambridge University Press. Ranis, G., Ramirez, A. and Stewart, F. (2000) Economic growth and human development, World Development, 28(2): Ross, M. (2006) Is Democracy Good for the Poor?, American Journal of Political Science, 50(October): Sen, A. (1981) Poverty and Famines: An Essay on Entitlement and Deprivation. New York: Oxford University Press. Sen, A.K. (1999a) Development as Freedom. Oxford: Oxford University Press. Sen, A.K. (1999b) Democracy as a Universal Value, Journal of Democracy, 10(3):3-17.

39 32 Democracy and Human Development Nexus: The African Experience Shandra, J.M., Nobles, J., London, B. and Williamson, J.B. (2004) Dependency, Democracy, and Infant Mortality: A Quantitative, Cross- National Analysis of Less Developed Countries, Social Science and Medicine, 59(July): Stasavage, D. (2005a) Democracy and Education Spending in Africa, American Journal of Political Science, 49(2): Tavares, J. and Wacziarg, R. (2001) How democracy affects growth, European Economic Review, 45: Tsai, M.-C. (2006) Does Political Democracy Enhance Human Development in Developing Countries, American Journal of Economics and Sociology, 65(2): UNDP ( ) Investing in Development A Practical Plan to Achieve the Millennium Development Goals, UNDP. UNDP (1991) Human development report 1991: Financing human development. NY: Oxford University Press. Vollmer, S. and Maria, Z. (2009). Political Institutions and Human Development Policy Research Working Paper 4818, World Bank, Washington, DC.

40 Journal of Economic Cooperation and Development 33 Appendix 1: List of the countries Algeria Gabon Niger Angola Gambia, The Nigeria Benin Guinea Rwanda Botswana Guinea-Bissau Senegal Burkina Faso Kenya Seychelles Burundi Lesotho Sierra Leone Cameroon Liberia South Africa Cape Verde Libya Sudan Central African Republic Madagascar Tanzania Chad Malawi Togo Comoros Mali Tunisia Congo, Dem. Rep. Mauritania Uganda Congo, Rep. Mauritius Zambia Cote d'ivoire Morocco Zimbabwe Egypt, Arab Rep. Mozambique Ethiopia Namibia

41

42 Journal of Economic Cooperation and Development, 37, 2 (2016), The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model Dr. Abdel Hakeem Ahmad Eltalla 1 The role of infrastructure as a catalyst of economic development is accepted. Among the different types of transport infrastructure, seaports are considered as especially strategic due to the increasing importance of maritime transport in deciding transport costs and international trade in the age of globalization. We analyze the economic impacts of constructing a Gaza Seaport. We use a computable general equilibrium model of the Palestinian economy to simulate the effects on the Palestinian economy brought about by the planned construction of a Gaza Seaport. According to the model's estimates, GDP will increase by 3.90 percent, household income by 3.11 percent, private consumption by 7.73 percent, imports by percent, and exports by percent and labor income will increase by 3.19 percent. Introduction The role of infrastructure as a catalyst of economic development is accepted. Among the different types of transport infrastructure, seaports are considered as especially strategic due to the increasing importance of maritime transport in determining transport costs and international trade in the age of globalization. Improvement in transportation and communication has increased mobility of goods and services, and economies in the world are quickly integrated into one global economy as a result. The oceans and seas, as a main international trade route, offer low cost and important transport means. Maritime cargo, measured in tonnes grew to about 8.7 billion tonnes in 2011 according to UNCTAD, while only 43 million tonnes were transported by air in 2011 (Boeing, 2012). Maritime transport handles over 80 percent of international trade as measured in weight (UNCTAD, 2012). Seaports have a crucial role in integrating national economies into the international economy. Seaports have been designed to assist local 1 Faculty of Business and Finance, Alaqsa University, Gaza, Palestine

43 36 The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model economic activities at early periods of industrialization. Heavy industries are dependent upon seaport services and low cost transportation means like shipping. Seaports and national economies have been dependent on each other. Seaports have been planned around industrial compounds, and industrial compound have been structured around seaports. Unconnected economies cannot ensure sustainable growth, thus, national economies are needed to integrate into the global economy for sustainable development. Seaports, as essential entranceway for international trade, are predicted to have positive effects on economic development. Export enterprises established near a seaport benefited from efficient logistics services of the seaport. Consequently, seaports have been considered to induce economic development in the cities where they are located. Seaports also have created substantial direct local employment and value added production effects (Jung, 2011). The Palestinian economy has become increasingly connected to the Israeli economy since the beginning of Israel s occupation in The existing trade regime (Customs Union) between the Palestinian Authority and Israel, as agreed under the terms of the Paris Protocol has increased the dependence of the Palestinian economy on Israel. The Protocol stipulates the freedom of movement of goods between the West Bank, Gaza Strip and Israel. Economic viability through increase trade relations and export potentials is the best means to improve political stability and lower unemployment in the Palestinian society. Supported by suitable growth strategies, trade can assist the structural change and transformation of the Palestinian economy toward higher productivity, leading to positive impact on unemployment, growth in income and standard of living. Although there are many bilateral trade agreements with different countries, the Palestinian enterprises have not been able to make significant advances in the international trade arena, mostly due to the barriers placed on the execution of these agreements by Israel. Israeli restrictions on the movement of both labor and goods have affected the normal flow of Palestinian exports and imports, leading to distortions in the economy, including the loss of international markets, and increased transaction costs. Exporters and importers from Gaza Strip must use Israeli transport companies when goods are going to or leaving Gaza Strip through Israel. Furthermore, shipments have to be reloaded at the crossing points with Israel and are subjected to special screening procedures. This has diminished the competitiveness of Palestinian

44 Journal of Economic Cooperation and Development 37 exports, establishing trade barriers of greater consequence than tariffs. Although Palestine has a seacoast of its own, it is effectively a landlocked economy. Palestinian enterprises are mainly dependent on Israeli seaports, and since 2000 Israel has been enforcing complex security measures. Israeli security measures at the main border crossings have undermined the ability of Palestinian industries. Israel s handling of Palestinian trade activities has been time consuming and expensive. The State of Palestine needs to have a seaport in Gaza in order to undo its forced landlocked economy status. Removing this forced status is essential for any real reduction in the high trade costs. The economic benefit of the Gaza seaport project in its regional context as a strategic project is the best option for the Palestinian State. The use of alternative maritime transport routes for Palestinian trade should not be considered as an alternative for the construction of a seaport in Gaza. The fact that land transport is the main cost part in the shipping of Palestinian trade from and to the rest of the world, and that more than 50 percent of the expenses related to this factor emerges from the Israeli closure policy and the security measures (UNCTD, 2006). Thus, a construction of a seaport in Gaza would be useful in restoring trade competitiveness and output growth level. To evaluate the impact of the construction of a seaport in Gaza on the Palestinian economy, we constructed a general equilibrium model that depicts the economic conditions and features of the economy of Palestine, and we constructed a 2012 social accounting matrix for Palestine. Literature Review Economic impact studies of seaports assist to generate understanding into the contributions of seaport activity to the economic development. The transport economics literature had emphasized that seaports induce economic development and growth because they increase competition through expansion of the markets, thereby reduction of prices for consumers (Goss, 1990; Acosta et al, 2011). Dooms et al. (2011) presented a meta-analysis of economic effect studies for seaports, and found that there is variety in terms of measures of economic effects. Ferrari et al. (2010) reviewed seaport impact studies. Ferrari et al. (2010) evaluated the effect of port activity on local development in terms of employment in Italian provinces. Ferrari et al. (2010) applied a

45 38 The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model two-stage econometric process which individually estimates traffic and an employment equation. Ferrari et al. (2010) observed that the bigger the port, the greater it s direct, indirect and induced effects and recognizing the relevance of their output. Danielis and Gregori (2013) summarized the results of a research project intended to detecting the main economic characteristics of the Friuli Venezia Giulia Region seaport system in Italy and the part it plays inside the economy. Danielis and Gregori (2013) concluded that the Venezia Giulia Region seaport system plays a important macroeconomic function in the region. Bottasso et al. (2013) studied the effect of seaport on local employment by studying a sample of 560 areas in ten West European countries between They found higher impact of port and regional employment is positively correlated to seaport. Deng, et al. (2013) examined the relationship between seaports and regional economy from a logistics perspective in China. That helps policy makers in their seaport decisions. They found that seaport supply had a positive impact on port demand and the value added activity in seaport had a positive effect on the development of regional economy. Wilson et al. (2003) used a gravity model to analyze the connection between trade facilitation, trade flows and economic growth in the Asia-Pacific area for the commodities sector. Seaport efficiency represented the quality of infrastructure of maritime. They found that enhanced seaport efficiency has positive impacts on trade. Gaza Seaport The Palestinian economy is a landlocked economy. In contrast to other land-locked economies, where access to world markets is delayed by the lack of a coast, Palestine s poor market access conditions are created by the prolonged delays in the building of a seaport in Gaza. Israeli control of the borders and transport routes made Palestinian trade completely reliant on political and security factors. Although participation in international trade is facilitated through Israeli seaport, high trade costs and complex procedures continue to diminish benefits derived from foreign trade and discourage investments in productive sectors. The State of Palestine needs to have a seaport in Gaza in order to undo its forced landlocked economy status. To remove this forced status is essential for any real lessening in the high trade costs. The economic benefit of the Gaza seaport project in its regional context as a strategic project is the best option for the Palestinian State. The use of alternative

46 Journal of Economic Cooperation and Development 39 maritime transport routes for Palestinian trade should not be considered as an alternative for the construction of a seaport in Gaza. The land transport is the main cost factor in the transport of Palestinian trade, and that more than 50 percent of the expenses related to this factor emerges from the Israeli closure policy and the security measures (UNCTD, 2006). The use of alternative routes and the future operation of the Gaza seaport are required complements in integrating Gaza seaport into the regional ports system, and the Palestinian economy into the regional and world markets. The construction of Gaza seaport began in November 1999, and in April 2000 the Gaza Seaport Authority signed an agreement with the European Gaza Development Group 2000, to execute the project. Finance needed for Phase 1 was approximately $70 million, with financial support from the Dutch and French governments, and a loan from the European Investment Bank. The construction of the Gaza Seaport has suspended since the eruption of the conflict in September The Gaza seaport was assumed to be operational in The port of Gaza was considered as an important element needed for the economic strength of the Palestinian economy and for the economic growth. The Gaza seaport is also important for integrating the Palestinian economy to the rest of the world (UNCTD, 2004). The deep-water Gaza seaport is planned to be built, around 200 kilometres west of Amman. The Gaza seaport will provide Palestinian traders and Jordanian traders with a substitute transport corridor to Europe and America. Future strategy consists of connecting its facilities with the areas ports of Port Said, Beirut and Cyprus. The Gaza seaport project has a strategic importance for the Palestinian economy; it will continue to be high on the priorities to construct the seaport of Gaza once conditions are allowed. The Gaza seaport will capitalize on the territory's location as a regional transit corridor, with plans to serve Jordan and establish links with neighbouring seaports in Egypt, Lebanon, Jordan and Cyprus. One important aspect of the corridor link between Gaza Strip and the West Bank will be the Gaza seaport (UNCTD, 2004). Methodological Framework and Data Computable general equilibrium models are an important tool of analysis in development economics. The computable general equilibrium methodology is a powerful methodological tool for

47 40 The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model examining the impacts across multiple markets of changes in policy variables or exogenous shocks, and an instrument for policy analysis. They are used for making predictions about the behaviour of economies in response to shocks and to different policies. Computable general equilibrium models are derived from economic theory. The competitive market equilibrium of supply and demand is determined by the demand functions of the consumers and the production functions of the firms. Computable general equilibrium models represent the direct and indirect interactions between all sectors of the economy. The computable general equilibrium framework provides a theoretical quantification that combines the general equilibrium arrangement organized by Arrow and Debreu with real economic data -provided by a social accounting matrix- to solve numerically for the quantities of supply, demand and price that preserve equilibrium across all markets. Markets respond to changes in prices. The general equilibrium happens when a set of prices make supply equal to demand in all markets at the same time (Shoven and Whalley, 1984). Computable general equilibrium models are a branch of economic models that work on actual economic data to evaluate the impacts of changes in policy, technology or external factors on the economy. They provide an economy-wide framework for policy analysis to assess a broad range of policy issues. This economy-wide, multi-market approach captures all sectorial and inter-sectorial price linkages simultaneously rather than analyzing each commodity market separately. The computable general equilibrium model that we use is neoclassical. Its framework is developed from the micro-economic foundations of optimization behaviour of rational economic agents. Consumers demand commodities and supply their endowments to maximize their utility, subject to their endowments. Producers (activities) demand inputs and supply outputs to maximize their profits, subject to production technologies. The optimizing assumptions emphasize the role of commodity and factor prices in affecting consumption and production decisions by households and producers. The model is formulated on a Walrasian system with the assumption of general equilibrium, which can be obtained when supply equals demand across all connected markets in the economy at a matrix of relative prices (Dervis et al., 1982). To undertake Computable general equilibrium analysis, a Palestinian Computable general equilibrium (CGE) model has built based on the standard model used by the International Food Policy Research Institute (IFPRI) (Lofgren et al.,

48 Journal of Economic Cooperation and Development ). Lofgren et al. (2002) has a complete description of the IFPRI's standard model. Social Accounting Matrix A social accounting matrix is a comprehensive, economy wide data framework, representing the economy of a country. Social accounting matrix is a square matrix in which each account is represented by a row and a column. The elements of the matrix represent the payment from the account of a column to the account of a row. A social accounting matrix accounts for the economy-wide circular flow of incomes and payments in the economy. It represents the structure, internal and external links of the economy, and the roles of agents and sectors in the economy. The sources of data for the social accounting matrix are an input-output matrix, national income accounts, household income, and expenditure statistics. Thereby, it is wider than an input-output matrix and national accounts. These data are from different time periods, but they still provide a good indication of the structure of the economy and the interactions among social and economic entities (King, 1985; Roland-Holst, 2008). A social accounting matrix is built on a walrasian general equilibrium framework. Walras' law is the principle for organizing the information in the social accounting matrix. It is assumed that agents earn incomes from selling their initial endowments to other agents. The agents spend part of their incomes to buy commodities or primary factors in the markets. All exchanges occur, in which for each income formed must be a corresponding expenditure. The revenues are located in the row accounts and expenditures in the column accounts. Since revenues must be accounted for by expenditures, the total of rows and columns must be equal for a given account (double entry accounting). Thereby for a consistent social accounting matrix, the sum of rows (revenues) and columns (expenditures) of each account must balance. However, because the data are often inconsistent with each other, thus it is likely the data will lead to unbalanced social accounting matrix. Thus balancing the social accounting matrix is needed which done by using a GAMS code to equate the sum of rows and their correspondent columns (Robinson et al., 2001; Fofana et al., 2005). A social accounting matrix contains most of the data required to implement a computable general equilibrium model analysis. The computable general equilibrium model has to be based on recent

49 42 The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model relevant available data to be credible for policy analysis. When historical data are used for policy analysis, it should be demonstrated that the structure of the economy has not substantially changed for the evaluation and analysis of policies to be credible and valuable. A 2012 social accounting matrix for Palestine is constructed. The 2012 social accounting matrix is used as the initial data for the calibration of the Palestinian computable general equilibrium model. See table 2: Macro 2012 social accounting matrix for Palestine million of dollars. Why Computable General Equilibrium Computable general equilibrium models are constructed to estimate the consequences of policy adjustment or exogenous shocks; their framework is more comprehensive and economy-wide than other models. Econometric models need reliable and lengthy time series data on economic variables for the evaluation of associations between economic variables. In contrast, computable general equilibrium models require fewer historical data. Most of the parameters for computable general equilibrium models are based on economic data (summarized in the social accounting matrix) of a benchmark year. The calibration process gives values to the parameters of the model equations using the base year social accounting matrix. Furthermore, the model has to be based on recent applicable existing data to be reliable for policy analysis. Therefore, when historical data are utilized, it should be verified that the structure of the economy has not considerably changed for the estimation and analysis of policies to be reliable and useful. This is not the case in Palestine. The environment of the West Bank and Gaza and their economic structure distorted dramatically after the Second Intifada in This is the reason we utilized computable general equilibrium model; it provides insights into the impacts of shocks even without time series data. The social accounting matrix of Palestine is a comprehensive, economy-wide data framework that provides a comprehensive representation of the socio-economic structure of Palestine. Trade Costs in the Model The notation principles make it possible to differentiate between variables (upper-case Latin letters) and parameters (lower-case Latin letters). The price system of the model assumes quality variations among

50 Journal of Economic Cooperation and Development 43 commodities of various origins and destinations: imports, exports, and domestic outputs used nationally. Endogenous prices are related to other prices (endogenous or exogenous) and to non-price model variables. The trade costs enter in the following equations: the price of imports and the price of exports. A reduction of the trade costs affects the model through those equations. The import prices paid by domestic consumers for imported commodities include import tariffs and trade costs per import unit icm. The world price of imports (pwm) transforms to the import price (PM) by considering the exchange rate and import tariffs plus trade costs icm. The equation of the import price of good c is: PM c = pwm c. (1 + tm c ). EXR + PQ c. icm cc Where c is a commodity, PM is the import price including trade costs, pwm is the world market import price, PQ is the composite price (the market price paid by domestic commodity consumers), tm is the import tariff rate, EXR is the exchange rate, and icm is the trade costs per imported unit. The import price (PM) is the price paid by domestic users for imported commodities. The import price is affected by the trade costs, which increase the price paid by the consumers. The export price (PE) is the price granted to domestic producers for their exports. The world price of exports (pwe) transforms to the export price (PE) by considering the trade costs and export tariffs plus exchange rate. The equation of the export price of good is: PE c = pwe c. (1 te c ). EXR + PQ c. ice cc Where PE is the export price, pwe is the world market export price, te is the export tax rate and ice is the trade costs per exported unit. The export price is the price received by domestic producers, which is affected by the export taxes (te), the trade costs and the exchange rate. The export price is affected by the trade costs, which reduce the price received by the domestic producers of exports (Lofgren et al., 2002). Simulations and Empirical Results We simulated a 25 percent reduction of trade costs. Possible reason for the reduction of trade costs is the construction of Gaza seaport. We used the General Algebraic Modelling System (GAMS) to perform the simulation. Table 1 shows the effects on selected variables of the

51 44 The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model Palestinian economy for a 25 percent decrease of trade costs. The baseyear (benchmark) values correspond to the values found in the Palestinian social accounting matrix. The impact of a 25 percent decrease in overall trade costs is to increase GDP by about 3.90 percent, household income by 3.11 percent, private consumption by 7.73 percent, imports by percent, and exports by percent. Government revenue increases by 5.43 percent. Overall absorption increases by 5.05 percent. The income of capital increases by 3.50 percent and labor income increases by 3.19%. Effects of a 25 percent decrease in trade costs on spending and income: Table 1: Effects of a 25 percent decrease in trade costs on spending and income millions USD As % of GDP Base line Change % Change Base line Change Absorption Private consumption Gov. consumption Investment Exports Imports Net Taxes GDP GDP at factors cost Trade Deficit Source: Authors calculations.

52 Journal of Economic Cooperation and Development 45 Source: Authors calculations. Table 2: The macro 2012 social accounting matrix of Palestine Total 1-Activities Commodities Factors Households Government Saving-Invest Taxes Rest of the World Total

53 The Economic Impact of Constructing a Gaza Seaport: 46 Evidence from a Computable General Equilibrium model Conclusion Palestine is in fact landlocked despite the long seacoast of the Gaza Strip, with total reliance on Israel transport facilities for participation in international trade. That makes Palestinian trade entirely reliant on political and security developments, which have lead to prohibitive transit transport costs and damage to the competitiveness of exports. The economy of Palestine is burdened by stifling trade costs resulting from the absence of a seaport, geographical fragmentation, frequent border closures, and checkpoint controls. Better political environment and construction of a seaport in Gaza may lower trade costs. We used a general equilibrium model to quantify the effect of a 25 percent reduction in trade costs. The effects are substantial. The simulation results show that GDP will increase by 4 %, import and export will increase by 12.93% and 27.41% respectively. Net taxes will increase by 5.43%. The level of private consumption will increase by 7.73 % from the base line and the labor income will increase by 3.19 %. Palestinian economic development policy framework requires identifying the circumstances of the Palestinian economy. Policies measures are necessary to assist evade instability and inefficiency, which consist of policies to promote exports, savings and investment. For the small and poor Palestinian economy, whose economic links were restricted for decades, establishing new trade links is vital for decreasing the reliance on trade with Israel; trade links between Israel and Palestine, which originated on the customs union under Paris Protocol, have intense political and economic consequences. The trade between Palestine and its partners should imitate the requirements of the Palestinian balance growth and development programme. A framework for Palestinian foreign trade that gives a level of independence to Palestinian trade flows to grant direct access to world markets and reduce the dependence on Israel which cannot be achieved beneath the current infrastructure. Basic facility for instance a seaport is necessary for creating normal trade links with the world. The simulation of the effects of a construction of Gaza seaport provides some policy lessons to the Palestinian Authority policymakers: there is real benefit from a construction of Gaza seaport on the economy. In addition, negotiating new trade agreements with various countries and facilitate broad trade by increasing investments in transportation infrastructure are essential measures to facilitate trade, development and balanced growth by creating links to the rest of the world. The results suggest that policies to lower trade costs (investment in infrastructure) have big benefits.

54 Journal of Economic Cooperation and Development 47 References Acosta, M., Coronado, D., & Cerban, M. (2011). The Economic Impact of the Port of Tarifa (Spain) in 2007 and the Forecast for International Journal of Transport Economics, 38(3). Boeing, (2012). World Air Cargo Forecast , September 2012, available at : Bottasso, A., Conti, M., Ferrari, C., Merk, O., & Tei, A. (2013). The impact of port throughput on local employment: Evidence from a panel of European regions. Transport Policy, 27, Danielis, R., & Gregori, T. (2013). An input-output-based methodology to estimate the economic role of a port: The case of the port system of the Friuli Venezia Giulia Region, Italy. Maritime Economics & Logistics, 15(2), Deng, P., Lu, S., & Xiao, H. (2013). Evaluation of the relevance measure between ports and regional economy using structural equation modeling. Transport Policy, 27, Dervis, K., de Melo, J. and Robinson, S. (1982). General Equilibrium Models for Development Policy. Cambridge: Cambridge University Press. Dooms, M., Haezendonck, E., & Verbeke, A. (2011). Towards a toolkit for port related socio-economic impacts: a meta-analysis of socioeconomic impact studies conducted for seaports. In European Conference on Shipping & Ports. Eltalla, Hakeem, and Luc Hens (2009). The Impact of Trade Transaction Costs on Palestine. In International Trade and Finance Association Conference Papers (p. 4). bepress. Eltalla, Hakeem, and Luc Hens (2010). The Economic Impact of Donor Aid to Reconstruct Gaza. GARNET Conference The European Union in International Affairs, Brussels.

55 48 The Economic Impact of Constructing a Gaza Seaport: Evidence from a Computable General Equilibrium model Ferrari, C., Percoco, M. and Tedeschi, A. (2010). Ports and local development: Evidence from Italy. International Journal of Transport Economics 37(1): Fofana, I., Lemelin, A. and Cockburn, J. ( 2005). Balancing a Social Accounting Matrix. Inter-university Center on Risk, Economic Policies and Employment (CIRPEE). Fox, A., Francois, J., and Londono Kent, P. (2003). Measuring Border Crossing Costs and their Impact on Trade Flows: The United States- Mexican Trucking Case. GTAP Resource no. 1282, Center for Global Trade Analysis. Purdue University. Goss, R. O. (1990). Economic policies and seaports: The economic functions of seaports. Maritime Policy & Management, 17(3), JACOBS, W., & LAGENDIJK, A. (2014). Strategic coupling as capacity: how seaports connect to global flows of containerized transport. Global Networks, 14(1), Jung, B. M. (2011). Economic contribution of ports to the local economies in Korea. The Asian Journal of Shipping and Logistics, 27(1), King, B. (1985). What is SAM? In Social Accounting Matrix: A Basis for Planning. Pyatt, G. and Round, J. (eds.). Washington D.C: The World Bank. Liu, H. (2010). Analysis of the Impact of Globalization on the Palestinian Economy. International Journal of Business and Management, 5(7), P204. Lofgren, H., Harris, R. and Robinson, S. (2002). A Standard Computable General Equilibrium (CGE) Model in GAMS. Trade and Macroeconomics, International Food Policy Research Institute, Washington, DC. Pallis, A. A., & Vitsounis, T. K. Port Economics, Policy and Management Content Classification and Survey. Transport Reviews, 31(4),

56 Journal of Economic Cooperation and Development 49 Reinert, K. A. and Roland-Holst, D. W. (1997). Social Accounting Matrices. In Applied Methods for Trade Policy Analysis: A Handbook, Francois, J. and Reinert, K., (eds.), pages Cambridge (UK): Cambridge University Press. Robinson, S., Cattaneo, A. and El-Said, M. (2001). Updating and estimating a social accounting matrix using cross entropy methods. Economic Systems Research, 13: Roland-Holst, D. (2008). Social Accounting Matrix for Pakistan, Berkeley: University of Berkeley. Sarkar, K. (2014). The port system in India: an analysis of the hierarchical changes and port performance ( ). Shoven, J. and Whalley, J. (1972). A general equilibrium calculation of the effects of differential taxation of income from capital in the U.S. Journal of Public Economics, 1: Shoven, J. and Whalley, J. (1984). Applied general equilibrium models of taxation and international trade: an introduction and survey. Journal of Economic Literature, 22(3): Shoven, J. and Whalley, J. (1992). Applying general equilibrium. New York: Cambridge University Press. UNCTAD (United Nations Conference on Trade and Development), (2012). The Review of Maritime Transport 2012, UNCTAD/RMT/2012, Geneva, December 2012, available at : PublicationWebflyer.aspx?publicationid=380 UNCTAD (2006). The Palestinian war-torn economy: aid, development and state formation. UNCTAD/GDS/APP/2006/1. New York and Geneva. UNCTAD (2004). Transit and Maritime Transport Facilitation for the Rehabilitation and Development of the Palestinian Economy. UNCTAD/GDS/APP/2003/1. New York and Geneva. Wilson, J. S., Mann, C. L., & Otsuki, T. (2003). Trade facilitation and economic development: A new approach to quantifying the impact. The World Bank Economic Review, 17(3),

57

58 Journal of Economic Cooperation and Development, 37, 2 (2016), Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Reza Ghazal 1 and Muhamed Zulkhibri 2 On the need to revisit the Islamic worldview and to develop a comprehensive measurement for Islamic economic performance and social progress based on Maqasid al-shari ah principle, this paper develops an Islamic Inclusive Growth Index (i-igi) - a composite index of growth inclusiveness - for the Organisation of Islamic Cooperation (OIC) member countries. It comprises of three sub-pillars to measure the countries performance based on three building blocks of inclusiveness: i) Islamic-adjusted economic growth incorporating Maqasid al-shari ah s principle; ii) performance of poverty reduction; and ii) performance of the society inequality and social inclusion. The results suggest that OIC member countries in the Central Asia, despite having a good performance on the three pillars of inclusiveness, is lagging behind in terms of its conformity with Maqasid al-shari ah principles. Contrarily, the low-income OIC member countries in the Sub-Saharan Africa, despite a low ranking on inclusiveness index is performing relatively betteraccording to Maqasid al- Shari ah principles. Disclaimers: All findings, interpretations, and conclusions are solely of the author s opinion and do not necessarily represent the views of the institutions. 1. Introduction The concept of inclusive growth has been debated substantially by policy-makers, researchers and practitioners in domestic and international political circles. The concept has inspired in many policies design and projects as well as impacted the views and the lives of many people in many countries. However, different interpretations of the concept of inclusive growth has been debated in the literature and policy 1 Corresponding author: Islamic Research and Training Institute (IRTI). khibri1974@yahoo.com. 2 Business and Management Sciences Dept., University of Kurdistan-Hawler (UKH).

59 52 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries circles. The concept is often used interchangeably with a suite of other terms, including broad-based growth, shared growth, and pro-poor growth. Nevertheless, by simple definition, inclusive growth focuses on both the pace and pattern of growth, a economic growth which is a necessary and crucial condition for poverty reduction. In Islam, development has specific dimensions of self and spiritual development, of the physical development, and of the development of society. The essential framework for individual and collective human progress are presented in the Qur an, and is, in turn, made operational by the traditions of the Prophet Muhammad (Peace be upon him). The core principles of Islam place great emphasis on social justice, inclusion, and sharing of resources among the society, rich and poor. From Islamic development perspective, Chapra (2008) classifies at least five essential elements for developing a human development and well-being model from the theory of Maqasid al-shari ah. Moreover, as classified by Imam Abu Hamid al-ghazali, the discourse on the meaning of the term Maqasid al-shari ah itself is focused more on the categories of the Maqasid. In order to measure the progress of development, creation of composite indicators to measure countries development performance has recently attracted the attentions of many scholars, policy-makers and institutions. Alternative measurements and concepts are being tested and increasingly used for policy-making at regional, national and international levels. However, the construction of these indicators have met with considerable criticism at least in three dimensions: (i) the ability of the index to give a suitable measure of development due to the quality and limitation of raw data (Stiglitz et al., 2009); (ii) the technical limits of the index, the so-called weighting and aggregation problem (Mazumbar, 2003); and iii) the need to take into account more information and other variables as well as accounted for what really matters for the society, i.e. sustainability, environment, happiness, religiosity (Nourry, 2008). The need for additional indicators to assess the economic and social progress has called for revisiting the worldview to take into consideration of a comprehensive approach to measure Islamic economics performance and social progress based on Maqasid al- Shari ah principles. Thus, the objective of the paper is to develop an

60 Journal of Economic Cooperation and Development 53 Islamic Inclusive Growth Index (i-igi), a composite index of growth inclusiveness based on Islamic principles for OIC member countries. It comprises of three sub-pillars to measure the countries performance based on three building blocks of inclusiveness: i) Islamic-adjusted economic growth incorporating Maqasid al-shari ah s framework; ii) Performance of poverty reduction; and iii) Performance of the society inequality and social inclusion. The paper is structured as follows. Section 2 provides brief concept of inclusive growth and Maqasid al-shari ah. Section 3 provides the related literature review on the construction of economic and social index based on Islamic principles. Section 4 describes the methodology and data for constructing the i-igi index. Section 5 analyses the results of the Inclusive Growth Index and Islamic Inclusive Growth Index (i-igi) and Section 6 finishes with the main conclusions. 2. Defining the Concept of Inclusive Growth and Maqasid al- Shari ah 2.1 Concept of Inclusive Growth Over the past few years, inclusive growth has become a very popular topic among development practitioners. Despite broad use of the term, there seem to be different views on the issue of what may be called inclusive growth. Often, the concept of inclusive growth is identified with pro-poor growth, which differs from inclusive growth (Klasen, 2010). The absolute definition of pro-poor growth suggests that what matters is the absolute rate at which the incomes of the poor are rising (Ravallion and Chen, 2003; DFID 2004; Ianchovichina and Lundstrom, 2009). Inclusive growth on the other hand, refers to the broader idea of a growth process that includes all segments of the society. It is about the extent to which economic growth creates opportunities for poor women and men through active participation in markets, communities and states. The Commission on Growth and Development (2008) notes that inclusiveness - a concept that encompasses equity, equality of opportunity, and protection in market and employment transitions - is an essential ingredient of any successful growth strategy. However,

61 54 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries attempts to measure inclusive growth have remained limited. Traditionally, poverty (or inequality) and economic growth analyses have been done separately. Recent work indicates that there may not be a trade-off between equity and efficiency as suggested by Okun (1975) and that it would be a big mistake to separate analyses of growth and income distribution (Berg and Ostry, 2011). Inclusive growth is important for very salient reasons: i) ethical considerations of equity and fairness, growth must be shared and should be inclusive across different segments of populations and regions.; ii) growth with persisting inequalities within a country may endanger social peace and further weaken other disadvantaged and vulnerable sections of population- resulting in a waste of vast human capital that could otherwise be used productively in creating economic outputs for sustainable growth; iii) continued inequalities in outcomes and access to opportunities in a country may result in civil unrest and violent backlash from people who are continually deprived, derailing a sustainable growth process. The concept of inclusive growth follows three policy pillars supported by good institutions and governance (Zhuang, 2010). As requirements to anchor inclusive growth strategy, the three pillars are aimed at high and sustained growth while ensuring that all members of the society benefit from growth: i) High, efficient, and sustained growth to create productive jobs and economic opportunity; ii) social inclusion ensures that all sections of the population, including those disadvantaged due to their individual circumstances, have equal opportunities; and iii) social safety nets are required to protect the chronically poor and to mitigate the risks and vulnerabilities associated with transitory livelihood shocks. Figure 1 depicts the three policy pillars of an inclusive growth strategy. In order to operationalize the concept of inclusive growth, ADB (2011) proposes a framework of inclusive growth indicators. The proposed framework is not rigidly prescriptive to take into account gaps in the availability of timely and comparable statistics of good quality for a majority of the developing economies in the region. It ultimately attempts to help achieve the goal of reducing poverty and inequality (income and non-income) by measuring income and non-income outcomes and their distribution across population, as allowed by available data. The framework presents 35 quantifiable indicators to

62 Journal of Economic Cooperation and Development 55 measure - outcomes and components of the three policy pillars of inclusive growth, as well as the components of good governance and strong institutions, which form the foundation of any inclusive growth strategy. 3 Figure 1: Policy Pillars of Inclusive Growth Inclusive Growth High, efficient and sustained growth to create productive jobs and economic opportunity Social safety nets to protect the chronically poor and to mitigate the risks of transitory livelihood shocks Social inclusion to ensure equal access to economic opportunity: investing in education, health and other social services to expand human capacity; eliminating market and institutional failures and social exclusion to level playing field Governance and Institutions Source: Adapted from Zhuang (2010) 3 Income - Proportion of population living below the national poverty line; Proportion of population living below US$2 a day at 2005 PPP$; Ratio of income/consumption of the top 20% to bottom 20%. Non-income - Average years of total schooling (youth and adults); Prevalence of underweight children under-five years of age; Under-five mortality rate.

63 56 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries 2.2 Concept of Maqasid al-shari ah Muslim scholars have developed a theory known as Maqasid al-shari ah (the objectives of the Shari ah). According to al-raysuni (1992), the theory can be traced back as far as the third century after Hijrah (9 th Century A.D.). Most of the discussions in the past literature on Maqasid centered on its legal dimensions. The pioneering works emanated from scholars such as al-shatibi (2004), al-ghazali (1901), al-juwayni (1979), and Ibn Ashur (1998). The application of al-maqasid in various disciplines including economics and finance has been gaining prominence in recent years. Among the leading economists who have written on the subject are Chapra (1985 and 2000), Siddiqi (2000), Ahmad (2000),, Atiyah (2008) and Al-Najjar (2008). The general objectives of Shari ah are commonly agreed by nearly all scholars of al-maqasid, which are Jalb al-masalih (to promote virtues) and Dar al-mafasid (to remove harm) (Ibn Ashur, 1998, 2006). Although there are some similarities in the principles (al-raysuni, 1992), some of these scholars differ in their classification of the specific objectives. Maqasid al-shari ah provides a comprehensive framework that can be used to measure development. This framework is commonly represented by the three levels of Maslahah and the five universal values or essential elements (al-daruriyat al-khams). The Maqasid al-shari ah is classified into five major categories according to Imam Al-Ghazali, a prominent and highly respected reformer in the eleventh century. He stated that the very objective of the Shari ah is to promote the well-being of the people, which lies in safeguarding of faith (Deen), self (Nafs), intellect (Aql), posterity (Nasl) and wealth (Mal). By promoting cooperation and mutual support within the family and society, it is also envisaged to spread the ethical values of compassion and guidance, establish justice, eliminate prejudice and alleviate hardship. Hence, there is a general consensus that the primary objectives of the Shari ah (al-maqasid) is to preserve the five essential elements, namely al-din (religion), al-nafs (life), al-aql (intellect), al- Nasl (family institution) and al-mal (wealth). Muslim scholars in the past, notably al-ghazali (1901), al-shatibi 2004) and Ibn Ashur (1998), besides having discussed the five essential elements of al-maqasid, they also discussed the levels of al-maqasid,

64 Journal of Economic Cooperation and Development 57 which termed as Maslahah. The three levels of al-maqasid are explained by earlier scholar in an ascending order as: Daruriyat (Necessity), Hajiyat (Complement) and Tahsiniyat (Embellishment). From a Maqasid perspective, the three levels of Maslahah represent a growth path and stages of development. The first level, Daruriyat represents the minimum level that every individual, institution and nation must achieve. This level embodies the preservation of the five universal values or essential elements mentioned above. The second level, Hajiyat provides avenues for the growth of the five elements and the third level, Tahsiniyat represents the stage for their sustainability. In recent years, the application of Maqasid al-shari ah has varied across several disciplines. Many of the contemporary applications have been in the legal circle where Maslahah has been used by Shari ah Advisory Councils of many Islamic banks as the basis for issuing fatwas. Although the parameters for the application of Maslahah have been defined by several Shari ah bodies and academies, these applications have hardly been extended beyond the fatwa sessions. On the other hands, in economics and finance, several literature on Maqasid al- Shari ah is essentially theoretical in nature, for example Chapra (1985; 2008), Siddiqi (2000), Hasan (2004) and Ahmad (2000). Nevertheless, there are few studies that have attempted to operationalize Maqasid al- Shari ah principles in the areas of economics and finance. 3. Related Literature on Measuring Islamic Economy and Social Progress There have been numerous attempts to construct indicators or composite indices in the mainstream economic to measure economic welfare, sustainable development quality of life and well-being. Among others are Human Development Index (HDI); the Quality of Life Index (QLI); the Happy Planet Index (HPI); Gross National Happiness (GHI); the Legatum Prosperity Index (LPI); the Multidimensional Poverty Index (MPI); Genuine Progress Indicator (GPI); and the Meaning of Life Index (MLI). HDI is a composite index widely used by international organizations to evaluate and rank countries in terms of three main indicators of economic and social welfare - income, health and education attainments.

65 58 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries In the Islamic literature, however, very few have attempted to integrate the religious aspect of development that reflects spiritual well-being, which has been recognized as an important component in the construction of the indices. Several indices have developed a methodological framework, while others have constructed the indices as potential alternatives to more commonly used indices based on conventional views of economic and social progress. However, there are few studies in the literature that develop Islamic indices among others are: i) the Ethics-Augmented Human Development Index (E-HDI) by Dar (2004); ii) the Islamic Human Development Index (I-HDI) by Anto (2009); iii) the Integrated Development Index (I-Dex) by Amin et al. (2013) and iv) the Islamicity Index (I 2 ) by Rehman and Askari (2010). Dar (2004) proposes an ethics-augmented human development index (E- HDI) as a new indicator of socio-economic change and development. The E-HDI incorporates freedom, faith, environmental concerns and the institution of family in the HDI and ranks countries of the world accordingly. The study constructs the E-HDI based on Maqasid al- Shari ah and it conceptualizes social change and development for all countries. However, the ranking of countries in the study is based on the ordinal measure using the Borda Rule instead of on the actual values of E-HDI computed for all countries. Hence, there is a need to develop a comprehensive development index based on Maqasid al-shari ah that can actually be computed for the ranking of countries level of development. Anto (2009) attempts to develop Islamic Human Development Index (I- HDI). The paper argues that the existing HDI is not fully compatible and sufficient in measuring human development from Islamic perspective. The index is comprised of Material Welfare Index (MWI) and Nonmaterial Welfare Index (NWI), representing the five basic needs in Maqasid al-shari ah. It also includes the Freedom Index and the Environment Index. The findings show that the rank composition between I-HDI and HDI is slightly different. A number of countries enjoy a better rank in I-HDI compared with HDI, while several countries suffer a marked deterioration of rank. The high score group in I-HDI is still dominated mostly by countries in the MENA region, whileat the bottom group is still dominated by countries in Africa region.

66 Journal of Economic Cooperation and Development 59 Rehman and Askari (2010) propose to measure the degree of religiosity of Islamic countries. It uses four sub-indices namely the Economic Islamicity Index (EI²), the Legal and Governance Islamicity Index (LGI²), the Human and Political Rights Islamicity Index (HPI²), and the International Relations Islamicity Index (IRI²) to measure government s adherence to Islamic principles in economics; legal integrity and governance environment; degree of civil and political rights; and relationship with the global community. The study shows that Islamic countries are not as Islamic as one might expect, at least in the realm of economics.; Instead it appears that the most developed countries tend to place higher on this Islamic Economic Index. Amin et al. (2013) proposes to develop an integrated Islamic development framework and index based on, and representing the Maqasid al-shari ah or noble objectives of the Shari ah for OIC and non-oic countries. The Integrated Development Framework is based on Maqasid al-shari ah principles and follows the works by al-ghazali and Abu Zaharah. Based on content analysis, the operational definitions of each component of Maqasid al-shari ah are derived, and the dimensions are identified based on the scope of the definitions. The elements have been selected for each dimension based on the relevant existing indicators. However, data availability poses a major constraint in selecting relevant indicators. Hence, the study does not proceed further to construct the actual index. On the other hands, several studies attempt to incorporate the concept of Maqasid al-shari ah approach into the field of Islamic finance. Dusuki (2007) and Hamdan (2014) attempt to relate Maqasid al-shari ah to corporate social responsibility (CSR) and argue that the concept of CSR is not alien to Islam and any corporation that claims to follow Shari`ahbased principles should naturally practice CSR, as it enshrines Islam s true spirit. Similarly, Hameed et al. (2005) develop Islamicity Disclosure Index to measure and compare the performances of two Islamic banks. Mustafa et al. (2008) refine the research and develop a quantitative performance measure for Islamic banking from the theory of al-maqasid. While Larbani and Mohammed (2012), develop a decision-making tool based on Maqasid al-shari ah and their levels of Maslahah for managers of firms to use in allocating their investible resources to vital sectors of the economy.

67 60 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Bedoui (2012) proposes a framework to manage ethical (financial and non-financial) performances based on the concept of Maqasid al- Shari ah. The paper argues that considering the fact that business ethics is one of the most needed in the market nowadays, this framework is a solution for Islamic financial institutions and can be used for other organizations looking for ethical investments. The proposed framework is considered akin to a rating model to benchmark organizations based on the five pillars of Maqasid al-shari ah. The graphical method used in the paper assigns a score the performance to each axis in which represent one objective of the five dimensions of Maqasid al-shari ah. 4.Data and Methodology 4.1 Data Descriptions In this section, we provide data descriptions used in the construction of Islamic-Inclusive Growth Index (hereafter i-igi) for OIC member countries. Table 1 presents the main indicators of i-igi, which are the Gini index, poverty index, GDP growth and ratio of non-interest income to total assets (see Appendix 1 for details) used in the index computation. According to Table 1, Comoros is the most unequal country in our sample based on the inequality indicator, with the Gini index at 64.3 percent, while Kazakhstan is the most equal country in terms of income distribution with the Gini index at 29 percent.

68 Journal of Economic Cooperation and Development 61 Table 1: OIC Member Countries Based on Three Pillars of Inclusiveness (average ) Poverty Headcount Ratio at $1.25 a day (PPP) (% of population) GINI index GDP Growth (% annual) Non-interest Income as a Ratio of Total Assets Albania Azerbaijan Benin N.A Burkina Faso Bangladesh Cote d'ivoire N.A Comoros N.A Djibouti N.A Egypt Gabon Guinea N.A Gambia Guinea-Bissau NA Indonesia Iran N.A Iraq Jordan Kazakhstan Kyrgyz N.A Morocco Maldives Mali N.A Mozambique Mauritania Malaysia Niger Nigeria Pakistan Sudan Senegal Sierra Leone N.A Syria Tajikistan NA Tunisia Turkey Uganda Yemen Average Note: N.A stands for not available. Source: World Bank Group Database; Bankscope

69 62 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries In terms of poverty, Nigeria has the highest percentage of population under the poverty line ($1.25 a day) at 68 percent, followed by Mozambique at 59 percent, while Azerbaijan has the smallest percentage of population under poverty line ($1.25 a day) at 0.4 percent, followed by Albania at 0.6 percent. In term of economic growth, Azerbaijan has experienced the higher growth of 10 percent on average over , while Cote d Ivoire has experienced the lowest growth at 1.1 percent. On non-interest income, interestingly, Maldives has the highest non-interest income as percentage of total assets at 11.3 percent, while Azerbaijan has the least percentage of non-interest income over total assets at less than percent. 4.2 Methodology There have been some methodologies in the literature suggested by different scholars and institutions such as Asian Development Bank 4 and Africa Development Bank to measure inclusive growth index. However, most of the methodologies require extensive data requirements, which can explain the reason why none has numerically developed inclusive development index in the literature to date. Having known these limitations on the extensive data requirements, we introduce an innovative simple Islamic inclusive growth index using three most important components that can be derived from the definition of growth inclusiveness. According to the classical definition of inclusive development (or growth), a growth is called inclusive if all the society will benefit from the growth process, especially the poor segment of the society. Alternatively, inclusiveness simply refers to the broader idea of a growth process that includes all segments of the society. In this context, the i- IGI is a summary measure of inclusive growth in OIC member countries. It intends to measure country s average achievements in three basic dimensions of inclusiveness: economic growth adjusted to take into account for Shari ah s aspects, country s performance on reducing proportion of population under poverty line, and performance in reducing the level of inequality. The i-igi is the geometric mean of normalized indices measuring achievements in each dimension. 4 See Appendix 2 for definition of indicators.

70 Journal of Economic Cooperation and Development 63 Using this definition as a starting point, our inclusive growth index comprises of three components; i) GDP growth representing the growth component, ii) inequality proxy by the Gini index to reflect the extent of inequality among the society, and iii) the poverty index representing the proportion of population under the poverty line at $1.25 a day (measured at PPP). Another innovation in the paper is to take into account the Maqasid al-shari ah s role in advancing growth inclusiveness. Since the aims on reducing poverty and inequality are universal across the countries irrespective of Muslim or non-muslim countries, we make adjustments to the growth dimension of the index to reflect the Islamic principle of growth by using the ratio of average non-interest income to total assets. In order to control for the cyclical effects in the components of the index (especially for the global financial crisis) and a way of preventing the under-representativeness or over-representativeness of a specific year, we use an average of those indicators over period. In addition, since an economy with higher equality, lower poverty and higher growth is the chief target and main objective of these countries, we need to transform the poverty and Gini indexes into comparatively consistent numbers. Figure 1 presents the flowchart in term of steps taken for computing the i-igi. In this regard, we make two transformations - inverse 5 and subtract the indicators from 100) - on the poverty and inequality data for the purpose of consistency and usability of the i-igi. Using two ways transformations also work as a robustness check for the final inclusive growth indices. In computing the index, we use two well-known but different methods: i) MinMax method as suggested in Human Development Index (HDI) report, and ii) standardization method. Under the MinMax method, the first step is to create sub-indices for each dimension. Minimum and maximum values (known as goalposts) need to be set in order to transform the indicators into indices between 0 and 1. Because the geometric mean is used for aggregation, the maximum value does not affect the relative comparison (in percentage term) between any two 5 Inversing means that we transform poverty and Gini indexes as follows: 1/Poverty ratio, and 1/Gini Index. In the same way, subtracting from 100 means: 100-Poverty Index, and 100-Gini Index.

71 64 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries countries or over period of time (HDI 2013). The maximum values are set to the actual observed maximum values of the indicators on a crosssection basis, average of from the countries. The minimum values are set at 32 for poverty headcount ratio at US$1.25 a day (PPP) using the inverse transformation (0.015 using second transformation), while for GINI index variables at 35.7 (0.016 using second transformation) and for GDP growth (percent annual) at 1.1. Hence, progress is measured against minimum levels that a society needs to survive over time. Following HDI methodology and having defined the minimum and maximum values, the sub-indices are calculated as follows: actual value min imum value Dimension Index (1) max imum value min imum value

72 Journal of Economic Cooperation and Development 65 Figure 2: Calculating the Islamic Inclusive Growth Index: A Graphical Flowcharts Reducing Inequality (GINI index) Inequality- Reduction Index Dimensions Economic Growth Poverty Reduction Inequality Reduction Islamic Inclusive Growth Index Indicators Economic Growth Adjusted by Absence of Riba Reduction of Poverty under Poverty Line ($1.25 per day) Dimension Index Growth Index Poverty- Reduction Index Source: Author own illustration based on HDI report. Islamic Inclusive Growth Index

73 66 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries For each dimension, Eq.1 is applied to each of the subcomponents and a geometric mean of the index is created. This is equivalent to applying Eq. 1 directly to the geometric mean of the sub-components. Because each dimension index is a proxy for capabilities in the corresponding dimension, the transformation function from income to capabilities is likely to be concave (Anand and Sen, 2000). Therefore, the i-igi is the geometric mean of the all dimension indices as follows: 1/ 3 1/ 3 1/ 3 i IGI ( I * I * ) (2) 1 2 I 3 where 3 represents the number of sub-indices, and I represent the subindices. Eq.2 embodies imperfect substitutability across all i-igi dimensions. It is thus addressed one of the most serious criticisms of the linear aggregation formula, which allowed for perfect substitution across dimensions. Some substitutability is inherent in the definition of any index that increases with the values of its components (HDI 2013). Under the standardization method the same methodology is applied except that instead of defining the goalposts (minimum and maximum values), each sub-index is standardized by the conventional standardization methodology. In other words, each sub-index is subtracted from the average of that sub-index for the whole group of countries and then it is divided by the standard deviation of the subindex. In addition, under standardization method since some negative sub-indexes will be appeared, we shift the distribution of the sub-indices merely to get positive numbers. It is worth mentioning that shifting the distribution will not affect the overall results. 5. Empirical Findings Table 2 presents the ranking of the OIC member countries in term of inclusive growth indices and Islamic inclusive growth indices. 6 The ranking is based on both MinMax and standardization methods. In general, the results are different from the findings of existing models in the literature (Rehman and Askari, 2010; Anto, 2009; Dar, 2004), but pointed to the same conclusion that majority of the OIC member 6 SeeAppendix 3 for the values correspondent to the rankings.

74 Journal of Economic Cooperation and Development 67 countries are not conforming to Maqasid al-shari ah and Islamic principles, at least in the realm of economic and development. According to the ranking, over period, Azerbaijan has ranked first on all growth indices using various methodologies due to its strong performance on growth dimensions during the period as well as very low proportion of population under the poverty line. However, when it comes to the Islamic Inclusive Growth Index (i-igi), due to extremely low level of non-interest income of this country (compared to its assets), its ranking drops significantly to 26 th place except for the case where MaxMin is built based on the second method of transformation. The same findings can be observed in the case of Kazakhstan. Having high growth and low poverty have helped the country to rank 2 nd based on the inclusive growth index, while it drops to rank around 20 th after taking into account the Maqasid al-shari ah aspects of the index (Table 3). On the other extreme, Nigeria ranks 26 th in terms of inclusive growth index due to its high levels of poverty and to some degree relatively high inequality, followed by Mozambique ranked at 25 th.

75 68 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Table 2: Ranking of Inclusive Growth and Islamic Inclusive Growth Index for OIC Member Countries IGMM T1 iigm MT1 IGMM T2 iigm MT2 IGSTD T1 iigst DT1 IGSTD T2 iigst DT2 Azerbaijan Kazakhstan Iraq Maldives Egypt Jordan Albania Syria Indonesia Morocco Turkey Tunisia Malaysia Gabon Sudan Bangladesh Pakistan Uganda Senegal Niger Mauritania Burkina Faso Yemen Gambia Mozambique Nigeria Source: Authors calculations Note: IGMMT1 refers to Inclusive Growth based on MaxMin method using the Transformation 1, and IGSTDT1 refers to Inclusive Growth based on standardization method using Transformation 1. i at the beginning refers to Islamic.

76 Journal of Economic Cooperation and Development 69 In general, interestingly the OIC member countries in Sub-Saharan Africa improves in terms of the ranking when it comes to i-igi. The main reason can mainly due to low levels of the overall bank assets. It can also be explained by the existence of foreign-banks that have been established by others OIC member countries particularly from the Gulf region. In addition, there are some countries, particularly Tunisia and Turkey for which a consistent pattern of ranking is observed where their rankings are remained almost similar for both IGI and i-igi indices.

77 Table 3: Inclusive Growth and Islamic Inclusive Growth Index for OIC Member Countries IGMMT1 iigmmt1 IGMMT2 iigmmt2 IGSTDT1 iigstdt1 IGSTDT2 iigstdt2 Azerbaijan Kazakhstan Iraq Maldives Egypt Jordan Albania Syria Indonesia Morocco Turkey Tunisia Malaysia Gabon Sudan Bangladesh Pakistan Uganda Senegal Niger Mauritania Burkina Faso Yemen Gambia Mozambique Nigeria Source: Authors calculations IGMMT1 refers to Inclusive Growth based on MinMax method using the Transformation 1. IGSTDT1 refers to Inclusive Growth based on standardization method using Transformation 1. i at the beginning refers to Islamic. 70 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries

78 Journal of Economic Cooperation and Development Conclusions This paper is the first step towards establishing and constructing an Inclusive Growth Index and Islamic Inclusive Growth Index (i-igi) based on Maqasid al-shari ah to gauge the performance of the OIC member countries on inclusive-related issues including economic growth, poverty and inequality. Moreover, this framework is one of the solution for OIC member countries to measure and benchmark their overall performance based on the Maqasid al-shari ah principles. In this study, different transformations and methods have been applied to ensure the data requirements for inclusive index calculation is consistent and at the same time can be used as a robustness check for the final results. The overall observation is that despite the significant performance of Central Asian OIC member countries on various pillars of inclusiveness, their performances in terms of conforming with the Maqasid al-shari ah (specifically riba-free economy) are considerably low. On the other hand, the low-growth of the OIC member countries in Sub-Saharan Africa have a promising performance based on the Shari ah-compliance aspects. The findings are different from the findings of existing models in the literature, but pointed to the similar conclusions that OIC member countries are not fully conforming to Maqasid al-shari ah principles, at least in the realm of economic and development. For future research, the most vital step in constructing a more accurate and comprehensive inclusive index based on the Maqasid al-shari ah principle is to compile more data covering the important dimensions of the principles. Furthermore, in order to add further research contribution, it is also vital to provide an empirical assessment and discussion on the i-igi s determinants for OIC member countries. This future research will answer the usefulness and relevance of the Islamic inclusive growth index (i-igi) in explaining the variation of economic development in the OIC member countries.

79 72 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries References ADB (2011), Key Indicators for Asia and the Pacific 2011: Framework of Inclusive Growth Indicators, special supplement. Manila: Asian Development Bank. Ahmad, K. (2000), Islamic Finance and Banking: The Challenge and Prospects, Review of Islamic Economics, 9, Al-Ghazali, A.H. (1901), Al-Mustasfa Min ilm Al-Usul. (1 st ed.). Al- Matba ah al-amiriyyah, Egypt. Al-Juwayni, A. (1979), Al-Burhan Fi Usul al-fiqh (ed.). Abdul Azim al-dib. Dar al-ansar, Cairo. Al-Najjar, U. A. (2008), Maqasid al-shari ah bi ab adin jadidah. Dar al- Gharb al-islami, Beirut. Al-Raysuni, A. (1992), Nazariyat Al-Maqid Inda Al-Imam Al-Shatibi. Herndon, International Institute of Islamic Thought. Al-Raysuni, A. (2005), Imam Al-Shatibi's Theory of the Higher Objectives and Intents of Islamic Law. International Institute of Islamic Thought, Herndon, USA. Al Sari, A. M. (n.d.), Basic Human Needs the Islamic Theory, unpublished thesis. Al-Shatibi, A. I. (2004), Al-Muwafaqat Fi Usul al-shari ah. Dar al- Kutub al- Ilmiyah, Beirut. Amin, R.M., Yusof, S.M., Haneef, M.A., Muhammad, M.O. and Oziev, G. (2013), The Integrated Development Index (i-dex): A New Comprehensive Approach to Measuring Human Development, 9 th International Conference on Islamic Economics and Finance, 9-10 September Anto, M. H. (2009), Introducing An Islamic Human Development Index (I-HDI) to Measure Development in OIC Countries, Islamic Economic Studies, 19(2),

80 Journal of Economic Cooperation and Development 73 Atiyah, G. E. (2008), Nahw Taf eel Maqasid al-shari ah (Towards Revitalizing al-maqasid). International Institute of Islamic Thought, Herndon, USA. Bedoui, M.H. (2012) Shari a-based Ethical Performance Measurement Framework, CEFN Working Paper, University of Paris. Berg, A., and Ostry, J.D. (2011), Inequality and Unsustainable Growth: Two Sides of the Same Coin? IMF Staff Discussion Note 11/08, Washington: International Monetary Fund. Chapra, M. U. (1985), Towards a Just Monetary System.Leicester: The Islamic Foundation. Chapra, M. U. (2000), The Future of Economics: An Islamic Perspective. Leicester: The Islamic Foundation. Chapra, M. U. (2008), The Islamic vision of development in the light of Maqasid al-shari ah. Islamic Research and Training Institute, Jeddah: Islamic Development Bank. Commission on Growth and Development (2008), The Growth Report: Strategies for Sustained Growth and Inclusive Development, Washington, DC: World Bank. Dar, H.A. (2004), On Making Human Development More Humane, International Journal of Social Economics, 31 (11/12), DFID (2004), What is Pro-Poor Growth and Why Do We Need to Know?, Pro-Poor Growth Briefing Note 1, U.K: Department for International Development. Dusuki, A. W., and Irwani, N. (2007), Maqasid As-Shari ah, Significance and Corporate Social Reponsibility, The American Journal of Islamic Social Sciences, 24(1), Hamdan, M. (2014). Corporate Social Responsibility of Islamic Banks in Brunei Darussalam, K.C.P Low et al. (eds.) Corporate Social

81 74 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Responsibility in Asia CSR, Sustainability, Ethics and Governance, Hameed, S., Pramano, S., Bakhtiar, A. and Bahrom, N. (2004). Alternative Performance Measures for Islamic banks. 2nd International Conference on Administrative Sciences. King Fahd University of Petroleum and Minerals, Saudi Arabia, April 2004 HDI (2013) Human Development Report 2013, New York: United Nations Development Programme. Ianchovichina, E. and Lundstrom, S. (2009). Inclusive growth analytics framework and application, Policy Research Working Paper 4851, Washington, DC: World Bank. Ibn Ashur, M. al-tahir. (1998). Maqasid al-shari ah al-islamiyyah, ed., al-misawi, Muhammad al-tahir, al-basa ir, Kuala Lumpur. Ibn Ashur, M. al-tahir (2006). Treatise on Maqasid al-shari ah. International Institute of Islamic Thought, Herndon, USA. Klasen, S. (2010), Measuring and Monitoring Inclusive Growth: Multiple Definitions, Open Questions, and Some Constructive Proposals,, ADB Sustainable Development Working Paper Series No. 12, Manila: Asian Development Bank. Larbani, M. and Mohammed, M.O. (2011), Decision Making Tools for Resource Allocation Based on Maqasid Al- Shariah, Islamic Economic Studies, 19(2), Mazumbar, K. (2003), A New Approach to Human Development Index, Review of Social Economy, 61(4), Mustafa, O.M, Razak, D. and Fauziah M. T. (2008), The Performance of Islamic Banking Based on The Maqasid Framework, IIUM International Accounting Conference (INTAC IV). PutraJaya Marroitt, 25 June 2008.

82 Journal of Economic Cooperation and Development 75 Nourry, M. (2008), Measuring Sustainable Development: Some Empirical Evidence for France from Eight Alternative Indicators, Ecological Economics, 67(3), Okun, A.M. (1975), Equality and Efficiency: the Big Trade-Off, Washington: Brookings Institution Press. Ravallion, M. and Chen, S. (2003), Measuring Pro-Poor Growth, Economics Letters 78, Rehman, S. S., and Askari, H. (2010), How Islamic are Islamic Countries? Global Economy Journal, 10(2), Siddiqi, M.N. (2000), Islamic Banks: Concept, Precept and Prospects, Review of Islamic Economics, (9), Stiglitz, J., Sen, A. and Fitoussi, J.P. (2009), Report by the Commission on the Measurement of Economic Performance and Social Progress. International Commission on Measurement of Economic Performance and Social Progress, Paris. The World Bank Group, database at data.worldbank.org. Zhuang, J., and Ali, I. (2010) Poverty, Inequality, and Inclusive Growth in Asia, in Zhuang, J. (ed) Poverty, Inequality, and Inclusive Growth in Asia: Measurement, Policy Issues, and Country Studies. Manila: Asian Development Bank.

83 Albania Azerbaijan Benin Burkina Faso Bangladesh Cote d'ivoire Comoros Djibouti Egypt Gabon Guinea Gambia, The Guinea-Bissau Indonesia Iran Iraq Jordan Kazakhstan Kyrgyz Republic Morocco Maldives Mali Mozambique Mauritania Malaysia Niger Nigeria Pakistan Sudan Senegal Sierra Leone Syria Tajikistan Tunisia Turkey Uganda Yemen Albania Azerbaijan Benin Burkina Faso Bangladesh Cote d'ivoire Comoros Djibouti Egypt Gabon Guinea Gambia, The Guinea-Bissau Indonesia Iran Iraq Jordan Kazakhstan Kyrgyz Republic Morocco Maldives Mali Mozambique Mauritania Malaysia Niger Nigeria Pakistan Sudan Senegal Sierra Leone Syria Tajikistan Tunisia Turkey Uganda Yemen 76 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Appendix 1: Gini Index, Poverty Index, and GDP Growth Gini (Inequality) Index ( Average) 0 Poverty Index ( average)

84 Albania Azerbaijan Benin Burkina Faso Bangladesh Cote d'ivoire Comoros Djibouti Egypt Gabon Guinea Gambia, The Guinea-Bissau Indonesia Iran Iraq Jordan Kazakhstan Kyrgyz Republic Morocco Maldives Mali Mozambique Mauritania Malaysia Niger Nigeria Pakistan Sudan Senegal Sierra Leone Syria Tajikistan Tunisia Turkey Uganda Yemen Journal of Economic Cooperation and Development 77 GDP Growth ( Average) Appendix 2: Comprehensive Framework of Islamic Inclusive Growth Index: Pillars and Indicators Dimension Indicator 1. Income poverty and inequality 1.1. Income 1. Proportion of population living below the national poverty line 2. Proportion of population living below $2 a day at 2005 PPP$ 3. Ratio of income or consumption of the top 20% to bottom 20% 1.2 Non-income 4. Average years of total schooling (youth and adults) 5. Prevalence of underweight children under five years of age 6. Under-five mortality rate 2. Pillar 1: Growth and Expansion of Economic Opportunities 2.1 Economic Growth and 7. Growth rate of GDP per capita at PPP (constant 2005 PPP$) Employment 8. Growth rate of average per capita income/consumption 2005 PPP$ (lowest quintile, highest quintile, and total) 9. Employment rate 10. Elasticity of total employment to total GDP (employment elasticity) 11. Number of own-account and contributing family workers per 100 wage and salaried workers 2.2. Key Infrastructure Endowments 12. Per capita consumption of electricity 13. Percentage of paved roads 14. Number of cellular phone subscriptions per 100 people 15. Depositors with other depository corporations per 1,000 adults

85 78 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries 3. Pillar 2: Social Inclusion to Ensure Equal Access to Economic Opportunity 3.1 Access and Inputs to 16. School life expectancy (primary to tertiary) Education and Health 17. Pupil-teacher ratio (primary) 18. Diphtheria, tetanus toxoid, and pertussis (DTP3) immunization coverage among 1-year-olds 19. Physicians, nurses, and midwives per 10,000 population 20. Government expenditure on education as percentage of total government expenditure 21. Government expenditure on health as a percentage of total government expenditure 3.2 Access to Basic Infrastructure Utilities and Services 3.3. Gender Equality and Opportunity 22. Percentage of population with access to electricity 23. Share of population using solid fuels for cooking 24. Percentage of population using improved drinking water sources 25. Percentage of population using improved sanitation facilities 26. Gender parity in primary, secondary, and tertiary education 27. Antenatal care coverage (at least one visit) 28. Gender parity in labor force participation 29. Percentage of seats held by women in national parliament 4. Pillar 3: Social Safety Nets 4.1 Social Safety Nets 30. Social protection and labor rating 31. Social security expenditure on health as a percentage of government expenditure on health 32. Government expenditure on social security and welfare as a percentage of total government expenditure 5. Pillar 4: Maqasid al-shari ah Principles 5.1 Hifdh al-din 33. Percentage of total expenditure in religious education 34. Corruption index 35. Zakat estimates 5.2 Hifdh al-aql 36. Primary enrollment rate 37. Secondary enrollment rate 38. Literacy rate 5.3 Hifdh al-nafs 39. Universal Human Rights Index 40. Incidence of crime 5.4 Hifdh al-mal 41. International Property Rights Index 5.5 Hifdh al-nasl 42. Divorce rate 43. Fertility rate 6. Good Governance and Institutions 6.1 Good Governance and 33. Voice and accountability Institutions 34. Government effectiveness Source: Adapted from ADB (2011)

86 Journal of Economic Cooperation and Development 79 Appendix 3. IGI and i-igi: MinMax and Standardization Methods Countries' Inclusive Growth (IG) and Islamic-Inclusive Growth (IIG) Values-MinMax Method (Transfromation 2) IGMMT2 IIGMMT2 Countries' Inclusive Growth (IG) and Islamic-Inclusive Growth (IIG) Values-MinMax Method (Trnasformation 1) IGMMT1 IIGMMT1

87 80 Islamic Inclusive Growth Index for the Organisation of Islamic Cooperation (OIC) Member Countries Countries' Inclusive Growth (IG) and Islamic-Inclusive Growth (IIG) Values-Standardization Method (Transfromation 1) IGSTD2 IIGSTD2 Countries' Inclusive Growth (IG) and Islamic-Inclusive Growth (IIG) Values-Standardization Method (Transfromation 2) IGSTD1 IIGSTD1

88 Journal of Economic Cooperation and Development, 37, 2 (2016), Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Ebaidalla Mahjoub Ebaidalla 1 This paper examines the determinants of youth unemployment in OIC countries during the period The study used a dynamic panel data method for a sample of 32 OIC countries, focusing on the impact of economic, demographic and institutional factors. The empirical results show that youth unemployment in OIC countries is influenced by economic environment measured by GDP growth, inflation and domestic investment. Fertility rate is found to be one of the significant factors behind the high rate of youth unemployment. Moreover, the results pointed out that bureaucracy quality has a negative impact on youth unemployment in OIC member countries. To improve the employability of young workers, the paper recommended that the economic environment in OIC countries should be ameliorated through effective fiscal, monetary and trade policies. In addition, public and private investment need to be expanded to provide job opportunities for young workers. Moreover, serious efforts should be made at vocational and technical education aiming at developing skills and experiences of young people. 1. Introduction Youth unemployment has been considered as one of the critical problems in developed and developing countries. Indeed, in all countries and regions young people (population ages between 15 and 24) have fewer opportunities to access the labor markets compared to the active adult population. Recent statistics show that young people in the World are estimated to be about 1.8 billion, constitutes one quarter of the world population (Population Reference Bureau, 2013). However, about 75 million of them are unemployed and most of youth workers engage in hazardous jobs and informal sector, and their transition to decent work is slow and difficult (International Labor Organization (ILO), 2013). 1 Assistant Professor, Department of Economics, University of Khartoum, Sudan. ebaidallamahjoub@yahoo.com

89 82 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis In the OIC countries 2, the youth accounts for about more than 30% of the total population (World Bank, 2014). Although, young population could be a potential demographic asset for the OIC countries, youth employment is a critical challenge that faces all the OIC states 3. In addition, since most of OIC countries located in the Middle East and North Africa (MENA), the recent statistics shows that this region holds the highest rate of youth unemployment in the World. For example, according to International Labor Organization s (ILO) statistics of 2011, the youth unemployment rate in the MENA region was about 24%, which is approximately twice the global average 4. Moreover, the other OIC countries that situated in Asia and Sub Saharan Africa also suffer from high youth unemployment. Youth unemployment may have multiple unfavourable economic, social and political effects. From the economic perspective, unemployment among youth means unutilised potential labor force, and then has a negative impact on production, economic growth and development. On the social front, failure of young persons to get a job and enjoy employment benefits may lead to frustration and social exclusion. From the political perspective, youth unemployment stimulates protests and demonstrations and, in turn, undermines the political stability. Therefore, tackling the problem of youth unemployment needs an indepth analysis of its determinants. This would reveal the effect of some factors that may be subject to the control of policy maker in OIC countries. 2 OIC countries include, Afghanistan, Albania, Algeria, Azerbaijan, Bahrain, Bangladesh, Benin, Bosnia and Herzegovina, Brunei, Burkina Faso, Cameroon, Central African Republic, Chad, Comoros, Côte d'ivoire, Djibouti, Egypt, Gabon, Gambia, Guinea, Guinea-Bissau, Guyana, Indonesia, Iran, Iraq, Jordan, Kazakhstan, Kuwait, Kyrgyzstan, Lebanon, Libya, Malaysia, Maldives, Mali, Mauritania, Morocco, Mozambique, Niger, Nigeria, Northern Cyprus, Oman, Pakistan, Palestinian territories, Qatar, Russia, Saudi Arabia, Senegal, Sierra Leone, Somalia, Sudan, Suriname, Syria, Tajikistan, Thailand, Togo, Tunisia, Turkey, Turkmenistan, Uganda, United Arab Emirates, Uzbekistan, and Yemen. 3 The youth unemployment rate is the proportion of youth (persons aged 15-24) who are unemployed. 4 In fact, the existent statistics on unemployment do not reflect the actual situation, since Arab countries lack systematic labor market surveys. Therefore, data on unemployment underestimates the problem of youth unemployment in the region.

90 Journal of Economic Cooperation and Development 83 This paper aims to identify the determinants of youth unemployment in the OIC countries, during last two decades. Specifically, the paper addresses the following questions: Why is youth unemployment so high in OIC countries? And What is the role of economic, demographic factors and institutional quality in explaining youth unemployment? Therefore, this paper would contribute to existing literature on youth unemployment in the OIC countries, as there is a dearth of studies on this issue. The contribution of this paper is to provide some policy recommendations that aim to improve the employability of young people in OIC countries. The paper would also contribute to the existing literature on youth unemployment in OIC countries, as to the best of our knowledge there are no empirical studies examine the causes of youth unemployment using OIC data. Moreover, the study investigates the effect of institutions like bureaucratic quality on youth employment in OIC countries. The paper is organised as follows: the next section outlines some stylised facts about youth unemployment and labor markets in the OIC countries. Section three reviews the empirical literature on the determinants of youth unemployment. Section four outlines methodology and data used in the study. While section five presents the empirical results, section six conclude with some policy implications. 2. Youth Unemployment in the OIC countries: Some Stylized Facts Before identifying the determinants of youth unemployment in OIC member countries, it is useful to overview the situation of unemployment in OIC countries. Recent statistics show that OIC countries recorded significantly higher unemployment rates compared to the world, developed and other developing countries (SESRIC, 2012). Figure 1 below presents the unemployment rates in some OIC countries during 2012.

91 84 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Figure 1: Unemployment Rate in OIC Countries (2010) (% of Labor Force) Source: SESRIC BASEIND Database (2014). The figure indicates that unemployment rate in OIC countries varies from one country to another. In Gabon, Palestine and Mauritania the rate of unemployment is estimated to be more than 20%. However, in the other countries except Qatar the rate of unemployment is ranging from 7 percent to 10 percent, which is equivalent to unemployment rate in most of developing countries. Regarding the youth unemployment, OIC countries host high rate of young unemployment, which is estimated around 17% in 2012 (World Bank, 2014). The problem of unemployment may be responsible for the unfavourable economic performance and low development outcomes as well as political instability in some OIC countries. As indicated in Figure 2, OIC countries have the highest rates of youth unemployment, among other regions, compared to low and high income countries. The Figure also show that youth unemployment in the OIC countries is persistent over the last two decades, estimated at about 17 percent, which is above the World rate. In other low income countries youth unemployment rates do not exceed 13%.

92 Journal of Economic Cooperation and Development 85 Figure 2: Youth Unemployment Rate by Region ( ) Source: World Bank Development Indicator (2014) and SESRIC BASEIND Database. Regarding the distribution of youth unemployment in OIC countries, the statistics show that there is a variation in the level of youth unemployment from one country to another. As indicated in Figure 3, in 2010 the youth unemployment rates vary from 1.3% in Qatar to about 45.2% in Palestine. This situation of high unemployment rate is attributed to the low absorption power of labor markets, besides low experience of young workers and skills mismatch. Indeed, in most of OIC countries, particularly the countries of Middle East and Northern Africa (MENA) the ratio of youth is considered among the highest in the world. This situation has attributed to many factors including the high unemployment ratio in the youth cohort compared to the total population besides the weak absorptive capacity of these economies (Ebaidalla, 2014).

93 86 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Figure 3: Youth Unemployment Rate in OIC Countries (2010) (% of labor force aged 15-24) Source: World Bank Development Indicator (2014) In accordance with the distribution of youth unemployment by gender, Figure 4 below presents the youth unemployment rate according to gender in some OIC countries. As in other developing regions, females suffer over proportionally as reflected approximately by the 2:3 ratio of male to female youth unemployment. The figure also shows that more than 40% of the female labor force aged is unemployed in Egypt, Guyana and Iraq. Also about more than third of the youth female labor forces are without jobs in Algeria, Bahrain, Gabon, Iran and Qatar. The high rate of female youth unemployment in the OIC countries may be attributed to low skills and experience of women. In addition most of OIC countries, particularly Arab states are dominated by some traditions and legislations which prevent women from work, besides early marriage and low education attainment (Ebaidalla, 2014).

94 Journal of Economic Cooperation and Development 87 Figure 4: Youth Unemployment in OIC Countries by Gender (2010) Source: World Bank Development Indicator (2014) 3. Determinants of Youth Unemployment: A Literature Review Due to unfavourable consequences of unemployment, the issue of youth unemployment has received considerable attention from both researchers and policy makers. Many empirical studies on the determinants of youth unemployment have been emerged in the last decades. In the literature, several variables have been considered as main factors influencing youth unemployment including aggregate demand, education, demographic change, wages, labor market policies and individual characteristics. Here we briefly review the most important causes cited in the literature. The level of aggregate demand is widely assumed as a significant variable that affecting youth unemployment. The reduction in aggregate demand due to economic crisis will increase youth unemployment rates. That is, in the time of recession, employers tend to reduce the number of workers particularly the young; hence, youth people are more likely to suffer from unemployment than adults. This also indicates that economic situation significantly affects youth unemployment, as economic downturn reduces demand for labor and in turn, increases youth unemployment (Ebaidalla, 2014). For example, during the recent global crisis the number of unemployed youth has increased from 73.5 million in 2007 to 77.7 million in 2010 (ILO, 2011). In addition, during that same period, the rate of youth unemployment in

95 88 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis European countries rose to 4.5 point. Moreover, as long as the excessive supply of labor occurs in the recession periods, the employers prefer adult workers rather than non skilled-youth workers (Borowski, 1986). Another factor affecting youth unemployment is the demographic transformations. It is well known that demographic transitions such as changes in the age structure of the population, fertility and child mortality affect the situation of youth labor markets (Gomez-Salvador and Leiner-Killinger, 2008). For example, an increase in the number of young people in the economy would have an adverse impact on the rate of youth unemployment. This is because young workers may complement adult workers, in terms of endowments and qualifications. Thus, an increase in the size of young workers relative to adult workers would then tend to raise unemployment, and in turn put a downward pressure on wages for young workers. Korenman and Neumark (2000) argued that, for a number of countries, an increase in the rate of young persons relative to prime age persons negatively influences employment and wages among the cohort size of young persons. On the other hand, the change in fertility and child mortality rates also results in a significant impact on youth unemployment. For example, the high rate of youth unemployment in some developing regions like South Asia and Sub Saharan Africa is attributed mainly to the high rate of fertility (births per woman). Assaad and Roudi-Fahimi (2007), who studied the youth unemployment in MENA countries, argue that the significant decline in child mortality leads first to an increase in the proportion of children under 15, and then to an increase in the proportion of young people aged 15 to 24. In addition, many empirical studies have claimed that youth unemployment responds highly to the changes in minimum wage paid by employers. An increase in minimum wages reduces youth employment, as wage increases raise the cost of production and make firms hire high skilled workers rather than youth. This can be explained by the assumption of a competitive labor market which claims that demand for labor decreases as the cost of real wage increases. If a minimum wage level increases, the demand for lower skilled labor will go down through two routes. First, employers will tend to substitute lower skilled labor by higher skilled workers. Secondly, the scale of production may be reduced due to increased costs of production, which will result in the reduction of the demand for labor including lower-

96 Journal of Economic Cooperation and Development 89 skilled labor (Ghellab, 1998). However, most of empirical studies on the relationship between minimum wages and youth unemployment have found ambiguous results. For example, Blazquez et al. (2009) found no significant relationship between minimum wages and youth unemployment in Spain. However, Neumark and Wascher (1999) found the effect of minimum wages on youth employment to be significant. In addition, Pereira (2003) studied the relationship between minimum wages and youth employment in Portugal in 1987 and found that an increase in minimum wages, inter alia, had a negative impact on youth employment compared with that of older workers and made firms substitute youth workers by older ones. In the same vein, the level of education is considered an important factor that influencing youth unemployment. Most of empirical literature argue that education significantly discourage youth unemployment. In fact, educated workers have better opportunities than illiterate people in seeking new jobs and gaining higher wages; hence, there is a lower risk of unemployment at higher educational levels (Kabaklarli, et al., 2011). However, another group of studies found that education may increases unemployment particularly of youth people (e.g. Galal, 2002). They attributed this phenomenon to a mismatch between the supply of education and demand for labor. In some cases the number of skilled jobs has not responded as quickly as the supply of educated workers, and thus high-skilled workers either had to accept jobs for which they were over-qualified or face unemployment (Venatus and Agnes, 2010). Therefore, investments in training institutions of vocational and technical education, which have a functional link with the labor market, are crucial and must be increased (Venatus and Agnes, 2010). Furthermore, individual characteristics exert important impact on youth unemployment. It has been argued that the youth unemployment rate varies in response to individual features of the youth such as, gender, age, race and region. For example, young women and girls have historically been more likely to be unemployed, but due to the recent recessions this trend seems to have changed (Higgins, 1997). Ethnic minorities and the poor also may suffer from the risk of youth unemployment. Geographical location in terms of rural vs. urban residence has a significant negative impact on employment opportunities. In addition, gender discrimination in rural areas is likely

97 90 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis to exclude girls from the education as well as labor markets (Ebaidalla, 2014). Finally, the labor market institutions and policies are widely acknowledged as the main determinants of youth unemployment. Labor market policies such as employment protection and regulation may lead to significant changes in youth unemployment. Many empirical studies found that employment protection regulations have a negative impact on employment in general and youth employment in particular (e.g., Venatus and Agnes, 2010). For example, a high level of protection against dismissal of workers tends to discourage firms from taking more workers during production booms, as it would be too costly to fire them when the economic situation is downturned. Hence, high levels of protection regulations make firms use only high-skilled workers as dismissals are costly (Ebaidalla, 2014). As a result, young workers have little opportunity to find jobs. In addition, if dismissing workers is unavoidable in time of recession, firms tend to dismiss young workers in higher numbers than old age workers as redundancy payments increase with job tenure (Gomez-Salvador and Leiner-Killinger, 2008). 4. Methodology and Data 4.1. Model Specification To analyze the youth unemployment in OIC countries, the paper uses econometric approach, employing a panel data method. The specification of unemployment model employed in this study follows the previous empirical studies on youth unemployment such as, Choudhry et al. (2012), Anyanwu (2013) and (2014). However, the model is extended by demographic and institutional variables. Thus, the model is specified as follows: YU it = β 0 + β 1 GDP it + β 2 INF it + β 3 TRD it + β 4 INV it + β 5 EDU it + β 6 FER it + β 7 BUR it + + µ it (1) Where the subscripts i and t represent the country and time period, respectively. The variable YU it is the dependent variable capture the youth unemployment. This model relates youth unemployment to a set of explanatory variables that hypothesized to influence youth unemployment. The explanatory variables include economic,

98 Journal of Economic Cooperation and Development 91 demographic and institutional factors. GDP is the growth of gross domestic product, INF is the inflation rate, TRD is the trade openness variable, INV is domestic investment, ED is education, FER is the fertility rate, BUR is the bureaucracy quality. The bureaucracy quality variable is used to capture the impact of institutional quality, since efficient institutions are conducive to labor productivity and growth (Acemoglu et al., 2004). Finally, µ is the error term with a zero mean and constant variance. The definition and sources of the variables are presented in Appendix I. All the variables are expressed in the natural logarithm except GDP growth because it bears negative values for some countries. According to theoretical and empirical literature, the impact of GDP growth is expected to be negative, as an increase in country s income will reduces the employment level. This also supported by the Okun s Law (1962) which is confirmed by many empirical studies 5. The inflation rate would be negative as expected, since there is negative association between the unemployment rate and inflation, as argued by Philips (1958). The impact of domestic investment is expected to be negative, as an increase in investment would encourage the employment of young people. The coefficients of trade openness and education are expected to be positive. The effect of fertility rate also is expected to be positive, because an increase of fertility level rises the numbers of young people, and hence their unemployment rates. Finally, the coefficient of bureaucracy quality is expected be negative, since a country with bad institution tend to suffer from unfavorable labor conditions and hence, high rate of youth unemployment. This is also supported by several studies on economic growth which assumed that good institutions are necessary for labor and human capital progress (e.g. Acemoglu et al., 2004). 5 Economist Arthur Okun (1962) who first started studying the relationship between unemployment and economic growth, and his research on the subject since then has become known as Okun's law. Okun law indicates a positive association between employment and output.

99 92 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis 4.2. Estimation Methodology To investigate the determinants of youth unemployment in OIC countries, the study uses a panel data method. Therefore, the error term µ it in equation 1 is a composite disturbance term that includes an unobserved country-specific effect η i, a time-specific effect λ t, and a error term ε it. The composite error could be specified as follows: µ it = η i + λ t + ε it (2) The possible endogeneity of unemployment as well as the correlation of the unobserved country fixed effects with the error term implies that the orthogonality condition is not likely to be met for Fixed Effects (FE) or random effects (RE) estimator to produce consistent estimates. To consider endogeneity of regressors the appropriate method is to use an instrumental variables (IV) estimator. Therefore, this study uses the dynamic panel models based on the Generalized Method of Moments (GMM), developed by Arellano and Bond (1991) The GMM dynamic method has many advantages over the conventional static panel data methods of fixed effects and random effects models. First, GMM model takes into account the time series dimension of the data and, hence it considers the short run effect. Second, it includes the non observable country specific. Finally, this method considers all the explanatory variables as endogenous, hence it account for the problem of endogenity that may result from the correlation between the error term and the lagged dependent variable. Empirically, there are two types of GMM models that have been widely used in estimating panel regressions: the first-difference GMM estimator, developed by Arellano and Bond (1991) and the system GMM estimator, developed by Arellano and Bover (1995) and Blundell and Bond (1998). However, the recent literature has shown that there are some possible statistical problems associated with the use of firstdifference GMM estimator (Bond et al., 2001). That is, when the regressors are highly persistent, the instrumental variables used in difference GMM, such as, lagged levels of the dependent variable and of the explanatory variables might be weak instruments. In this situation, the first-differenced GMM model potentially suffers from a downward

100 Journal of Economic Cooperation and Development 93 bias, especially when the time periods (T) is small (Blundell and Bond, 1998). On the other hand, the system GMM estimator overcomes the weak instruments problem by allowing the use of the lagged differences and lagged levels of the explanatory variables or other variables as instruments. Therefore, the analysis in this paper relies solely on the system GMM technique. The System GMM estimators are drawn from the estimation of a system of two simultaneous equations, one in levels (with lagged first differences as instruments) and the other in first differences (with lagged levels as instruments). However, one possible problem that might be arising when adopting the system GMM is the invalidity of the lagged differences of the explanatory variables as instruments. Therefore, we examine the validity of the instruments by two tests: Sargan test of overidentifying restrictions and Arellano and Bond (AB) test of serial correlation. Furthermore, the study tests the endogeneity problem applying Durbin (1954), Wu (1974) and Hausman (1978) endogeneity test Data Sources The paper uses annual data for a sample of 32 OIC countries over the period The data on employment, macroeconomic and demographic variables are obtained from several sources including World Bank s development indicators and IMF Financial Statistics (IFS). Data on institutional quality (bureaucracy quality) is sourced from the International Country Risk Guide (ICRG) statistics 6. The summary statistics of the variables that used in our analysis is presented in Appendix II. It is clear that the inflation rate, trade openness and education vary greatly across countries. The descriptive statistics also shows that the youth unemployment in OIC countries is relatively high and varies across member states. Interestingly, the 6 The International Country Risk Guide (ICRG) political stability indicators comprises 12 institutional measures - government stability, socioeconomic conditions, investment profile, internal conflict, external conflict, corruption, military in politics, religious tensions, law and order, ethnic tensions, democratic accountability, and bureaucracy quality.

101 94 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis standard deviation of economic growth is small, implies that the income of OIC countries grow slowly and unequally distributed. 5. Empirical Results and Discussion The estimation results of equation (1) using system GMM method are presented in Table 1 below. Column 3 presents the results of GMM model while column 2 contains the results of fixed effect estimates which presented for the purpose of comparison. Table 1: The Results of System GMM estimation ( ) Variable Fixed Effects GMM Constant (17.81) Youth unemployment (-1) (2.14) GDP Growth (-3.25) Inflation (3.36) Trade openness (-2.93) Domestic Investment (3.13) Education (-2.87) Fertility (-0.485) Bureaucracy (17.81) (0.50) 0.998*** (15.20) * (-1.88) ** (-2.21) 0.018** (2.46) *** (-4.00) (-1.20) 0.817* (1.82) (-0.55) Observations Sargan Over-identification Test (0.426) Durbin Wu Hausman Test (0.023) AB- test for AR(1) (0.000) AB- test for AR(2) (0.547) Notes: Figures in Parentheses are t-statistics ***, ** and * denotes significance at the 1, 5 and 10 percent level The estimation is two steps System GMM, instruments used in the analysis are the first lagged difference and the second lagged level of dependent and explanatory variables.

102 Journal of Economic Cooperation and Development 95 Durbin Wu Hausman and Sargan tests are asymptotically chi-sq distributed, with p values reported in brackets. The results of Table 1 show that the Sargan and Arellano and Bond tests suggest that GMM regression do not suffers from any problems with the instruments, and there are no second order serial correlation problems in the estimated model. The Durbin Wu Hausman test statistic also rejects the null hypothesis that all regressors in the model are exogenous at any reasonable degree of confidence. Thus, we conclude that all variables in the model are endogenous and this constitutes a suitable justification for using GMM method. Focusing on the results of system GMM model in column 3, we find that most of the estimated coefficients bear the expected sings and in line with the theory. The results also indicate that all the variables are statistically significant except education and bureaucratic quality. The results show that GDP growth has negative and significant impact on youth unemployment, implying that increasing economic growth reduce the youth employment in OIC countries. This result confirms the validity of Okun s Law for OIC countries. The coefficient of inflation is also negative and significant, indicating that Philips Curve hypothesis holds in the OIC countries. Unexpectedly, the trade openness is found to have positive impact on youth unemployment, implying that a country with high degree of trade openness tends to experience more youth unemployment. This result could be explained by the fact that trade openness in OIC countries are mainly due the high imports which hurts the local industries and increase unemployment in general and youth unemployment in particular. Moreover, the results show that the coefficient of domestic investment is negative and significant, suggesting that increasing investment tends to discourage youth unemployment in OIC countries. Therefore, we conclude that economic environment plays a significant role in explaining youth employment in OIC countries. This also implies that the demand side of labor market has an important role in influencing youth unemployment in OIC countries. The results of Table 1 also reveal that the coefficient of education is negative but it is not significant. This finding implies that labor supply factors like education has no significant impact on youth unemployment

103 96 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis in OIC countries, since demand side factors are the most important variables influencing youth unemployment rate. In addition, the results of system GMM model indicate that the impact of fertility is positive and significant, suggesting that high fertility rate increases the opportunity of youth people to confront unemployment. This result also indicates that a country with high level of fertility tends to suffer from high rate of youth unemployment. Furthermore, the results reveal that bureaucracy quality has a negative effect on youth unemployment, but it is not significant Conclusion and Policy Implications Motivated by high rates of youth unemployment in OIC countries, this study aims at identifying the determinants of youth unemployment in the region. The study used a dynamic panel data method for a sample of 32 countries during the period ( ). The analysis focused on the impact of economic, demographic and institutional quality. The empirical results show that real GDP growth, inflation and domestic investment exert negative and significant impact on youth employment. The results also show that trade openness has positive and significant effect of youth unemployment in OIC countries. These findings imply that economic environment play a significant role in explaining youth unemployment in OIC countries. In addition, the results also show that demographic changes measured by the fertility rate have positive and significant effects on youth unemployment in OIC countries. Furthermore, the results pointed out that institutional quality measured by the quality of bureaucracy has a negative impact on youth unemployment. This finding indicates that a country with an inefficient bureaucratic system is more likely to suffer from youth unemployment. Based on the above findings, many policy recommendations can be provided aiming at improving the employability of young people in OIC countries. First and foremost, the economic environment needs to be given considerable attention using sound economic policies. Thus, effective fiscal and monetary policies should be adopted to foster the GDP growth and attain economic stability via reducing the inflation 7 Bureaucracy quality index is scaled from zero to six. Higher scores indicate more bureaucracy quality.

104 Journal of Economic Cooperation and Development 97 rates. The trade sectors need to be paid serious efforts to improve the exports capacity and competitiveness so as to create more jobs for young people. In addition, public and private investment should be enhanced in projects that provide opportunities for youth. Moreover, attractive incentives should be offered to foreign investment aiming at increasing the production capacity and job opportunities. On the demographic front, health and family planning programs should be at the top of policy agenda. This is because OIC countries suffer from high level of fertility and population growth. Besides, OIC countries need to pay great attention to educational policies. Therefore, serious efforts should be paid to vocational and technical education aiming at developing skills and experiences of young people so as to help them to enter the labor markets.

105 98 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis References Acemoglu, D., Johnson, S. and Robinson, J. (2004), Institutions as the Fundamental Cause of Long-Run Growth: Handbook of Economic Growth, North Holland. Anyanwu, J. (2013). Characteristics and Macroeconomic Determinants of Youth Employment in Africa African Development Review, Vol. 25, No. 2, pp Anyanwu, J. (2013). Does Intra-African Trade Reduce Youth Unemployment in Africa? African Development Review, Vol. 26, No. 2, PP Assaad and Roudi-Fahimi (2007) Youth in the Middle East and North Africa: Demographic Opportunity or Challenge? MENA Policy Briefs, Population Reference Bureau, Washington. DC. Blazquez, M.; Llorente, R. and Moral, J. (2009), Minimum Wage and Youth Employment Rates in Spain: New Evidence for the Period , Working Papers in Economic Theory. Universidad Autonoma, Madrid, Spain. Borowski, A. (1984) A Comparison of Youth Unemployment In Australia And The United States, Monthly Labor Review, 107 (3), Choudhry, M., Marelli, E. and Signorelli, M. (2012) Youth unemployment rate and impact of financial crises, International Journal of Manpower, Vol. 33, No. 1, pp Ebaidalla, E. (2014) Youth Unemployment in the Arab World: An Analysis of Causes and Possible Ways Forward In Naoual Belakhdar, et al. (ed), Arab Revolutions and Beyond: Change and Persistence, Center for North African and Middle Eastern Politics, Freie University Berlin, Berlin., pp

106 Journal of Economic Cooperation and Development 99 Galal, A. (2002) The Paradox of Education and Unemployment in Egypt, A paper presented at the ECES conference on Employment and Unemployment, Cairo, Egypt, Ghellab, Y. (1998),Minimum Wage and Youth Unemployment. Geneva, ILO. Gomez-Salvador, R. and Leiner-Killinger, N. (2008) An analysis of Youth Unemployment in the Euro Area,European Central Bank Occasional Paper Series, No. 89. International Labor Organization s (ILO) statistics (2011) ILO (2013),Global Employment Trends for Youth. ILO, Geneva. Kabaklarli, E., Hazel Er, P. and Bulus, A. (2011) Economic Determinants of Turkish Youth Unemployment Problem: Co Integration Analysis Presented at the International Conference on Applied Economics, Held in Dubai, May 16-18, Korenman, S. and Neumark, D. (2000) Cohort crowding and youth labor markets: a cross-national analysis, in: Blanchfl, D. and Freeman, R. (ed), Youth Employment and Joblessness in Advanced Countries, The University of Chicago Press, Chicago. Neumark, D. and Wascher, W. (2004) Minimum Wages, Labor Market Institutions and Youth Employment: A cross-national Analysis Industrial and Labor Relations Review, 57 (2), Neumark D. and Wascher, W. (1999) A cross-national Analysis of the Effects of Minimum Wages on Youth Unemployment NBER Working Paper, no. 7299, Cambridge, MA. Pereira, S. (2003) The Impact of Minimum Wages on Youth Employment in Portugal European Economic Review, 47(2), Statistical, Economic and Social Research and Training Centre for Islamic Countries (SESRIC) (2012) Annual Economic Report on the OIC countries SESRIC, Ankara, Turkey.

107 100 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Statistical, Economic and Social Research and Training Centre for Islamic Countries (SESRIC), BASEIND statistics (2014). Available at accessed 11, November, United Nations, World Population Prospects: The 2010 Revision (New York: UN, 2010). United Nations, World Population Prospects: The 2004 Revision (New York: UN, 2005). Venatus, K. and Agnes, I. (2010) Youth Unemployment in Nigeria: Causes and Related Issues, Canadian Social Science, 6, (4), World Bank (2014), World Bank Development Indicators. Accessed 13\9\2013. Available at: accessed 10, November, 2014.

108 Journal of Economic Cooperation and Development 101 Appendices Appendix I: Description and Sources of Variables used in the Regression Analysis Variable Definition Source Youth Unemployment Measured as residual of Employment to Population Ratio, Ages Total (%)*. Real GDP Growth Growth of gross domestic product Inflation Rate Annual inflation rate measured by the change in consumer price index. Trade Openness Measured by the ratio of (Exports +Imports) to GDP Domestic Investment Domestic Investment, measured by fixed capital formation as share of GDP % Fertility Rate Measured by births per woman Education Measured by ratio of total secondary enrolment to the population Bureaucracy Quality Bureaucracy quality, measures autonomy from political pressure and strength and expertise to govern without drastic changes in policy or interruption in government services. (scale from zero to six) World Bank s World Development Indicators World Bank s World Development Indicators IMF Financial Statistics and World Bank s World Development Indicators IMF Financial Statistics and World Bank s World Development Indicators World Bank s World Development Indicators United Nations, World Population Prospects and World Bank s World Development Indicators. World Bank s World Development Indicators International Country Risk Guide (ICRG), (2013) Note: All the variables are expressed in the natural logarithm, except GDP growth. * Due to the lack of systematic labor surveys in OIC countries the data on unemployment of total youth as % of total labor force ages is not available.

109 102 Determinants of Youth Unemployment in OIC Member States: A Dynamic Panel Data Analysis Appendix II: Summary Statistics of Sample Data Variable Mean Std. Dev. Min Max Youth Unemployment GDP Growth Inflation Trade Openness Domestic Investment Education Fertility Bureaucracy Appendix III: List of OIC Countries that Considered for the Study No Country Name No Country Name 1 Albania 17 Malaysia 2 Algeria 18 Mali 3 Bahrain 19 Morocco 4 Bangladesh 20 Mozambique 5 Brunei 21 Niger 6 Burkina Faso 22 Nigeria 7 Cameroon 23 Pakistan 8 Egypt 24 Qatar 9 Gabon 25 Saudi Arabia 10 Gambia 26 Senegal 11 Guinea 27 Sudan 12 Indonesia 28 Syria 13 Iran 29 Togo 14 Jordan 30 Tunisia 15 Kuwait 31 Turkey 16 Lebanon 32 Uganda

110 Journal of Economic Cooperation and Development, 37, 2 (2016), Determinants of Tax Revenue: Does Liberalization Boost or Decline It? Mansour Zarra-Nezhad 1, Majid Sheikh Ansari 2 and Mahvash Moradi 3 Trade liberalization due to its improving role has been at the center of economies in recent years. Against public finances of many developing and emerging market countries, it still acts as develop alternative source of revenue. Using a panel of 83 countries, over , we perform Generalized Method of Moment regression to test the hypothesis whether trade liberalization boosts tax revenue. Also, this paper assesses the statistical significance number of potential determinants of tax revenue as a share of GDP. The results of dynamic panel estimation show more trade liberalization is accompanied by more tax revenue. Among the variables that exert a statistically significant influence on tax revenue are the following ones: GDP growth rate, share of agriculture over GDP, official exchange rate, urbanization and democracy. Therefore, the study concludes that there is the need for appropriate macroeconomic policy to enhance the trade liberalization in order to accelerate government revenue. 1. Introduction Trade liberalization is a hotly debated issue in recent years in economies. However, the early classical theorists; David Ricardo and Eli Heckscher have long pointed out the possible gains from trade. They suggest that these gains result from specialization in production due to international trade. After that, the related issues of trade liberalization became focal point of interests for many authors. As Longoni (2009) stated, one of the most widely accepted theories in economics claims that there exists a positive relationship between a higher degree of openness to international trade and economic growth. As a consequence, 1 Department of Economics, Shahid Chamran University, Ahvaz, Iran 2 Department of Economics, Yazd University, Yazd, Iran 3 Department of Economics, Shahid Chamran University, Ahvaz, Iran moradi.mahvash67@gmail.com

111 104 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? a trade liberalization reform is largely considered as a growth-enhancing strategy because of its positive effect on the promotion of efficiency, the improvement in international competition and the expansion of the trade volume. Agbeyegbe et al., (2004) specified that trade liberalization has frequently been the centerpiece of an economic development strategy. Trade liberalization often entails a reduction and unification of tariffs and relaxation of quantitative barriers, and may be accompanied by domestic tax reform. Also, Pupongsak (2009) claimed trade liberalization has outstanding advantage which induces most countries to walk toward free trade regime. Globalization is a phenomenon which involves increases in the flows of trade, capital, information and technology, mobility of labor across borders, substantial expansion in world production, and consequently, a rise in world economic welfare. Trade liberalization is normally associated with the reduction, removal and elimination of taxes on goods and services (including tariffs and import duties), and other trade barriers such as quotas on imports, subsidies, and non-tariff barriers to trade. There are many examples which strongly support the notion that openness to international trade brings more benefits to the country, for instance Harrison (1996), Harrison and Hanson (1999), Rodríguez and Rodrik (2000), which suggest the positive association between trade liberalization and economic growth. In addition, World Bank (2002) reported that almost half of the developing countries which have lowered their average tariffs by about 30 percentage points, experienced growth of per capita income by 4 percent in Thus, over the past few decades, liberalizing the external trade regime has been one of the central and most visible elements of many less developed and developing countries to achieve accelerated economic growth. Against the advantages a country can achieve from liberalization process, it has been questioned "Whether all countries have benefited from the gains of trade liberalization?" or "Whether countries which rely heavily on tax revenue as a source of government revenue benefit from liberalization?" The relationship between trade liberalization and tax revenue is therefore an issue of great practical importance, since; trade liberalization is mainly thought to be linked to tax revenue through its effect on international trade tax revenue. On one hand, it has been argued that trade liberalization is likely to lead to a considerable decrease in international trade tax revenue through the reduction of tariffs, especially in developing countries. The fiscal drawback is serious

112 Journal of Economic Cooperation and Development 105 if a country is highly dependent on international trade tax. In this way, economists recommend that, in order to mitigate the loss of international trade tax revenue, one strategy is to boost both domestic direct and indirect taxes, mainly increasing revenue from goods and services tax, by implementing domestic tax reform. Leading revenue sources from international trade tax to broad-based domestic taxes, economists believe that the negative impact of trade liberalization can be offset or reduced. As Greenaway and Milner (1993) found that there is a wide range of possible revenue outcomes from trade liberalization, depending on initial conditions, the components of the reform package, the effects of changes in tariff rates, changes in the import base, and changes in the exchange rate. On the other hand, in the last several decades, there have been ambitious efforts in much of the countries to liberalize trade, because they recognize that trade reform is vital for economic development, poverty reduction, more efficient allocation of resources, enhanced productivity, and higher economic growth. The increasing trend of liberalization around the world is illustrated in Figure 1. Figure 1: Trend of Liberalization around World Source: Own calculation As discussed earlier, the results around tax revenue and trade liberalization did not converge to a global conclusion. All in all, in this paper we aim at scrutinizing the effect of trade liberalization on tax

113 106 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? revenue across a selected group of countries during the period of The reminder of this paper is as follow: Section 2 offers a brief literature review. Section 3 presents theoretical relation between liberalization and revenue. Section 4 contains econometric procedure and results. Finally, section 5 concludes and suggests policy recommendations. 2. Literature Review Several studies have been done on trade liberalization as an area that deserves attention in international trade. In this part, we aim at introducing some relevant literature. Agbeyegbe, et al., (2004) examined the linkages between trade liberalization, exchange rates, and tax revenue. Using a panel of 22 countries in Sub-Saharan Africa, over , they performed Generalized Method of Moment regressions to test this relationship. They found evidence that the relationship between trade liberalization and tax revenue is sensitive to the measure used to proxy trade liberalization, but that, in general, trade liberalization is not strongly linked to aggregate tax revenue or its components though with one measure, it is linked to higher income tax revenue. Currency appreciation and higher inflation showed some linkage to lower tax revenues or its components. Baunsgaard and Keen (2005) in a paper have focused on the simple question: Over the last 25 years, have countries actually managed to offset reductions in trade tax revenues due to liberalization by increasing their domestic tax revenues? Based on results, for high-income countries, the answer was clearly yes. For middle-income countries, there was also evidence of significant recovery: there were strong signs that this had been in the order of cents of additional domestic tax revenue for each dollar of trade tax revenue, with apparently full recovery when separately identifying the episodes in which trade tax revenues fell. For low-income countries, however, recovery had been far from complete. At best, they had on average recovered no more than around 30 cents of each lost dollar. Gupta (2007) investigated revenue performance of a large set of developing countries over the past 25 years. He found that several

114 Journal of Economic Cooperation and Development 107 structural factors like per capita GDP, share of agriculture in GDP and trade openness are statistically significant and strong determinants of revenue performance. The results indicated that although foreign aid improved revenue performance significantly, debt did not. Among the institutional factors, corruption had a significantly negative effect on revenue performance. Political and economic stability also affected revenue performance, but only across certain specifications. Longoni (2009) empirically investigated the effect of trade liberalization on trade tax revenues applying panel-data methods to a large sample of African countries from the period He found that there exists a large tradeoff between a greater degree of openness to international trade and the revenue collected from import and export taxation. Moreover, he also found that the relationship between trade taxes and trade tax revenues is nonlinear, giving credit to the existence of a Laffer effect. Dioda (2012) in a study aimed at investigating the structural or longterm determinants of tax revenue by applying standard models to the case of Latin America and the Caribbean. Through panel econometric methodologies, the paper assesses the statistical significance of a number of potential determinants of tax revenue as a share of GDP, using data from 32 Latin American countries over the period The results indicate that, among the variables that exert a statistically significant influence on tax revenue are the following ones: civil liberties, female labor force participation, the age composition of the population, the degree of political stability, the level of education, the population density as well as the size of the shadow economy. In that year, a study by Nwosa et al., (2012) examined the relative contribution of trade liberalization on trade tax revenue in Nigeria for the period 1970 to The findings of the study showed that trade liberalization, public debt, trade openness, gross domestic product and labor force affected positively on trade tax revenue while exchange rate had negative effect. The Wald test showed that labor, public debt and exchange rate had significant influence on trade tax revenue while the Beta coefficient showed that trade liberalization policy was the major determinant of trade tax revenue in Nigeria. The study concluded that there is the need for appropriate macroeconomic policy to enhance the success of trade liberalization policy in Nigeria.

115 108 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? Muibi and Sinbo (2013) attempted to examine the most relevant macroeconomic policy variable that can serve as an anchor variable for stimulating tax revenue and boost the revenue profile of the government. The paper used secondary data from Nigeria economy for the period 1970 to 2011 and adopted the error correction mechanism to establish both the long run and short run relationships among the variables. The main finding of the empirical analysis was that tax revenue tends be significantly responsive to changes in income level, exchange rate and inflation rate. The income elasticity of tax showed that a unit percent increase in income level probably lead tax revenue increase by 0.63% in the immediate and 0.33% in the second year. Epaphra (2014) studied the impact of trade liberalization proxied by reduction in collected tariff rate and other determinants of tax revenue that are associated with trade liberalization and reforms. In estimating the import duty revenue model, cointegration analysis and error correction modeling were applied over the 1979/ /10. The study findings showed that trade liberalization is a potential source of fiscal instability in Tanzania because it relies heavily on revenue from international trade. Trade liberalization eventually results in reduced import duty revenue. 3. Tax Revenue-Trade Nexus Countries collect taxes in different ways. It is therefore not possible to generalize about the effect of changes in trade liberalization and the surrounding macroeconomic environment on tax revenues without examining the structure of the different components of revenues. Tax systems encompass a wide variety of taxes, which can be divided into three general categories: taxes on income and profits, taxes on goods and services, and international trade taxes (Agbeyegbe et al., 2004). To investigate the effect of trade liberalization on tax revenue the underlying analysis that are clearly defined in Adam et al., (2001) is used. We examine a small open economy facing world prices of P X and P M for its exports and imports that produces and consumes a non-traded domestic good with price P N. Capital stocks are sector-specific while labor, denoted L; moves between sectors to equalize real consumption wages. Public expenditure is financed through the three taxes: t on income; d on non-tradable production and consumption; and the tariff on

116 Journal of Economic Cooperation and Development 109 imports, φ. Relative prices are defined in terms of market rather than factor prices, hence the import real exchange rate is denoted: Q = P N P M = P N P M + τ (1) Where τ denotes the tariff and an increase in Q denotes an appreciation. The export real exchange rate is Q X = P N /P X from which it follows, Q Q X = P X P M = P X P M + τ = T (1 φ) (2) Where T = P X /P M is the (exogenously determined) small country's terms of trade in international markets, and φ= τ/ (P M + τ) is the tariff expressed as a proportion of the tariff inclusive price. Equilibrium holds when aggregate spending equals aggregate income at full employment and the trade balance is equal to the exogenous aid inflow (for simplicity we assume there are no private international capital movements or changes in reserves). Using revenue and expenditure functions (see Dixit and Norman, 1980), and denoting public and private expenditure functions and utility by lower- and uppercase letters respectively, we express this relationship as: e (P N, P M, u) + E (P N, P M, U) = R (P N, P X ; L) + P M A (3) In which u, U and A denote public and private sector utility and aid measured in units of imports. There is no government production in this economy: the government simply consumes the imported and nontradable goods, although its preferences are not necessarily the same as those of the private sector. For convenience, we normalize on the domestic price of imports (P M ) and re-express (3) as: e (Q, 1, u) + E (Q,1, U) = R (Q, T (1- φ); L ) + A (4) By the properties of the revenue and expenditure functions, letting subscripts denote partial derivatives with respect to the relevant arguments, we can express the supply and compensated demand functions for non-traded goods as R Q, e Q and E Q respectively, leading to market-clearing condition in the non-traded goods market:

117 110 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? e Q (Q, 1, u) + E Q (Q, 1, U) = R Q (Q, T (1- φ); L) (5) By Walras Law, (4) and (5) imply equilibrium in the tradable goods sector. GDP measured in imported goods is: R= QR Q + T (1- φ) R T (6) Where R T represents the supply of tradable goods. Public expenditure is financed through a tax on private sector income (excluding net aid inflows), the domestic sales tax and the tariff on imports. With no changes in foreign reserves and aid as the only capital inflow then the tax base for tariff revenue can be expressed in terms of the capacity to import (i.e. total exports plus aid valued at domestic import prices). Finally, we allow for Tanzi effects so that the actual real revenue yield from each tax instrument is a (negative) function of the inflation rate, π. Tanzi effects are assumed to differ across the three taxes and are denoted α(π), β(π), and γ(π) respectively for income taxes, indirect taxes, and trade taxes. These Tanzi effects represent the only price nonhomogeneity in the model. Thus, with a balanced government budget, we define the revenue/expenditure of the public sector as: e (Q, 1, u) = t(1 α (π)) (QR Q + T(1- φ)r T ) + d(1-β(π))qr Q + φ( 1- γ ( π)) (T(1- φ)r T + A) (7) Defining the revenue yield for each tax as Y i for i = t; d; and φ, we obtain: Y t = t(1 α(π)) (QR Q+ T(1 φ)r T ) Y QR Q + T(1 φ)r d = d(1 β(π))qr Q T QR Q + T(1 φ)r T Y φ = φ(1 γ(π))(t(1 φ)r T+A) QR Q +T(1 φ)r T (8) Totally differentiating these revenue yields, we get the following expressions for the evolution of the component tax yields, under the assumption that the tax rates are fixed at t, d and φ. For convenience, we define the own- and cross-price elasticity of supply for tradable and nontradable as: σ QQ = QR QQ, σ R TT = TR TT Q R T, σ QT = TR QT, and σ R TQ = QR TQ (9) Q R T

118 Journal of Economic Cooperation and Development 111 Denoting the non-tradable component of GDP by n=qr Q and the tradable component as s = (1 φ)tr T, the total derivatives for each tax component are as follows: for income taxes, dy t = -tα π dπ (10) dy d = -dβ π ( n ) dπ + n+s s d(1 β(π))r Q [1 + σ QQ σ TQ ] ( (n + s) 2) dq + d(1 β(π))(1 φ)r T [σ QT (1 + σ TT )] ( n (n+s) 2) dt (11) for indirect taxes, and dy φ = - φγ π ( s+a n+s ) dπ + φ(1 γ(π))r Q [(1 φ)(n A)σ TQ (s+a)(1+σ QQ )] (n+s) 2 dq + φ(1 γ(π))r T [(1 φ)(n A)(1 + σ TT ) (s + A)(σ QT )] (n + s) 2 dt + φ(1 γ(π)) da (n + s) (12) for tariffs. As mentioned before the effect of trade on tax revenue is not clear. Based on the theoretical concept, i) When the own-price elasticity of supply for tradable is sufficiently large relative to cross-price elasticity, such that (1+σ TT )>σ QT, an improvement in the terms of trade, denoted by an increase in T; will reduce the yield from domestic indirect taxes and vice versa for a deterioration in the terms of trade; (ii) when the own-price elasticity of supply for tradable is sufficiently large relative to cross-price effects and the share of tradable in total GDP is not too large, such that (1+σ TT )>σ QT (s+a)/((1- φ)(n-a)), an improvement in the terms of trade will improve the trade tax yield and vice versa for a deterioration in the terms of trade; (iii) movements in the terms of trade have no impact on the income tax yield.

119 112 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? 4. Model Specification and Econometric Results What affects tax revenue has been the subject of a long debate. Before turning to the evidence, we discuss factors that matter most for explaining tax revenue. It is possible to classify the factors into three main groups: Economic indicators (Such as GDP growth rate, trade liberalization, share of agriculture over GDP and exchange rate), Socio demographic indicators (For instance urbanization), Political indicators (Like democracy). The construction of this relationship in the framework of a dynamic panel can be specified as: Tax i,t = α i + γ t + β 1 Tax i,t-1, + β 2 GDP i,t + β 3 Trd i,t + β 4 Agr i,t + β 5 Exc i,t + β 6 Urb i,t + β 7 Demo i,t + ε i,t (13) Variables are expressed across a series of countries (i=1,,n) and time periods (t=1,,t). The first two terms on the right side of the equation are intercept parameters, which change among the various countries i and years t. They allow for specific effects across countries (α i ) and across time (γ t ). εit shows random disturbance. As a dependent variable, we use tax revenue as a share of GDP. Explanatory variables include the following: GDP- GDP growth rate Trd- trade liberalization Agr- share of agriculture over GDP Exc- official exchange rate Urb- urbanization Demo- democracy GDP growth rate: Based on the Wagner law, since the demand for public services is sensitive to income (it is elastic), economic development is accompany with an increased request for public goods and services which need to be financed by increasing tax revenue. Also, development is related to larger capability to levy and collect taxes (Chelliah, 1971). Hinrichs (1966) and Tanzi (1992) point out to a positive link between development and tax revenue of a country.

120 Journal of Economic Cooperation and Development 113 Trade liberalization: Openness degree, which is measured as the share of international trade in GDP, may also have a significant impact on tax revenues. It could be considered as an indicator of liberalization level of the economy. Certain features of international trade make it more amenable to taxation than domestic activities. In developing countries, the international trade sector is typically the most monetized sector of the economy. Entrance and exit to the country takes place in specified locations. Thus import and export shares could be an important determinant of tax revenues (Karagöz, 2013). Share of agriculture over GDP: This variable is used as a proxy to control for the difficulty in collecting taxes. A strong negative relation between agriculture s share in GDP and tax revenue could be expected. In the developing countries, it is difficult to tax the agricultural sector, since an outsized part of it consists of subsistence and little farmers, notoriously tough to impose tax on the massive numbers that sell their merchandise in informal markets (Stotsky and WoldeMarian, 1997). On the other hand, since many public sector activities are urban based, a declining share of agriculture in GDP tends to be linked to an increase in demand for public expenditures and thus put pressure to raise tax revenue (Dioda, 2012). Also, if agricultural exports have a dominant role in the exporting sector of a country, it may lead to a positive relationship. Official exchange rate: Tanzi (1989) presented several wide-ranging hypotheses of the relationship between various macroeconomic variables, including exchange rate and tax revenue. He observed that there is often an inverse relationship between a country s tax revenue and the real level of its official rate of exchange. He argues that overvaluation has a direct effect by suppressing import and export bases measured in domestic currency terms. This reduces collections of international trade taxes and sales and excise taxes, which are usually levied on domestic and imported consumption. Overvaluation also has indirect effects by reducing the incentive to produce goods for export, encouraging capital flight and currency substitution, weakening the balance of payments, encouraging black markets, and encouraging trade restrictions (Agbeyegbe, et al., 2004) 4. 4 For more detailed see Reisen (1990) and Seade (1990)

121 114 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? Urbanization: Urbanization is crucial for its political, social and cultural features. It is expected that urbanization and tax revenue must exhibit a positive link. Since, it increases citizens demand and needs for public products and services (Tanzi, 1987). The density of population ought to be absolutely connected with tax revenue, because it leads to a reduction in the cost of tax collection and controlling for tax evasion (Ansari, 1982) and government s ability to collect taxes is enhanced by structural changes, which are concomitant with urbanization. Democracy: There is no consensus in the empirical literature supporting the significance of political variables such as the level of democracy and the duration of a political regime as determinants of tax revenue. On the one hand, in line with authors like Boix (2003) and Acemoglu and Robinson (2006), democracy is a significant factor for distributing income from the rich to the poor that creates an enlarged welfare state, and a stronger and more efficient tax system, based more on direct taxes than on indirect taxes. Also, under a non-democratic regime the size of the public sector would be relatively small, because a large part of citizens are excluded from the decision making process. Thus, a transition towards a democratic government would coincide with an increase in taxes and public spending in accordance with the theory of the median voter, moving in the direction of a better redistribution of wealth (Dioda, 2012). On the opposite hand, some authors like Barro (1979) and Wittman (1989), consider that the most drivers of public policy are not political factors. Also, Mulligan et al., (2004) did not find evidence that democracy can explain the changes in tax revenue. Based on literature, three common measures are exist about measuring democracy: The first measure of democracy is derived from the data on political rights published by Freedom House. The second measure is derived from the democracy index published in Polity IV. The third measure is the measure of democracy published in the International Country Risk Guide. Our focus attention leads us to the fact that most of the scant literature on democracy debate use Polity Project data. Therefore, in order to increase the credibility of our results, we use polity which is derived from the democracy index published by Polity IV. The Polity IV Project has rated the levels of democracy for each country and year using coded information on the general qualities of political institutions and processes, including executive recruitment, constraints on executive action, and political competition. These ratings have been combined into a single, scaled measure of regime governance:

122 Journal of Economic Cooperation and Development 115 the Polity score. The Polity scale ranges from -10, fully institutionalized autocracy, to +10, fully institutionalized democracy. The data is mainly taken from World Bank indicators. Due to reasons related to data availability, we restricted the analysis to the period and 83 developing middle-income countries (See appendix for included countries). Table 1 contains descriptive statistics for all variables. Table 2 provides the correlation between variables. Table 1: Descriptive Statistics of Variables Variables Mean Maximum Minimum St. deviation tax revenue GDP growth rate trade liberalization share of agriculture over GDP official exchange rate urbanization democracy Source: Own calculation Table 2: Correlation Matrix of Variables Tax Taxt-1 GDP Trd Agr Exc Urb Demo Tax Taxt GDP Trd Agr Exc Urb Demo 1.00 Source: Own calculation

123 116 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? In the first step of our analysis, it is crucial to ascertain the integrational properties of the data series. In a panel data model, when the presence of a unit root in a model is admitted, one may obtain apparently significant relationships from unrelated variables. This phenomenon is called the spurious regression problem. In order to test the data stationary and the order of integration of variables, we apply two conventional unit root tests, Im et al., (2003) and levin el al., (2002) (hear after IPS and LLC). These tests are widely known and understood, so we refrain from repeating the methodology here. The results in table 3 indicate that there is no presence of unit root. The IPS and LLC tests reject the null hypothesis of a unit root, showing that all variables used in the study are stationary at level. Table 3: Unit Root Test Variables IPS LLC tax revenue (0.000) (0.000) GDP growth rate (0.000) (0.000) trade liberalization (0.000) (0.000) share of agriculture over GDP (0.000) (0.000) official exchange rate (0.000) (0.000) urbanization (0.000) (0.000) democracy (0.000) (0.000) Note: P values in parentheses. Source: Own calculation Various econometric approaches have been used to estimate the before mentioned function relying largely on cross-sectional (and more recently, panel) data by OLS or GLS. However, the possibility of existing endogeneity of variables in a model using macro variables should be bear in mind. In this case, two stage least squares (2SLS) or generalized method of moments (GMM) are suggested as remedy.

124 Journal of Economic Cooperation and Development 117 GMM does not require complete knowledge of the distribution of the data and it ensures consistency and efficiency while dealing with heteroskedasticity and serial correlation. It undertakes the issue of persistence of tax revenue performance over time by including lagged dependent variable in the regression, since, a simple panel analysis, either with fixed or random effects, is generally not sufficient to fully investigate the lag structures inherent in a macroeconomic variable. It also handled the potential bias that could be created by inclusion of dependent variable in the regression. The particular approach we adopt is based on the GMM estimators for panel data model and is due to Arellano and Bond (1991). The Sargan test is designed to test the validity of the instruments, employed to estimate the model, by analyzing the sample analog of the moment conditions. It attempts to answer the question, given that a subset of instrumental variables is valid and exactly identifies the coefficients, are the extra instrumental variables valid? Failure to reject the null should be interpreted as favoring the specified model. Also, we apply Arellano Bond test for serial correlation in the first-differenced errors. Table 4 presents the outcome of examining the determinants of total tax revenue. Based on the Sargan test, the null hypothesis of the validity of the instrumental variable cannot be rejected, i.e. the instrument passes the test and errors are independently distributed. The results for 1 st and 2 nd order serial correlation report the fact that the assumption of serially uncorrelated errors is appropriate. Many of the estimated coefficients for the explanatory variables are in line with the predictions and largely coincide with previous findings in the literature and the overall fit of the panel model is also reasonable. First of all, it is interesting to note that the coefficient on the lagged dependent variable is positive and significant in the regression, suggesting that there is a sharp adjustment over time in the tax revenue. GDP growth as a good indicator of the overall level of economic development and sophistication of the economic structure shows positive and significant sign in revenue equation. As countries develop, they will improve their public administrations, judicial systems and promote structural and institutional reforms, so that, the costs of the tax system will be gradually reduced. Also, there is a shift in taxable income and taxpayers move into higher tax brackets, and this possibly results in

125 118 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? higher amount of personal income tax collection. This result is in line with Hinrichs (1966), Chelliah (1971), Tanzi (1992, 1987), and Ghura (1998) who found a positive correlation between the level of development and tax revenue. Table 4: GMM Estimation Variables Coefficient Z P> Z Lagged tax revenue GDP growth rate Trade liberalization Share of agriculture over GDP Official exchange rate Urbanization Democracy Wald Chi Sargan (1.000) A(1) (0.0051) A(2) (0.2709) Source: Own calculation *Sargan is asymptotically distributed as a Chi2 under the null of instrument validity, with p-value in parentheses. **A1 and A2 are tests for first-order and second-order serial correlation in the first differenced residuals, asymptotically distributed as a Chi2 under the null of no serial correlation, with p-value in parentheses. Openness as a proxy for trade liberalization has a positive significant effect on the total tax revenue. As an increase in the ratio of imports plus exports to GDP of one percent, increases revenue performance by up to percent. This is because, as countries liberalize their trade, trading with the rest of the world become relatively easier, raising the variety of goods and services that help boost corporate profit and the flow of goods and services within countries. In this way, trade-related taxes are easier to impose because the goods enter or leave the country at specified locations and tax revenue is mainly obtained from taxes on the exports and imports of country. The positive effect of trade liberalization on trade revenue is similar to the findings by Ebrill et al., (1999), and Khattry and Rao (2002).

126 Journal of Economic Cooperation and Development 119 The share of agriculture over GDP is statically significant and inversely related to tax revenue. One percent growth in the share of agriculture reduces tax revenue by 0.14%. The impact is relatively strong and it is in line with previous findings by Stotsky and WoldeMariam (1997). Agricultural activities are difficult to tax, because large part of the agricultural sectors are small-scale with limited number of taxpayers paying tax on income or profits and agricultural products are exempted from indirect taxes. Also, substantial part of the output is consumed and not marketed. The significant negative relationship between exchange rate and tax revenue indicates the fact that depreciation of the domestic currency leads to a decline in the volume of imports and hence leads to a loss of trade tax revenue. Also, it may cause to a reduction in the personal real income and consequently hamper the potential amount of income tax. Earlier studies have found that total tax revenue increases when a society becomes more urbanized, but somewhat surprisingly, it appears significant with negative sign. Democracy emerges significant and positively correlated with tax revenue. A more democratic and peaceful political regime enjoys more legitimacy and loyalty among taxpayers which leads to a higher degree of voluntary compliance. In low level of democracy, large part of the citizens may be excluded from the key decision making process, and perhaps there are few or practically no political parties that represent the interests of the electorate, being largely influenced by the vested interests of lobbies and elites (Grossman and Helpman, 1994). According to this view, elites exert power and pressure on political parties in order to defend their interests, and in particular to prevent taxes from rising and keep their special privileges and exemptions. 5. Conclusion and Policy Recommendation The macroeconomic consequences of trade liberalization have generated a great deal of debate among scholars in recent years. There are bodies of theoretical and practical research on the role of trade liberalization as a source of government revenue. The aims of this paper are twofold, at first we try to shed some lights on the basic question "Do middle income countries can benefit trade liberalization in order to promote tax

127 120 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? revenue?" Secondly, introducing those factors that matter most for explaining tax revenue. In so doing, to fulfill these aims we use a dataset, which includes 83 developing countries over time period covering the years from 1990 to The main conclusion of GMM estimation is that there exists a tradeoff between a greater degree of openness to international trade and the revenue collected by government. A set of factors that can potentially influence tax revenue has been divided into various groups like economic, socio demographic and political indicators. The empirical results of the panel model here built and econometrically tested indicate that GDP growth and democratic system of the economy are positively related to tax revenue in a statically significant way. The share of agriculture over GDP, official exchange rate and urbanization are also statically significant, but negatively associated with tax revenue. The result has important implications for countries that have been reluctant to undertake trade liberalization for fear of the revenue consequences. Bear in mind, there is not much economic policy can do to change an economy s relative structure (at least in the short run), for most countries no such limitations exist regarding openness. There is the need for government to emphasize the focus on liberalization in order to ensure the successful performance of trade liberalization policy and achieve the enormous revenue from trade tax. In addition, middleincome countries profit in terms of upper tax-revenue if formal activities, like the manufacturing sector, is growing faster than the agricultural sector. Developing countries must actively strive to increase the opportunities for more growth and development and improve the degree of democracy.

128 Journal of Economic Cooperation and Development 121 Reference Acemoglu, D., and Robinson. J.A. (2006), Economic Origins of Dictatorship and Democracy, Cambridge University Press. Adam, C.S., Bevan, D.L., and Chambas, G. (2001), "Exchange Rate Regimes and Revenue Performance in Sub-Saharan Africa," Journal of Development Economics, 64(1), Agbeyegbe, T., Stotsky, J.G., and WoldeMariam, A. (2004), "Trade Liberalization, Exchange Rate Changes, and Tax Revenue in Sub- Saharan Africa," IMF Working Paper, No Ansari, M.M. (1982), "Determinants of Tax Ratio: A Cross-Country Analysis," Economic and Political Weekly, 17(25), Arellano, M., and Bond, B. (1991), "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations," Review of Economic Studies, 58(2), Barro, R.J. (1979), "On the Determination of Public Debt," Journal of Political Economy, 87, Baunsgaard, T., and Keen, M. (2005), "Tax Revenue and (or?) Trade Liberalization," IMF Working Paper, No Boix, C. (2003), Democracy and Redistribution. Cambridge University Press. Chelliah, R.J. (1971), "Trends in Taxation in Developing Countries," IMF Staff Papers, 18(2), Dioda, L. (2012), "Structural Determinants of Tax Revenue in Latin America and the Caribbean, ," Comisión Económica para América Latina y el Caribe (CEPAL), No Dixit, A., and Norman, V. (1980), The Theory of International Trade. Cambridge University Press.

129 122 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? Ebrill, L., Stotsky, J., and Gropp, R. (1999), "Revenue Implications of Trade Liberalization," IMF Occasional Paper, No Epaphra, M. (2014), "The Revenue Implications of Trade Liberalization in Tanzania," Journal of World Economic Research, 3(3), Ghura, D. (1998), "Tax Revenue in Sub-Saharan Africa: Effects of Economic Policies and Corruption," IMF Working Paper, No Greenaway, D., and Milner, C. (1993), The Fiscal Implications of Trade Policy Reform: Theory and Evidence, UNDP-World Bank Trade Expansion Program Occasional Paper, 9, Washington, DC. Grossman, G.M., and Helpman, E. (1994), "Protection for Sale," American Economic Review, 84(4), Gupta, A.S. (2007), "Determinants of Tax Revenue Efforts in Developing Countries," IMF Working Paper, No Harrison, A. (1996), "Openness and Growth: A Time-Series, Cross- Country Analysis for Developing Countries," Journal of Development Economics, 48(2), Harrison, A., and Hanson, G.H. (1999), "Who Gains from Trade Reform? Some Remaining Puzzles," Journal of Development Economics, 59(1), Hinrichs, H.H. (1966), A General Theory of Tax Structure: Change During Development. Cambridge, MA: Harvard Law School International Program. Im, K.S., Pesaran, M.H., and Shin, Y. (2003), "Testing for Unit Roots in Heterogeneous Panels," Journal of Econometrics, 115(1), Karagöz, K. (2013), "Determinants of Tax Revenue: Does Sectorial Composition Matter?" Journal of Finance, Accounting and Management, 4(2),

130 Journal of Economic Cooperation and Development 123 Khattry, B., and Rao, J.M. (2002), "Fiscal Faux Pas?: An Analysis of The Revenue Implications of Trade Liberalization," World Development, 30(8), Levin, A., Lin, C.F., and Chu, C. (2002), "Unit Root Rests in Panel Data: Asymptotic and Finite-Sample Properties," Journal of Econometrics, 108, Longoni, E. (2009), "Trade Liberalization and Trade Tax Revenues in African Countries," Working Paper Series, No Muibi, S.O., and Sinbo, O.O. (2013), "Macroeconomic Determinants of Tax Revenue in Nigeria ( )," World Applied Sciences Journal, 28(1), Mulligan, C.B., Gil, R., and Sala-i-Martin, X. (2004), "Do Democracies Have Different Public Policies Than Non-Democracies?" Journal of Economic Perspectives, 18(1), Nwosa, P.I., Saibu, M.O., and Fakunle, O.O. (2012), "The Effect of Trade Liberalization on Trade Tax Revenue in Nigeria," African Economic and Business Review, 10(2), Pupongsak, S. (2009), The Effect of Trade Liberalization on Taxation and Government Revenue. A thesis submitted to the University of Birmingham for the degree of Doctor of Philosophy. Reisen, H. (1990), Interaction between the Exchange Rate and the Public Budget in Major Debtor Developing Countries. In Fiscal Policy in Open Developing Economies, ed. by Vito Tanzi (Washington: International Monetary Fund), Rodríguez, F., and Rodrik, D. (2000), "Trade Policy and Economic Growth: A Skeptic s Guide to the Cross-National Evidence," NBER Macroeconomics Annual, 15, Seade, J. (1990), Tax Revenue Implications of Exchange Rate Adjustment. In Fiscal Policy in Open Developing Economies, ed. by Vito Tanzi (Washington: International Monetary Fund),

131 124 Determinants of Tax Revenue: Does Liberalization Boost or Decline It? Stotsky, J.G., and WoldeMariam, A. (1997), "Tax Effort in Sub-Saharan Africa," IMF Working Paper, No Tanzi, V. (1987), Quantitative Characteristics of Tax Systems of Developing Countries. In D.M.G. Newbery and N. H. Stern Editions, The Theory of Taxation for Developing Countries, New York, Oxford University Press. Tanzi, V. (1989), "The Impact of Macroeconomic Policies on the Level of Taxation and the Fiscal Balance in Developing Countries," IMF Staff Paper, 36(3), Tanzi, V. (1992), Structural Factors and Tax Revenue in Developing Countries: A Decade of Evidence, in Open Economies: Structural Adjustment and Agriculture, ed. by Ian Goldin and L. Alan Winters (Cambridge: Cambridge University Press), pp Wittman, D. (1989), "Why Democracies Produce Efficient Results," Journal of Political Economy, 97,

132 Journal of Economic Cooperation and Development 125 Appendix Countries included in analysis Albania Ghana Panama Algeria Grenada Papua New Guinea Angola Guatemala Paraguay Argentina Honduras Peru Armenia Hungary Philippines Azerbaijan India Romania Belarus Indonesia Samoa Belize Iran, Islamic Rep. Sao Tome and Principe Bhutan Jamaica Senegal Bolivia Jordan Serbia Bosnia and Herzegovina Kazakhstan Seychelles Botswana Kiribati South Africa Brazil Kyrgyz Republic Sri Lanka Bulgaria Lao PDR St. Lucia Cabo Verde Lebanon St. Vincent and the Grenadines Cameroon Lesotho Sudan China Macedonia, FYR Suriname Colombia Malaysia Syrian Arab Republic Congo, Rep. Maldives Thailand Costa Rica Mauritius Tunisia Cote d'ivoire Mexico Turkey Dominica Moldova Ukraine Dominican Republic Mongolia Vanuatu Ecuador Morocco Venezuela, RB Egypt, Arab Rep. Namibia West Bank and Gaza El Salvador Nicaragua Yemen, Rep. Fiji Nigeria Zambia Georgia Pakistan

133

134 Journal of Economic Cooperation and Development, 37, 2 (2016), On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco EL Mostafa Bentour 1 Using input-output model, we analyze the effect of removing energy subsidies on prices in Morocco. We set three scenarios of increasing oil prices by 25%, 50% and 75%. We show that the effects on inputs prices are high in intensive oil products sectors such as transports, manufacturing industries, fishing and aquaculture, and, electricity and water sectors; with respectively 19.6%, 13.4%, 9.5% and 8.1% for 75% oil price increase scenario. Using the weights of the sectors, we deduce the overall cost increase generated by direct and indirect requirements for the total economy. An increase in oil prices by 75% generates an overall increasing cost between 5.5% and 8%. The generated cost-push inflation may alter the stable path of inflation recorded over the past fifteen years putting pressure on the monetary authorities. Therefore, the change of strategy from fixed exchange rate regime towards a flexible regime is an urgent option. 1. Introduction Reforming subsidies constitutes a real challenging task for the Governments regarding their economic, social and political implications. Economically, removing subsidies of imported products means increasing the cost of intermediary inputs consumption for producers and increasing the prices of the final goods for households consumption. This also means deteriorating the competitive advantage for domestic exporting firms from the producers' perspective. Socially, this has direct negative consequences on the households as it may erode their purchasing power and increases poverty and inequalities if such reforms have not been accompanied by policies mitigating these effects. Indeed, a recent study of the World Bank simulated that removing subsidies in Morocco, increases poverty by 4.4 percentage points, for an 1 The Arab Planning Institute, P.O.B: 5834, Shuwaikh Educational, Kuwait. embentour@gmail.com / mostafa@api.org.kw

135 128 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco equivalent poverty line of US$ 1.44 per day, and the Gini inequality index by 2.1 percentage points (Verme et al., 2014). Politically, lobbies against removing subsidies, represented by political opposition parties and associations, are motivated by personal and group interests and play an aggressive role against any attempt of reforming subsidies. The presence of a strong government coalition and consensus building at the time of reform is a prerequisite condition for its success (Sdralevich et al., 2014). The continuing increase of the cost of subsidized imported products in recent years due to soaring energy and food prices had push a big pressure on the Government to urge reforming such subsidies. The bill of subsidies has continuously increased from 1.3% as percent of GDP in 2002, to 6.7% of GDP in The main share is for petroleum products subsidy with DH48.2 billion (US$5.8 billion) representing 5.8% of GDP in 2012 (table 2). At the same time, the share of imported oil products increased from 0.4% in 1998 to 10.6% in Consequently, budget deficit including subsidies moved from -2.2% in 2009 to -7.3% in Without subsidies, this deficit was only about - 0.4% in 2009 and reached 0.7% in In percent of expenditures, subsidies represented 25.4% of total expenditures in 2012 overcoming the capital expenditures, which represent around 24.6% of the total expenditures. The current account deficit stood at high records in 2012, with -9.7%. Consequently, international reserves are reduced to less than three months of imports in 2012 and the debt ratio increased as the country resort to external financing. Concerning these high weights of petroleum products and the instability of oil prices, the reform of such subsidies becomes an urgent and mandatory option for the Government. The motivation of removing price subsidies is not only driven by the high burden cost that weighs on the Government budget, it is also argued that such system of subsidies does not benefit the poor but cover the rich, as well. In fact, many studies have shown that subsidies are pro-rich; approximately, the first quintile of the rich people benefits from the largest part of the total subsidies (Verme et al., 2014). For example, the poorest quintile benefits from diesel subsidies by only 1/6 the amount of the benefit of the richest quintile (7% for the poor versus 42% for the rich) (Sdralevich et al., 2014).

136 Journal of Economic Cooperation and Development 129 Recently, Morocco adopted a gradual approach as to the removal of subsidies especially petroleum products subsidies. Starting by the energy bill, a system of domestic prices indexation to international oil price was introduced in July In January 2014, the Government succeeded to remove the total subsidy for the "gasoline" product, followed by the removal of the total subsidy over the "diesel" product in January The subsidies of other oil products such as the industrial fuel have also been reduced but still not totally removed. This constitutes an important gain contributing mainly to the decline of the budget deficit from -7.3% in 2012 to around -5.4% in The total removal subsidy of gasoline and diesel has coincided with a downward of oil prices starting on September This has to avoid the inflationary pressures that may arise due to the removal of subsidies. Nevertheless, the assessment of such pressures due to the increase of the cost of production should warn policy makers from the adverse effects of inflation generated by large swings of international oil prices. The recent history of inflation evolution in Morocco has shown very low stable prices around 1.73% as an average over the period of , that we could qualify by a great moderation inflation era in the country. It is the lowest inflation rate amongst the MENA region whether this is due to the central bank successful policy in controlling domestic prices or other factors 2. Having said that, we use an input-output framework to analyze the passthrough effects of international oil prices to domestic prices as a result of removing subsidies in a small open oil importing developing economy, i.e. Morocco. This paper constitutes an added value to the empirical literature focusing on the inflationary pressures as a consequence of subsidy reform in an oil importing country in the MENA region. In what follows, we present in section two, a brief literature review of the effects of subsidies on prices and welfare. The third section describes the Input-Output models methodology used to assess the effect of the removal of subsidies on prices for twenty producing sectors in Morocco, as well as data and figures about the important weight of such 2 For details about this issue, an analysis is briefed in section 5: Discussion of monetary policy implications.

137 130 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco subsidies. The fourth section draws results and comments. The fifth one discusses the monetary and fiscal policy involvements and the sixth one concludes. 2. Review of Literature The role of subsidies in the economy is unclear. From different views of many economists, subsidy system could disrupt producers and consumers prices and welfare. The subsequent section discusses the effects of subsidies and their removal on prices and welfare for a relatively small, open and net energy importing country; i.e., Morocco. 2.1 The impact of subsidies on prices and welfare The major empirical studies and reports about subsidies are those of the IMF and the World Bank institutions. Canceling subsidies for an oil importing country means exposing local producers and consumers to the risks of prices instability for goods and services generated by international oil price variations. Many economists study the impacts of energy price swings. Examples of those are: Adjemian and Darracq Pariès (2008), Anand et al. (2013), De Gregorio et al. (2007), Elder and Serletis (2010), Finn (2000), Hamilton (1983, 2005), Hooker (2002), Hunt and Laxton (2001), Kilian (2008, 2014), OECD (2011), Peersman and Van Robays (2012), Tang et al. (2010), etc. On oil importing countries, the impacts vary across countries following the oil dependency of the economy, the degree of the development and the methodology of assessment. As example of a developed country for which the impacts were estimated, Hanson et al. (1993) analyzed the effects of international oil prices shocks on the agricultural sector in the United States, using input output and general equilibrium models. For a developing country, Lofgren (1995) used a CGE model for Egypt to analyze the short run effects of removing price-distorting subsidies for oil products traded locally and for other subsidized commodities. These effects are assessed to be negative on GDP, private income, consumption and employment while increasing government savings. As for the micro impacts, there was a reduction in oil use by 6 to 8 percent giving an additional push up to oil exports and reducing environment pollution. For the simulation of the consumer subsidy reduction of foodstuffs, this seems to be not very responsive due to low price and income elasticities for household food demand.

138 Journal of Economic Cooperation and Development 131 Recently, Coady et al. (2006) and Del Granado et al. (2010) used models based on input-output tables to assess the real income losses associated with reducing subsidies on fuel products. To assess the long run macroeconomic effects of subsidized energy products for a small, oilimporting developing country, Plante (2014) calibrated a general equilibrium model. The main conclusions are that, in the long run, the presence of subsidies increases the supply of hours worked and real wages leading to an over emphasis on traded goods while distorting the relative price of non-traded goods. By contrast, a general equilibrium simulation for Egypt economy involving 56 sectors with 11 energy sectors shows that energy subsidies shift resources from labor-intensive resources to capital-intensive resources leading to higher unemployment (World Bank, 2014). Accordingly, reducing subsidies in a similar economy should lead to higher labor and or wages especially in laborintensive sectors promoting the overall employment. Many studies and reports confirm the negative and direct role of the subsidies on the budget and the current account balance, threatening the sustainability of the public finances. The presence of subsidies also creates distortions on the economy and could increase inequalities. Especially, untargeted subsidies have tendency to favor a group of rich consumers over the poor, missing the target of protecting vulnerable people. About 1% in Egypt and Mauritania and less than 3% in Sudan (Sdralevich et al., 2014), and about 0.86% of diesel and gasoline in Morocco (Verme et al., 2014) are the shares of petroleum products subsidies consumed by the poorest 20% of the population. These figures show that energy subsidies do not benefit the poor. The positive effects of the subsidies are assumed to benefit the final consumers through lowering prices of energy and food, and benefit the producers by reducing the cost of intermediate inputs. As a result, subsidies cushion the direct price increase effects of imported goods as well as the subsequent price increase generated by the cost of the imported inputs. Despite the fact that energy subsidies are, effectively, pro-rich, their removal in the short term, could be highly felt by the vulnerable population, i.e. the middle and poor classes. The adverse effects are felt, for example, by the likely inflationary pressures due to price increase, resulting in income losses and other associated effects. The concerned

139 132 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco sectors are generally those of intensive using of petroleum products, such as transport, electricity and water production, food processing industries and manufacturing industries. The short run effects could be extended to medium and long terms effects via wages-prices spiral and due to subsequent higher expectations of future inflation. Furthermore, exporting sectors could see their foreign competitiveness threatened due to an increase in the production costs because of the removal of subsidies. This is why reforming subsidies is advisable to be gradual and accompanied by policies cushioning such negative externalities. Furthermore, removing subsidies could increase energy smuggling across neighboring countries, due to price differences between these countries (El-Katiri and Fattouh, 2015) Pass-through of oil prices from producer prices to consumer prices The prices effects of cancelling subsidies of oil products on the final consumers are of two origins: The first effect is when using these oil derivatives in their final consumption. Before removing subsidies, the price difference is supported by the government budget by compensating suppliers in charge of refining and distributing petroleum products. The second effect reflects the producers increased costs following the removal of subsidies on other products and services using oil products as inputs, such as transports, food and other manufacturing products. In the second case, both consumers and producers carry the charge of the price increase. The producers will try to pass on the maximum part they could of the increase to the consumers. However, this depends on the price elasticity of demand of the consumers to producers products reflecting the responsiveness to prices increase. Figure 1 explains the situation where producers and consumers share the burden of removing subsidies. We define P as the price by which the quantity Q is supplied before removing subsidies and P 1 and Q 1 respectively the new price and quantity supplied in the new situation of canceling oil subsidies. The reform shifts the supply curve (S) to the left, and the new supply curve (S without Subsidies) intersects with the demand curve (D) in the point (b) resulting in a higher price P 1 (reflected by the impact from international oil price increase) and lower quantity Q 1. We define P 0 as the price of the new quantity Q 1 if subsidies would have been kept.

140 Journal of Economic Cooperation and Development 133 Figure 1: Producer and consumer price changes and welfare after energy reform Prices S without P 1 b S P c a P 0 f D Q 1 Q Quantities For the new equilibrium "b"= (Q 1, P1), the load on the consumers is determined by the area bcpp 1 equal to (P 1 -P)*Q 1, and the load on the producers is defined by the area fcpp 0 equal to (P-P 0 )*Q 1. The division between consumers and producers depends on the price elasticity. The welfare gain for the government from this reform is reduced by the amount of tax base decrease due to the quantity reduction (Q-Q 1 ). See for example, Coady et al. (2015) for further explanation about fiscal and welfare gain from subsidy reform. At the aggregate level, the point "a" is the equilibrium of the long run determined as the intersection between the aggregate demand and the aggregate supply. Once subsidies are removed, the new equilibrium is in "b" and is characterized as a short run equilibrium where prices are high and the aggregate output is reduced. This situation is called stagflation as a result of a coexistence of high inflation and output below its normal potential. If the aggregate demand stands stable, the return to the long run equilibrium in "a" could take place if salaries decreased, allowing

141 134 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco firms to reduce costs and re-expand the capacities. However, this equilibrium is not always immediately reachable by an automatic decrease of salaries driven by market forces. High commodity prices are likely to trigger a wage-price spiral, as labor unions will ask for higher salaries following a response to higher consumer prices. In developing economies, market competition is generally less intensive and labor unions less organized and less strong. This makes producers more likely to pass on increased costs to consumers by increasing prices. Thus, reform subsidies are more likely inflationary in developing than in developed countries. Other transitional effects could be the involvement of firms in efficient energy use and the shift from producing intensive labor goods towards intensive capital goods. To sum up, many studies have revealed that oil shocks can cause inflation. Oil price increase combined with domestic energy prices deregulation will directly upraise the cost of production, especially across energy intensive sectors, disrupting firms from raising their investment capacities if not reduced. The higher cost of inputs will be passed on to consumer prices, increasing the headline consumer price index (CPI). The pass-through to the headline CPI, if oil prices were sustained, could be transmitted to the core inflation by raising inflation expectations especially if these expectations are not well anchored. The surge of inflation will reduce households real income and consequently will reduce their consumption. This situation of high prices and disrupted production capacities results in higher unemployment and difficult policy implications. 3. Methodology and Data 3.1. Input-Output based model methodology Based on Input-Output techniques, each economy is disaggregated into a number of sectors, n. In order to produce, each of the n sectors uses some or all different commodities produced by the other sectors as inputs including its proper commodity. The coefficients link the intermediate inputs used by others sectors represented by a ( n n) matrix, A ( ) a ij n n, commonly called the technological matrix or technological coefficients of the entire economy. So, for a fixed i and j

142 Journal of Economic Cooperation and Development 135, a ij is the proportion of input i required to produce one unit of the good j. Using the input output tables of a country, several simple formulae are developed to answer important questions such as; calculating inputs required for a given vector of final demands or, calculating final production for given available inputs or also assessing the impacts of an input or several inputs prices changes over the other inputs prices. The latter is known as the price input-output model. Noting X (n 1) a vector of total produced output by the n sectors of the economy and D (n 1) the vector of final demand, leads to: X-AX D (1) Where AX is the part of the production served as input (intermediary consumption) and the left hand side is then, net production. Assuming that (I-A) is a nonsingular matrix, the production vector is deducted as: -1 X (I-A) D (2) For the price input-output model, we follow (Hanson et al., 1993). The derivation of change in unit costs, to measure direct and indirect cost linkages in an input-output model starts from the definition of cost prices: P A'P VW (3) Where, P is an n-element vector of sector prices, V is an n-element vector of value-added coefficients, W is an n-element vector of value added prices and A is a transpose of the technological matrix A. Each sector output unit is allocated to purchase the intermediate inputs from other economic sectors ( A'P ) and the primary factors (VW ) (wages). Deriving the equation according to the price P yields: dp A'dP VdW (4) Rearranging the equation above yields: dp -1 (I - A') VdW (5)

143 136 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco In assessing the impact of price changes of sectors over other economic sectors, the nature of the impact differs whether the price of the studied sectors should be treated as exogenous or endogenous. For example, for many commodities, such as oil products, prices are internationally determined and are treated as exogenous for importing countries. In practice, treating the k sectors, experiencing the price shock as exogenous, in the first step, is eliminating the k sectors from the structure of the production by removing their corresponding rows and columns from the technological matrix A. the second step is to include the removed rows of the direct requirements of the k exogenous sectors in the value added vector V. This allows the estimation of the effect of the prices shocks on cost prices of other sectors (or sectors' prices if all the higher costs are passed on to buyers). The reduced technological matrix is of dimension ( n k) by ( n k) and V is now a ( n k) by ( 1 k) matrix. Assessment of the exogenous price shocks effects is deduced by applying the precedent formula (5) to the new matrices Data and subsidy burden in the economy Morocco published input-output tables for the period , issued by the High Commission for Planning 3. The data are disaggregated into 20 economic sectors. The definitions and codes of these sectors are presented along with the weight of each sector in table 1. The weight for each sector is defined as its total use of all products inputs to the all inputs used by the economy to produce the whole output. Table 1 presents also the refined oil and other related products intensiveness of the 20 sectors of the Moroccan economy. We choose to present data for three input-output tables by selecting three years, where oil prices are low (1998), medium (2005) and high (2012). The intensiveness in oil products, for each sector, is defined as the intermediary consumption of refined oil divided by the total inputs used by that sector in the process of production. 3

144 Journal of Economic Cooperation and Development 137 The most intensive oil sectors are fishing and aquaculture, transports, electricity and water, mining industries, other manufacturing industries and trade and repairing. Taking the weights into accounts, which are varying between the three years, the most consuming sectors in refined oil products are transportation, mining industries, manufacturing industries, electricity, and water production sectors, with differences between 1998, 2005 and 2012, due to the weights variation.

145 138 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco Table 1: Oil intensiveness, sectors' weights and weighted intensity in percent (%) Rank Code Economic sectors Oil intensity i Sectors' weight ii Weighted intensity iii Oil intensity i Sectors' weight ii Weighted intensity iii Oil intensity i Sectors' weight ii Weighted intensity iii 1 A00 Agriculture, hunting, & related services B05 Fishing and aquaculture C00 Mining Industry D01 Food Industry and Tobacco D02 Textile industry and cook D03 Chemical industry and para D04 Mechanical, electric & metal industries D05 Other manufacturing industries D06 Oil refining and others E00 Electricity and water F45 Building and public works G00 Trade and repairing H55 Hotels and restaurants I01 Transportation I02 Post and telecommunications J00 Financial activities and insurance K00 Real estate, rental & services to enterprise L75 General public administration & security MNO Education, health and social work OP0 Other non financial services Total Source: Author s calculation from the input-output tables for the Moroccan Economy for the years 1998, 2005 and i: The oil intensity or intensiveness is the quantity of refined oil used as input among other inputs. ii: The weight for each sector is defined as its total use of all products inputs to the all inputs used by the economy to produce the whole output. Following the same n n n notations in section 3, we formulate the weight i Wi A ij A ij. j 1 i 1 j 1 iii: weighted intensity is the oil intensiveness taking into account the weight of the sector in the economy in terms of using inputs. It is the results of the oil intensity multiplied by the weight: WI i OI i W i. W for each sector i as:

146 Journal of Economic Cooperation and Development 139 Table 2 summarizes some interesting facts about the food and energy subsidies trends in Morocco as well as the associated imports of energy products and oil prices. The first part of the table shows the total subsidies, which was about DH5.835 billion in 1998, representing 1.5% of total GDP, 5.4% of total imports and about 9.1% of government expenditures. These figures moved up following an ascending trend of international commodities prices (oil and food prices), recording more than DH55.6 billion in 2012; which is equivalent to 6.7% of GDP, 13.4% of total imports and 34.9% of total government consumption. By category of subsidies products, the second and the third parts of the table 2, show the importance and the evolution of food and energy products subsidies in the economy. We observe that the trends of shares in total subsidies and in GDP are reversed, by time, between the two principal categories of subsidized products; food versus energy products. Food subsidy, which was about 90% of total subsidies in 1998, decreased to around 13% in 2012 and 15% in Their level grows by only 1.2% on geometrical average over the past fifteen years while the total subsidy grows by 17% over This huge growth is dragged by the energy subsidy that grows at an annual geometric rate of 38% following a high-recorded oil prices especially in 2011, 2012 and The oil imports have considerably increased as a share of GDP from 0.6% in 1998 to 7.4% in 2012 following a sustained increasing trend in oil prices over this period. The important increasing of the bill energy compared to the food and the sensitivity of vulnerable population to the food subsidy are the principal reasons behind starting with reforming energy subsidies first.

147 On the Removal of Energy Products Subsidies in an Importing Oil Country: 140 Impacts on Prices and Policy Implications in Morocco Table 2: Evolution of subsidies in levels and its shares in percent of main economic aggregates Total subsidies Food products subsidies Energy products subsidies Oil imports Oil Prices (Brent, Europe) U.S. $ per barrel % of years GDP Millions of % of % of % of government Millions of % of % of Millions of % of % of total Dh GDP Imports consumption Dh total GDP Dh GDP , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Sources: Author s calculation from Input-Output tables, and the International Energy Agency website for oil prices.

148 Journal of Economic Cooperation and Development 141 The subsidy reduction policy has started in July 2013, allowing for reducing the subsidies recorded in 2012 by about DH13 billion (US$1.6 billion) representing 2.1 percentage of GDP. The year 2015 is expected to witness a reduction in the energy subsidies of more than 80%. This percentage reflects approximately the share in the total energy products of the gasoline and diesel products which have been totally unsubsidized starting from January 2015, as well as, and other fuel products partially liberalized. 4. Results and Comments Morocco imports all its energy products from which the major share is crude oil. The country is supplied by the domestic refinery, SAMIR 4 with a refining capacity of 150,000 barrel/day, and a storage capacity of 2 million m 3. Therefore, international prices are directly reflected in the cost of production before tax and subsidies. 4.1 Treating oil sector as endogenous We consider three scenarios for the rise of oil prices from their observed levels as baseline scenarios by respectively 25%, 50% and 75%. Regarding the tendency of the asymmetry found in most of the empirical literature 5, we couldn t expect any results for the oil price decrease. Assuming that removing subsidies is equivalent, in an oil importing country, to exposing the domestic producers and consumers of energy to direct impact of international prices, we translate the three previous scenarios of oil changes as changes in the proportions of refined oil sector (sector 9, table 1) used by other sectors. The assessment is, therefore, to increase the intermediary consumption of other sectors in terms of oil products, and applying the formula (5) (section 3.1) to study the impacts. We consider the oil sector as endogenous at this level. Assuming that the input-output tables structure linkages could change over time leading to instable coefficients and multipliers, we choose to assess the scenarios effects considering three input-output matrices: the 4 =2&Itemid=104 5 See for example (Borenstein et al. 1997) and (Kristoufek and Lunackova, 2015) for a survey of the articles on the asymmetry subject.

149 142 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco first of the period is the table for the year 1998, the middle is for 2005 and the last of the period is for The properties of the selected matrices correspond to three different levels of actual recorded oil prices: low level in 1998 (12.8 US$ per barrel), medium level in 2005 (54.6 US$ per barrel) and high level in 2012 (111.6 US$ per barrel). This allows diversifying scenarios results under different actual prices. Table 3 presents the results for the studied scenarios for the three matrices. The first results show that in the absence of subsidies, international oil prices affect the most intensive sectors in oil products such as transportation services, electricity and water production, industries such as food processing, manufacturing and mining sectors. For example, transport prices have increased by around 10% in 1998 and 22% in 2005 and 2012 following a scenario of an increase in oil prices by 75%. We also show that, the effects are more pronounced in recent economic structure (2005 and 2012) than in fifteen years ago (1998). This could be a mixed effect of low level of actual oil prices in 1998, as well as the difference of structures of the productions (weights of sectors in term of total inputs use and intensiveness of energy use). Furthermore, observed linear movements of effects between the three scenarios are automatic due to the linear form of the input-output models.

150 Journal of Economic Cooperation and Development 143 Table 3: Scenarios effects of oil price increases by respectively 25%, 50% and 75% on the structure of production inputs of 1998, 2005 and Effects on the structure of 1998 (%) Effects on the structure of 2005 (%) Effects on the structure of 2012 (%) Increases by Increases by (%) Increases by (%) Rank Economic sectors 25% 50% 75% 25% 50% 75% Agriculture, hunting, and related services Fishing and aquaculture Mining Industry Food Industry and Tobacco Textile industry and cook Chemical industry and para Mechanical, electric and metal industries Other manufacturing industries Oil refining and others Electricity and water Building and public works Trade and repairing Hotels and restaurants Transportation Post and telecommunications Financial activities and insurance Real estate, rental and services to enterprises General public administration and security Education, health and social work Other non-financial services

151 On the Removal of Energy Products Subsidies in an Importing 144 Oil Country: Impacts on Prices and Policy Implications in Morocco 4.2. Results for exogenous oil sector In the previous results, we count for oil refining sector as endogenous. However, it is more suitable to consider oil swings as foreign shocks for an oil importing country. We, therefore consider the oil sector exogenous, following the approach of Hanson et al. (1993). Using this approach, we produce three scenarios of increase effects and, for the two matrices of 2005 and The prices changes are quite reduced compared to the situation where we considered the oil sector as endogenous (Table 3). To deduce the overall increasing inputs cost, we account for the weight of each sector in the total economy in terms of intermediary consumption. The contribution of each sector to the overall cost increase is its price change multiplied by the corresponding weight. The overall increase is then a sum of the all sectors contributions; a weighted sum of the 19 sectors prices change excluding the oil sector as it is exogenous. The results, presented in Table 4, are sorted from largest to smallest sector's change based on 2012 results. For a scenario of 75% of oil price increase, the most affected sectors are transportation by 19.6% (1% as contribution), other manufacturing industries by 13.4%, fishing and aquaculture by 9.5% and electricity and water by 8.1. Furthermore, weights, as for oil intensiveness, play an important role in contributing to the overall production cost. For example, in 2012, fishing and aquaculture sector which have the highest oil intensity of 49.1% is the least weighted sector by 0.8% (Table 1) making its contribution to the overall cost increase insignificant (0.1%).

152 Journal of Economic Cooperation and Development 145 Table 4: Prices change of the inputs production structures of 2005 and 2012, following exogenous shocks on the refined oil sector by respectively 25%, 50% and 75%. Effects of oil price increase on 2005 inputs structure table for Effects of oil price increases on 2012 inputs structure Weight i % ii Weight i % Economic sectors Scenarios effects (%) Weighted effects (%) Scenarios effects (%) Weighted effects (%) 14 Transportation Other manufacturing industries Fishing and aquaculture Electricity and water Building and public works Trade and repairing Mecanical, electric and metalurgical industries Mining Industry Food Industry and Tobacco Textile industry and cook General public administration and security Chemical industry and para Post and telecommunications Hotels and restaurants Agriculture, hunting, and related services Other non-financial services Financial activities and insurance Education, health and social work Real estate, rental and services to enterprises Total Weighted Inflation (oil sector exogenous) i: Weights Total Weighted Inflation (oil sector endogenous) in table 4 are a little different from those presented for 2005 and 2012 in 2.4 1, as 5.1 they are 8.0 newly calculated to adjust only sectors after 3.6 the 5.5 withdrawal of oil sector considered as exogenous in this table. ii: All the data in the table are sorted from largest to smallest according to the column reporting the effects of oil price increase by 75%.

153 On the Removal of Energy Products Subsidies in an Importing 146 Oil Country: Impacts on Prices and Policy Implications in Morocco 5. Monetary and Fiscal Policy Involvement In September 2014, oil prices have started to decline from their high levels above 100 US$ and reached the level of 50 US$ at the beginning of the year This constitutes a huge decrease of more than 50% in just a quarter of year. In the beginning of 2016, the prices decreased to around 30 US$. This constitutes a huge decrease of more than 70%. The decrease is a consequence of supply and demand shocks. Form a supply side, United States production capacities augmented. From a demand side, world growth is expected to slow in The recent oil prices downward trend coincides with the total removal of subsidies over gasoline and diesel products in Morocco. This avoided to the economy, the inflationary pressures that could have occurred had the oil prices continued with their high levels of Economic policies and inflation trend in Morocco By the expected back of oil prices increase, inflationary pressures will arise which could threaten the great moderation era of Morocco. Indeed, the country has enjoyed low stable inflation rates over the period The average inflation rate over this period is 1.7% bounded with a minimum of 0.6% in 2001 and a maximum of 3.9% recorded in It is the best stable inflation path in the MENA region in terms of average and variability (table 5). The important questions that remain are: what has to be done, henceforth, to cushion the inflationary effects? Does the Central Bank in its current status armed to control such additional cost inflation?

154 Journal of Economic Cooperation and Development 147 Table 5: Inflation rates distribution over the period for the Middle East and North African Countries. Mean STDEV Minimum Maximum Range Morocco Tunisia Kuwait Bahrain Saudi Arabia Mauritania Djibouti Jordan Oman Lebanon United Arab Emirates Algeria Pakistan Egypt Syria Qatar Libya Iran Yemen Afghanistan Iraq Sudan Source: Calculated from, IMF World Economic Outlook Database, October Data are sorted from smallest to largest according to Standard deviation column (STDEV) Generally, some believe that the great moderation era is not due only to the success and achievements of the Central Banks; it is rather import prices and wages moderation (Perry and Cline, 2013). In Morocco, this performance is seen by some as a forced stability at the expense of sustainable growth and full employment promotion (Taouil, 2010). The monetary policy and the fiscal policy altogether have contributed to insufficient growth and low level of employment achievements. The country maintains an average growth rate of 4.5 percent over the last two decades. This is judged insufficient to absorb radically high unemployment rates and alleviate poverty (World Bank, 1995; 2006). Despite a downward trend, the unemployment rate still stand at around 10 percent.

155 148 On the Removal of Energy Products Subsidies in an Importing Oil Country: Impacts on Prices and Policy Implications in Morocco The insufficient achievements of both monetary policy and fiscal policy sounds, to some extent, arguable if we examine some monetary aggregates, such as, the public debt and the lending interest rate. The composition of the public debt and its trend are characterized over the period by a growing share of debt acquired from the domestic market and a decreasing share of the public foreign debt as described in Figure 2. This gives a signal of a constriction of the credit market by the Government. Figure 2: The evolution of treasury debt in Morocco, as percent of GDP. Total Treasury debt, % of GDP Source: Data are from The Ministry of Economy and Finance, Morocco ( As for the real lending rate, the real cost by which the private sector is financed from the banking sector, it is the among highest in the MENA region, despite an opposite low level of Central Bank policy rate compared to the MENA region (Figure 3). This constitutes a brake on the demand and access to credit especially for SMEs, which affects the growth and thus reduces inflation by demand shocks. The classic response to the positive demand shocks that increases the inflation and output is tightening the monetary policy. However, for a foreign positive price shock, the implications are only inflationary and their effects are negative on output making the task not straightforward for the Central Bank (De Gregorio, 2012).

156 Journal of Economic Cooperation and Development 149 Figure 3: Average nominal and real lending rates over the period for selected MENA countries (%) Nominal lending rate, % Real lending rate, % Source: Constructed by the author from World Development Indicators database, the World Bank. The fiscal policy have also played an important role by subsidies, as proven in this paper, in cushioning the effects of high prices of imported energy and food products. Consequently, the low level of inflation and its stability in Morocco could not be solely the merit of the monetary policy Discussion of the pass-through from headline to core inflation Inflation is by definition a sustainable increase in the general level of prices. Starting from this point, could we argue that removing subsidies is by nature inflationary process? Especially that Central Banks target core inflation calculated removing volatile products such as energy and food products? What factors could ease the pass-through to core inflation? Central Banks mostly focus on core inflation instead of headline inflation. In calculating core inflation, Central Banks excludes food and energy prices from calculations, as they consider that their prices changes are mainly driven by short-term fluctuations generally in supply side and could not fuel the formation of economic agents expectations