The effectiveness of external financing sources on economic growth case of the developing countries of the MENA Region

Transcription

1 The effectiveness of external financing sources on economic growth case of the developing countries of the MENA Region Miss. Manelle LAHDHIRI Doctor in Economic Sciences, Faculty of Economic Sciences and Management of Sousse University of Sousse -TUNISIA Mr. Mohamed Amine HAMMAS Conference Master, Faculty of Economic Sciences and Management of Sousse University of Sousse -TUNISIA Abstract Face to the difficulty of achieving the Millennium Development Goals (MDGs), due to lack of sufficient funding and the dramatic drop of the Official development assisstance (ODA). One wonders about how to overcome this decline through the exploitation of other international financing resources for development, for example through the use of financial transfers from emigrants (remittances) and foreign direct investment (FDI). Building on this new basis of international development finance and given the trend decline in ODA, we seek in this work to test the effectiveness of the external financing resources on economic growth in developing countries of the MENA region. JEL Classification :F35 ;B2 ;F22 ;F23 ;F24 Key words: Economic growth, official development assistance; remittances, foreign direct investment, developing countries and MENA region. 1. Introduction The MENA region is mainly composed by developing countries where remittances are a significant external financial flows. Given this importance of the remittances (WR) and given the decline of the official development assistance (ODA), international financial institutions (IMF and World Bank) are wondering to what extent remittances from emigrants they could serve as a source of financing for economic growth in these countries. According to the World Bank, these financial flows have largly increased in this region, they spent from 861 to 4759 million between 2000 and This increase has continued even in 2008, followed by a slight decline in 2009 and then a return to increase in However; during the same period ; the Official Development Assistance (ODA) has crossed a crisis related to a combination of several factors 24. This crisis has led to the questioning of ODA fundamentals, accompanied by a sharp fall of its flows to developing countries. 24 As the changing environment marked by the end of the Cold War and the problem of indebtedness growing countries receiving assistance. 1228

2 This challenging environment has revived the debate about the effectiveness of ODA by the publication of a World Bank s report : "Assessing Aid" based on the work of Burnside and Dollar ( ).This works supports the thesis that the effectiveness of ODA s impact on economic growth depends on economic policies quality of the ODA s receiving developing countries.thus, this work paves the way for selectivity criteria 25 that should be applied to these countries. However ; despite this conditionality attached to ODA flows by the criteria of selectivity ; ODA s efficiency isn t always achieved. Indeed, its impact remains are limited both in terms of growth and in terms of reducing poverty 26 This thesis has been adopted by the International Financial Institutions 27 (IFIs), witch encourage the financing of the MDGs through the use of the remittances, and witch announce that remittances have grown faster than FDI. Therefore; they argue for the assumption that ; if the remittances wich are sent to their families living in the devlelopping countries ; are effectively and productively used by these countries, they could be an effective factor in reducing poverty and stimulating economic growth. However, all the studies mentioned above have been criticized because of their limitation to test the impact of one extern financial resource on promoting economic growth and development in the receptor country and ignored the potential role of the other sources. Moreover, comparative studies on the effectiveness of the three finacial development resources are poorly theorized. Especially, despite the growing importance of remittances in total international capital flows the relationship between remittances and economic growth has not been sufficiently tested. Following this finding ; the aim of our research is to determine what would be among these three external resources of financing development ; the most effective and / or what would be the combination of these sources, which would be the most optimal to support economic growth and development in developing countries in the MENA region. To do this, we construct a new model that multiplies the resources of international capital flows to identify the effect of their mixing on stimulating economic growth and to show those remittances; both through formal channels and informal channels 28 ; could be a significant source of development of finance of developing countries the MENA region. 25 This principle states to give more ODA to developing countries who are making political, economic and institutional reforms in order to improve their "good governance ". The basic idea is that improvement efforts of "good governance" greatly increases the productivity of ODA, this same productivity is insignificant or even negative in poorly governed countries (World Bank, 1998;Burnside and Dollar, 2000 & 2004). 26 Jacky Amprou et Cauvet Lisa, (2007), Débats sur l Efficacité de l aide : fondements et nouveaux enjeux, Agence française de développement, The World Bank and FMI 28 Informal channels of the remittances: Such transfer eludes official statistics and does not account banking, especially in terms of profit, they are items which are directly or through stakeholder informal (friends, family, hawala, etc)

3 The outline of this paper is as follows: Section 1 shows the three sources of international capital flows, and describes their evolution over the past two decades in developing countries of the MENA region.in section 2 we review the theoretical and empirical literature of ODA, remittances and FDI s roles in financing the development in the developing countries. Empirical results are presented and discussed in Section Sources of external financing development Under the current debate on financing the Millennium Development Goals (MDGs) and on their achievement in developing countries of the MENA region, three main sources of external funding are preferred by the IFIs: remittances (WR), official development (ODA) assistance and foreign direct investment (FDI) The remittances Remittances have been increasingly important because they constitute a stable source of funding for all developing countries, especially those from the MENA region. The new data show that remittances officially recorded in the destination developing countries developing countries have now reached $ 307 billion. Indeed; because of their quantities to the developing countries; remittances have become increasingly essential as the IFIs, who regard them as a potential source of external financing for development.these remittances have reached significant levels in all regions of the world and contribute to financal growth at different levels from one region to another. According o the World Bank, these flows will increase by 6% in 2010 to $ 325 billion, 6.2% in 2011 and 8.11% in 2012 to $ 374 billion dollars. The flows of the remittances from migrants to developing countries are as follows: Table No 1: According to this table, we note that the MENA region attracts attention with the highest rate of 1.6% as a percentage of GDP and ranks first on the list of beneficiary developing countries from remittances.the World Bank predicts an increase subsequently. These forecasts are represented in the table below: 1230

4 Table No 2: Source: Author s calculation based on data from IMF Balance of payments Statistics Yearbook 2008 and Data realises from central banks, national statistical agencies, and Word Bank country desks It should be noted that the MENA region has indeed had a significant increase in remittances in recent years; however; this development differs from one country to another, Lebanon, Jordan and Egypt are the countries that have received benefits from these flows as outlined in the following diagram. 1231

5 However, the data already presented; are underestimated and don t correspond to the exact mount of transfers sent by migrants to their countries.this reflects the fact that the flow of remittances recorded by the IMF in the Balance of payments are only part of what is actually transferred, because they considered only the transfers sent to developing countries through e channelsofficiels 29. This accounting therefore neglects the remittances through informal channels, because this type of transfers eludes official statistics and does not require a bank account, especially for he bénéficiaires 30.Thus, by their volume, remittances could be a potential source of external financing in many developing countries of this region.it should be remembered that remittances from migrants include three types of monetary transfers between countries, wich are reported by the International Monetary Fund (IMF) in its annual statistical yearbook of the balance of payments. They are remittances by migrant workers, compensation paid to employees and remittances migrants 31. The international organization of migration 32 (IOM) defines remittances as "The remittances that migrants make to their destination country."remittances have become a popular subject in the literature on external financing development and economic growth, because of their size and their potential to promote economic growth in developing countries. On this basis, the developing countries in the region MENA is an example of this potential evidence.firstly, due to the fact that remittances have continued to increase either through official channels or through unofficial channels; and secondly; because of the increased number of migrants from the MENA countries, especially to countries of the European Union. This explains the fact that four developing countries in the MENA region belong to the list of the top beneficiaries in terms of the share of transfers in GDP in 2008 and in terms of transport volume U.S. dollar in 2009, as presented by both charts below. 29 Official channels: include bank transfers 30 Op cit n 8 p2 31 OCDE (2009), perspectives des migrations internationales SOPEMI IOM (2005) (International Organization for Migration), World Migration Geneva. 1232

6 2.2. The official development assistance The official development assistance (ODA) includes grants and loans to developing countries whichare given to privileged financial conditions imposed by public agencies of the Organization for Cooperation Economic Development (OECD) met in the Development Assistance Committee (DAC). According to this committee, ODA represents all inputs and resources which are provided to developing countries, including the grants which are conditioned by the following criteria: (i) ODA must be provided by government agencies, including state and local governments or organizations acting on behalf of public agencies. (ii) ODA must have as objective that s the economic promotion and development and improve the living standards in developing countries. ODA can be either bilateral (if they are intended directly to a developing country) or multilateral when they pass through an international organization, such as the World Bank or the European Commission. This ODA, comes entirely from DAC member countries through two channels: The first is directly given by the Member States as part of their bilateral aid. The second is managed by the multilatéral institutions Multilateral institutions: the World Bank and International Monetary Fund. 1233

7 Bilateral ODA represents about two-thirds of ODA and multilateral assistance accounts for the rest. This assistance has declined remarkably during the last decade. Indeed, official development assistance granted by 22 countries member of OECD belonging to the Assistance Committee (DAC), which totaled $ billion in 2007, are down from billion in 2006 to billion dollars in This decline in official development assistance (ODA) in recent years has affected all developing countries without exception. This decline began in the 1990s, following the crisis of legitimacy of the ODA and it is largely explained by the strategy implemented debt relief following the two extraordinary agreements Paris35. In developing countries in the MENA region, the fall in ODA has been remarkable and has been significantly accompanied by a large increase in the flow of remittances to these countries especially during the period , as shown in the following graph: 2.3. The foreign direct investment The International Monetary Fund (IMF) defines foreign direct investment (FDI) as "the investments done by a resident entity in one economy (direct investor) in order to acquire a lasting interest in an entrprise resident in another economy (the company direct investment). 34 According to the report of the World Bank : «Global Development Finance» published in Paris s extraordinary agreements: ODA, debt and international trade placed in the center of the High level Meeting on Financing for Development June ; AG/10364 press release. 1234

8 The lasting interest implies the existence of a long-term relationship between the direct investor and the company, and the investor exerts a significant influence on the management of this company. The direct investment comprises not only the initial transaction establishing the relationship between the investor and the entrprise, but also all subsequent transactions between them and among related entrprises, whether or not incorporated and therefore possesses a personal separate legal entity. " FDI represents all the resources (inputs to the capital; endowments, loans, making available cash, trade credit or reinvested earnings) that a direct investor leave available to companies with which there is a direct investment relationship.the OECD defines Foreign Direct Investment as follows: "FDI is an activity in which a resident Investor in one country obtains a lasting interest and a significant influence in the management of an entity resident in another country. "So ; in general; FDI is an important form of international capital movements, through which the business of a country creates or develops a subsidiary in another country. According to IMF data, foreign direct investment (FDI) in developing countries reached a record $ 325 billion in 2006 up by 44 billion dollars compared to 2005.Inflows of foreign direct investment reached 2.9% of GDP of developing countries in2006, up 2.3% in 2003 but still below the peak of 3.1% in 1999.For developing countries belonging to the MENA region, foreign direct investment (FDI) continued to influence the economic growth in recent years. In this region, FDI continued to flow at high levels, about $ 45 billion, slightly down from a record $ 52 billion in The impact of financing development ressources on the promotion of economic growth : review of the empirical litterature A large empirical literature has focused on the study of the effect, that can lead the external financing resources of development to the promotion of economic growth. Thus, 1235

9 each source has a specific impact on promoting growth. However; according to the previous studies ; this impact is tempered by economic policies that determine its effectiveness The impact of the remittances on economic growth Workers Remittances are considered by some authors of the migration s word Bank rport 36 as a significant source of external finance for many developing countries. Indeed, given the growing trend in the amounts of money sent by migrants to their origin country some economists are interested in studying the effect of these funds to stimulate economic growth in recipient countries. The development economists from the World Bank including Dilip Ratha and Sanket Mohapatra think that international migration can generate welfare gains significant for migrants, both for origin and destination countries, to reduce poverty. These remittances are an important external resource for financing development. Their amounts exceed official development assistance 37.The importance of remittances from emigrants for developing countries emerges more clearly in proportion to Gross Domestic Product (GDP) of the recipient countries. Indeed, according to the Word Bank s data38, emigrants remittances can reach between 10% and 30% of GDP in some recipient countries. It is the same for most developing countries belonging to the MENA region. The table below shows the remittances as percentage of GDP to some developing countries in this region. Table No 3 Remittances % GDP Algeria 2.2% Djibouti 3.7% Egypt. Arab Rep. 5.0% Jordan 20.4% Lebanon 22.9% Morocco 8.3% 36 World Bank, 2006, Global Economic Prospects: Economic Implications and Migration, (Washington, DC). 37 Dilip Ratha and Sanket Mohapatra,(2007), Development Prospects Group, The World Bank Washington D.C November 26, Op.cit n 3 p

10 Syrian Arab Republic 2.4% Tunisia 5.0% Gaza 14.7% Yemen. Rep. 6.7% Based on previous studies on economic growth, few studies have tested the role of remittances on economic growth and reducing poverty. Among these studies were notably those of Lucas and Stark (1985) Adams (1991), Sander (2004) Gubert and Azam, 2005 and Adam (2006). In fact, it is noted that the macroeconomic effects have been accounted for at least two reasons: A theory suggests that these transfers are mainly used for consumption and therefore they have a little impact on investment. Therefore, remittances are considered as transfers of compensation for family members who have lost skilled workers due to migration. However, Stahl and Arnold (1986) 39 argue that the use of funds for consumption can have a positive effect on growth because of their possible multiplier effect. In addition, these transfers respond to investment opportunities in the country of origin. These possibilities depend on charitable motives or insurance. Moreover, many migrants invest their savings in small businesses, real estate or other assets in their own country because they know local markets better than those of their host countries.adams and Page (2005), Insights (2006), Siddiqui and Kemal (2006) and Gupta, Pattillo and Wagh (2009) believe that remittances have positive effects on the welfare and growth. And they also directly alleviate the poverty levels, improving the living standards of families receiving these funds. Similarly, remittances have a direct and adverse macroeconomic indirect. Faini (2002) and Ang (2007) found that the impact of remittances on economic growth is positive. Indeed Faini argues that transfers overcome market imperfections and capital enabling migrant households to accumulate positive assets. Ang has referred the identification of the relationship between workers remittances and growth at national and regional level and found that remittances affect positively and significantly to economic growth. Ang and al (2007) concluded that workers' remittances should be introduced to the added value of economic activities and investment, which are considered the main sources fundamental to the 39 Stahl, Charles W. and Fred Arnold, 1986), Overseas Workers Remittances in Asian Development, International Migration Review, 20 (4):

11 development and growth. By cons, it should be noted that critics of emigrants remittances reckon that impact on Economic growth is not significant and even negative. This is the case of Barajas, A., Chami, R., Fullenkamp, C. Gapen, M. and Montiel (2009), who have recently found that the effects on growth are insignificant or even negative. That s opposed what is expected by some economists of development thinking in particular that remittances are similarto foreign direct investment and capital inflows and ODA Official development in their effects on growth. Thus, much of the current empirical literature on remittances from emigrants was largely concentrated.on their effects on macroeconomic development and more specifically on economic growth, although these studies have led to controversial conclusions. This is the case Chami R., Fullenkamp, C. and Jahjah, S. (2003) 40 who have shown through their study covering 113 countries that remittances have had a negative 41 effect on economic growth which is caused by asymmetric information and uncertainty of remittances that limit their ability to help the promotion of the investment in human capital in developing economies. According to the IMF s study (2005) 42 on the effect of remittances on economic growth for 101 developing countries over a long period ( ), there is no link between remittances and growth in GDP per capita or between transfers funds and other variables such as enrollment rates or investment. This result weakly finding stems from the fact that remittances may behave cyclical manner with respect to growth.it follows from the various studies cited above, that consensus on the effectiveness of remittances is far from being established. The implication of the findings of these studies is that other policies can be used to increase the rate of growth in the long term, instead of depending solely on increased remittances The impact of ODA on economic growth Given the importance of development assistance for both donors and for beneficiaries, the literature on this concept is voluminous. In practice, the method for evaluating the ODA s economic effectiveness is to meet the question: "Has the flow of ODA have been broadly associated with improved economic performance of recipient countries of these flows and more specifically, is that are they associated with economic growth?» 40 Chami, R., Fullenkamp, C. & Jahjah, S. (2003), Are Immigrant Remittances Flows a Source of Capital for Development?, IMF Working Paper, No In fact, they concluded that the income of emigrants 'remittances allow countries to reduce receptionists their own work for developing countries. 42 IMF International Money Fund, (2005), World Economic Outlook: Globalization and External Imbalances, Washington, DC. 1238

12 The utility of ODA in promoting economic growth has been a subject of controversy since Rosenstein-Rodan (1943). Subsequently, during the last two decades, "the usefulness of ODA" has been the topic of several empirical studies based on panel data of developing countries receiving ODA, and the result from these studies lead to a consensus that has formed arguing that "ODA is useful" (ODA works) However, the usefulness of ODA to boost growth in developing countries has been a subject of several controversies. First, based on the work of Chenery and Strout (1966), the criteria of ODA allocation is based on the level of need otherwise the marginal effectiveness of aid depends on the country's deficit (two-gap model), connected to two constraints: lack of savings and foreign trade deficit. Then, Hansen and Trap (2000) have done their work in an overview of empirical analyses 43 developed over the past 30 years, with some regressions on a group of countries to assess the effectiveness of foreign ODA. And from 131 regressions, the authors lead to the following results: (i)oda increases aggregate savings. (ii) ODA increases investment. (iii)oda has a positive effect on the rate of economic growth, when pulled by capital accumulation. Nevertheless; the Boone s ( ) study 44 ; which focused on these relationships has shown that ODA has no impact on investment or growth in the neoclassical model. This finding was drawn from their regressions on a sample of more than 90 countries covering a period of 20 years. And following the legitimacy crisis of the ODA ; that has occurred during the 1990s; a new generation of studies on the ODA s effectiveness seems to focus on the relationship between aid, the economic policies and economic growth, while trying to identify factors whom generate the effectiveness of ODA. Thus, those studies 45 succeeding Boone were based on relationships rather standard growth; they added support as a fraction of GDP to analyze its effectiveness. Unlike Boone (1994), these studies found a positive impact of aid on growth. This is the case of Burnside and Dollar s study ( ) 46, which was developed while the ODA is down sharply. This analysis was a central to the debate on the effectiveness of ODA 24. The contribution of their work marks a real turning point in studies on ODA since these authors address firstly, the question of favorable 43 Henrik Hansen and Finn Tarp, (2000), Aid and Growth Regressions, CREDIT Research Paper No. 00/7. 44 Boone, P., (1996), Politics and the Effectiveness of Foreign Aid, European Economic Review42: Including Hadjimichael (1995), and Durbarry (1998), Lensink White (1999), Burnside and Dollar (2000). 46 Burnside, C., and D. Dollar, (2004), Aid, Policies and Growth: Revisiting Evidence, Policy Research Working Paper World Bank. Washington, D.C. Cassen, R., and Associates Does Aid Work? 2 nd ed. Oxford: Clarendon Press. 1239

13 macroeconomic conditionalties to greater efficiency help; secondly; they defend and support recommendations expressed in the report "Assessing Aid"published by the World Bank. Burnside and Dollar think that the ODA s effectiveness in terms of growth depends on the quality of economic policies implemented by developing countries. Their conclusion is based on an econometric work in which they estimate the coefficients of growth equations including an ODA s variable in interaction with indicator of economic policy. They hypothesize that the impact of foreign ODA on economic growth of developing countries is conditioned by economic policies that influence growth. They found that the interaction between ODA and a policy index 47 (Policy Index) is statically positive.this index policy is obtained through their regressions. (See Annex 1 and Annex 2) Similarly, the impact of ODA remains limited both in terms of growth in terms of reducing poverty. Among the economists; who have criticized this conditionality: Chauvet and Guillaumont (2001) and Collier and Dehn (2001); defend the thesis that the developing countries are structurally vulnerable so that other factors may influence the ODA s effectiveness terms of economic growth. These additional factors may be shortcomings in the ability to absorb ODA (Hansen and Trap ); the volatility of ODA flows; the structural vulnerability of the developing countries to external shocks (Guillaumont and Chauvet, 2001), the state of country's poverty (Collier and Dollar 2001), the level of borrowing of the twin deficits (Chenery and Strout,1966) and the existence of a post-conflict situation (Collier-Hoeffler 2004). The factors listed above illustrate the rich debate on the determinants of ODA s efficiency in promoting economic growth and selectivity conditions applied in the allocation of aid Therefore, several empirical studies are born each seeking empirically the impact of a factor influencing the effectiveness of ODA. Among the most relevant works, those of Chauvet and Guillaumont (2001), who think that vulnerability is a factor that may influence the ODA s effectiveness. Thus, they have argued that this effectiveness depends on environmental factors. They tried to show that the ODA will help to mitigate the negative influence of vulnerability due to the External shocks in many developing countries. They tested the following hypothesis: the worst environment, have the most need for assistance, and the marginal effectiveness of ODA will be the highest. 47 The index implies that the marginal effect of ODA on growth depends on the quality of policies economic. 1240

14 In addition, they believe that the work of Burnside and Dollar limits 48 present the following: The ODA does not differ from other financial flows. The ODA s conditionality creates a problem of endogeneity between aid and political support to take into account. The selectivity s argument is ambiguous: the ODA isn t effective if the policy isn t good.this argument will only be moral and assistance will be the reward of good policies. The ODA is thus presented as being an incentive to adopt good policies where the argument of selection opens to debate. Their new assumption is: a country that faces shocks (durable or transient), does eed assistance, insurance, to avoid the interruption of its growth process. The ODA will be more effective in these countries, since it allows absorbing the negative effects induced by shock contributes to the defense of growth. And in this context they suggest the following findings: -ODA is a support of good policies (planned as a reward / incentive) -ODA is a tool to cope with the shock / vulnerability (as planned insurance scheme) To justify what these findings,p. Guillaumont and L. Chauvet introduced a regression in their vulnerability s vector 49. The factors listed above illustrate the rich debate on the determinants of the ODA s effectiveness in promoting economic growth and selectivity s conditions in the allocation of this aid Foreign direct investment and economic growth The growth of foreign direct investment flows to developing countries and especially those from the MENA region during the last decade, growing to wonder on the potential role of these investment flows in stimulating economic growth. In recent economic literature, much attention has been devoted to the impact of FDI on economic growth of the host country of investment. Theoretically; in the neoclassical economic growth 50 ; increased volumes of FDI efficiency in 48 Lisa Chauvet and Patrick Guillaumont (2003), AID and growth revisited : Policy,Economic vulnerability and Political Instability, CERDI, CNRS, Université d'auvergne E The vulnerability has three dimensions: the magnitude of potential shocks that countries face and their exposure Shock-their ability to react to these shocks 50 According to Borensztein E., Gregorio, J., & Lee, J. (1998),''From How does foreign direct investment affect Economic Growth.''Journal of International Economics, 45 (1) p

15 promoting growth is done by generating distribution technology in developing countries receptionist investment. According to Balasubramanyam and Sapsford (1996) and De Mello(1999),"FDI is a composite b undle of capital stock, know-how and technology, and may consequently increase the existing stock of knowledge economy in host countries through training, skills development, disseminating the introduction of alternative, management practices and organizational arrangement. Blomstrom M., Lipsey, R., & Zejan, M. (1996) find that the evolution of FDI has empirically a positive effect on the growth of receiving countries, because using the FDI inflows into a country allow developing countries to make the exchange with other countries. UNCTAD (1999) finds that FDI can have either a positive or a negative effect on growth. It depends largely on variables that are introduced into the equation to test the impact of these investments.these variables can include the initial GDP per capita, level of education, the ratio domestic investment, political instability, terms of trade, the black market and the state of financial development. Borensztein E., Gregorio, J., & Lee, J. (1998), suggest that the effects of FDI on growth vary in the absorptive capacity of the country. Their analysis is based on the fact that the level of human capital determines the capacity to adopt foreign technologies. Thus, large endowments of human capital are assumed to induce high levels of growth economic, given the amount of FDI. This hypothesis is tested by their empirical findings. They also suggest that countries must have a minimum stock of human capital in order benefit from the positive effects of FDI. Durham (2004) fails to identify a positive effect generated by FDI on growth, but rather suggests that this effect is subject to the "technological absorption capacity of home countries, as alleged by Borensztein, E., Gregorio, J., & Lee, J. (1998).However, there is anecdotal evidence to suggest that the beneficial effects of FDI are notnecessarily related to the absorption capacity. For example, Bende-Nabende, A., Ford, J., Santoso, B., & Sen, S. (2003) find that the direct long-term impact of the FDI on growth is positive and significant for relatively less developed, such as the Philippines and Thailand. By cons, this effect is negative in more economically advanced countriesas Japan and Taiwan. However, the impact of FDI on growth remains controversial in empirical an study that s in theoretical studies. Although some studies find a positive impact of FDI on economic growth, others detect a negative relationship between these two variables. 1242

16 Empirical studies presented above show that the effect of FDI on economic growth is inconclusive. The role of FDI can be positive, negligible or even negative depending on economic institutions, and technological conditions of the receiving country. This shows that the results of studies that have examined the separate analysis of the relationship between each external sources of funding and economic growth are limited and can be improved by studying the effect of mixing these three sources. That's what we intend to test later in this article. 3. Empirical analysis of the effectiveness of external financing sources on economic growth in developing countries Our study aimed at identifying the contribution and effectiveness of the three main sources of funding external development in promoting economic growth. The three sources of external funding are preferred by financial institutions international, namely: the official development assistance, remittances and foreign direct investment. It should be noted that several empirical studies have attempted to measure the impact of each of these external financing sources on economic growth in developing countries. However, these studies have been criticized because of limitations in testing the impact of a single source to promote growth and economic development of developing countries receiving this source while ignoring the role that could the other source have. Indeed, despite the diversity of studies on the positive impact played by the sources in developing countries, the effect of the integration of remittances has not been adequately tested. Given this criticism, and in order to monitor the impact of these various sources on the promotion of economic growth of countries in our sample. We also consider the traditional resources of economic growth. We use fixed effects and random models methods to estimate the impact of the external capital flows and the traditional resources. This method will allow us to identify their contributions to economic growth. Thus, in our study, we will estimate the contribution of the three external financial resources(the official development assistance (ODA foreign direct investment (FDI) and remittances (wr)).and the three traditional sources The gross fixed capital formation as a percentage of GDP (GCF), the terms of trade index(tot) and the enrollment rate in secondary (SC). The promotion of economic growth. To do this, we will first introduce the external sources of financing for development. Subsequently, we introduce the traditional explanatory variables of economic growth one after the other. And finally, we will build a new model that multiplies the external financing sources and the traditional sources of economic growth. 1243

17 3.1.Sample and variables Our estimate will be conducted on a sample composed of 13 developing countries belonging to MENA51 region and beneficiaries of official development assistance from the Committee's list of aid Development OECD / DAC. The three extern capital flows are 52 : Remittances (wr): Are defined as current private transfers from migrant workers living in the host country for more than a year to recipients in their country of origin. Foreign direct investment (FDI): Are defined as capital flows in which an enterprise of one country creates or develops a subsidiary in another country. Official development assistance (ODA): Represents grants or loans on preferential financial terms to developing countries by bodies of the Organization for Economic Cooperation and Development (OECD) gathered in the Development Assistance Committee (DAC). The other traditional sources of economic growth are: The gross fixed capital formation (Gross fixed capital formation) (gcf):gcf includes land and property acquisitions, purchases of plant, machinery and equipment, and construction of buildings, infrastructure and equipment. The index of terms of trade in goods net (2000 = 100) (tot) is calculated as the ratio of the indices of export unit values relative to unit value indices for imports, measured relative to the reference year. The enrollment rate in secondary (sc): Is the ratio of total enrollment, regardless of age, compared to the population of age group that officially corresponds to a level of education. Secondary education completes the provision of basic education that began at the primary level. To determine the reactivity of the annual growth rate of GDP compared to income of remittances, flows of foreign direct investment, official development assistance to development and three other conventional sources of economic growth defined above, we use panel data on the main exogenous factors including: Lwr i, t : Denotes the log of remittances from migrant workers as a percentage of GDP in the country i at the time t. 51 We will be limited to data on only 13 developing countries in the region of MENA; there is no considered Gaza and Iraq, although they benefit from high flows in remittances, lack of availability data. 52 Three basic factors in our estimation. 1244

18 Recent literature states that there is a positive impact of remittances on economic growth. This study will explore the overall impact on economic growth in MENA region. Lfdi i, t : Denotes the log of the net inflows of foreign direct investment as percentage of GDP of the country i at the time t. This impact has been fully studied, and among economists who argue for an effect called positive and significant economic growth were particularly Blomstrom M., Lipsey, R., & Zejan, M. (1996), Hansen and Rand (2006); de Mello, (1999). LODA i,t : denotes the log of the amount of aid received as a fraction of GDP for the period i t (Offical net aid per capita Received:% of GDP). It should be noted that supporters of aid (including Chenery and Strout, 1966;Papenek, 1973;Levy, 1987; and Islam, 1992; Fayissa and El-Kaissy, 1999) argue that aid flows are necessary for economic growth in developing countries. However, the opponents of ODA (Heller 1975 and Boone, 1994) stipulate that these external capital flows can generate a negative effect on domestic savings and economic growth in developing countries. And in this context several empirical studies have emerged focusing on factors that may influence the effectiveness of ODA. Among these studies, the work of Burnside and Dollar ( ) those are central to the debate on factors causing the effectiveness of aid. The idea developed by them and defended by the World Bank report Assessing Aid (1998) is that the effectiveness of aid on growth depends on the quality of the institutional and political environment. The log of ODA, the log of WR and the log of FDI are used to capture the effect of these three sources of funding development on economic growth. Lgcf: is the log of gross fixed capital as a percentage of real GDP used as a proxy for capital investment physique 53. Ltot: is the log of the term of trade for each country in the sample, measured by the ratio of export to import price index to capture the impact of trade or economic openness on economic growth. Lsc: is the log of enrollment in secondary education as a measure of investment in human capital that has a positive effect on economic growth (Schultz, 1980; Romer, 1986; Lucas, 1988, and Barro, 1990). These three log are used to capture the effect of three traditional sources on economic growth 53 Our estimate is based on empirical studies in the context of the new theory of growth (Lucas, 1988; Barro, 1990, Benhabib and Spiegel, 1994, Grossman and Helpman, 1991, Barro and Sala-i-Martin, 1992b; Barro and Lee, 1994, and Temple, 1999). 1245

19 g i, t :is our endogenous variable is represented by the annual growth rate GDP of the country i at time t (GDP annual growth). 4. Specification and estimation issues Our model is based on panel data. The panel data are used to monitor an individual effect i (i = 1 to N) over a period T (t = 1 to T). Data are doubly indexed. We note as well, yit = xb it + µ it, where αi have represented the effect of specific countries and εit is an error term. Our model is based on panel data witch has the advantage of addressing the joint common and specific effects. However, models based on panel data raise the problem of the correct specification of the effect that is either common or specific. This first necessary to distinguish the specific effects and the common effect. To perform the regressions, two possibilities must be taken into consideration: fixed effects estimation (MEF) or random effects (MEA). The fixed effects estimates are calculated based on differences within each country and through time. Indeed, the individual effects specific to each country are represented by deterministic constants. We do not care about the difference in average countries against each other. But we are interested in the variation of the estimator coming from the name "Within" estimates of random effects are the most efficient, since they incorporate individual information across countries and across periods. Thus, we use only the interindividual s variability which is based on the calculation of average individual. She locates the position of individual compared to the individual average position. We will adopt the model wich considers economic growth as measured by the annual growth rate of GDP according to two categories of variables: Sources of financing for development including: direct investment inflows (FDI), remittances from emigrants (wr) and official development assistance (ODA) The traditional sources of economic growth that is the gross fixed capital percentage of GDP (GCF), the index in terms of trade (TOT) and the enrollment rate in secondary (SC). Our model is then written as follows: (g)it=αi+α1log(fdi)it+α2log(wr)it+α3log(oda)it+α4log(grcf)it+α5log(sc)it+ α5log(tot)it+ εit 5. Estimation Our model is based on panel data witch has the advantage of addressing the joint common and specific effects. However, models based on panel data raise the problem of the correct specification of the effect that is either common or specific. It is first necessary to distinguish the specific effects and the common effect. 1246

20 5.1 Existence test of specific effects This is a test of homogeneity constants α i assuming b i =b (the coefficients explanatory are assumed constant). Under the null hypothesis, the sample is homogeneous and there only a common constant and no individual effect. Thus, H : α = α =... = α =... = α i N It is remarkable for the regressions: (1), (2), (3) and (5) of the table below, that Fisher's test allow us to accept the null hypothesis that the panel is homogeneous since the p-value is greater than 5%. It therefore asserts that there was no specific effect of the countries. Thus it affects OLS estimation. However, for the regression in column (4), it should be noted that Fisher's test allowed us to reject the null hypothesis that the panel is homogeneous since the p-value is less than 5%. It thus affirms the existence of individual effects. In this case, the question is how these effects should be specified?this amounts to answering the following question: Should we accept the hypothesis of randomness or otherwise the hypothesis of fixed effect? It examines the model and sees its nature, that is to say, if the fixed effects (Estimator within) or random (between estimators).the choice between these two effects is done through the Hausman test (1978). According to our estimation Annex N (6) the statistics of the Hausman test equals 9.63 with a probability of less than 5%. So we retain the fixed effect model to estimate. 5.2 Empirical results Exogenous Variable Constant *** (9.11) Log (fdi) *** (2.47) Log (wr) *** Log (oda) (1) (2) (3) (4) (5) (4.29) (0.24) (-0.42) *** (2.33) *** (0.101) (0.74) (6.23) *** (2.70) *** (3.35) (-0.21) (-0.20) *** (2.09) ** (2.06) (0.28) (-0.18) ** (2.46) ** (2.23) (0.09) Log (tot) ***

21 (2.86) (1.31) Log (gcf) *** (5.89) *** (5.25) Log (sc) *** (1.77) (0.44) Obs N R 2 Within R 2 Between R 2 Overall Note: the endogenous variable is the annual rate of GDP growth *, **, *** Indicate statistical significance respectively at 10%, 5% and 1%. The main objective of this study is the investigation of the remittances mixed with two other external funding sources and three conventional sources of economic growth. The results of our estimation show that shipments of immigrants have a positive impact on economic growth in developing countries. For the entire sample and referring to all the columns in the table above, the elasticity the rate of annual growth of GDP relative to foreign direct investment (FDI) is positively significant at 1%. In addition, the results of our model show that remittances from migrants have a positive impact and statistically significant at 1% level. Thus an increase in remittances (in percentage of GDP) by 1% leads to an increase of 29.63% annual growth rate GDP of developing countries in the MENA region. If FDI flows to developing countries of the MENA region increases by 1%, this will cause changes in the rate of GDP growth of 0.67%. This is consistent with results obtained by; Blomstrom M.,Lipsey, R., & Zejan, M. (1996); Hansen and Rand (2006); de Mello, (1999); who found that the evolution of FDI have an empirical positive effect on the growth of receiving countries, because of the use the FDI inflows in developing countries to allow this country to exchange with other countries. 1248

22 In addition, our results indicate that remittances have a positive and statistically significant impact at the 1% level. Thus an increase in remittances percentage of GDP by 1% leads to an increase of 29.63% of the annual growth rate of GDP in developing countries in the MENA region. This result is predicted by Faini (2002) and Ang (2007) who found a positive impact done by the remittances.this is the same view of Stahl and Arnold (1986)54; who argue that despite the factthat transfers funds are intended for consumption, they can have a positive effect on growth economic multiplier effects because of their potential. In addition, remittances appear as a response to investment opportunities in countries of origin.similarly, Adams and Page (2005), Insights (2006), Siddiqui and Kemal (2006) and Gupta, Pattillo, and Wagh (2009) and the development economists of the World Bank Dilip Ratha and Sanket including Mohapatra 55 argue that remittances s effects on the welfare and economic growth are positive. If we compare the two coefficients assigned to the variable (FDI) and the variable (WR), we note that there is a contribution of flux transfer to the annual GDP growth well above that of FDI, despite the fact that FDI flows are much larger compared to the remittances. For the elasticity of the annual growth rate of GDP compared with official Development Assistance ODA, the results reported in columns (1) (2) (4) (5) tells us that variable (ODA) has a positive but not significant annual growth in GDP. This is predicted by Boone ( ) who studied the relationship between ODA and economic growth and demonstrated that the ODA does not impact on investment or growth in the context of neoclassical models and that this impact may even be negative (column (4)) on economic growth.heller (1975) and Fayissa and El-Kaissy (1999) support the findings of Boone (1994) and state that external capital flows generate a negative effect on domestic savings and economic growth in developing countries. Indeed, it is noted that developing countries are structurally vulnerable to so many factors whose may influence the effectiveness of ODA. And therefore, these factors determine the effectiveness of this assistance granted to developing countries. Among these factors there are: the quality of the institutional and policy study by Burnside and Dollar (2000) who hypothesize that the impact of foreign aid on economic growth in developing countries is conditioned by economic policies that influence economic growth. They found that the interaction between aid and an index of policy (Policy Index) is statistically positive. The quality of macroeconomic policies is actually apprehended by the control of inflation, balanced budget and the implementation of a policy of trade openness. 54 Op cit n 19 p.8 55 Op cit n 16 p.8 56 The index implies that the marginal effect of aid on growth depends on the quality of policies economic. This indicator of economic policy is as follows: Policy = x Budget surplus x Inflation x Openness

23 The conclusion that the effectiveness of aid depends on the quality of economic policies then results from the demonstration, based on estimates of growth, a significantly positive indicator of economic policy on economic growth. In addition, there are other factors influencing this effectiveness as: The level of poverty (Collier and Dollar 2001); Economic vulnerability ( Chauvet and Guillaumont, (2001) ; The absorption capacity of ODA ((Hansen and Tarp 2000, 2001). It is therefore important to study, the impact of the economic policy index as well as the factors stated below, on the effectiveness of ODA to promote economic growth. By introducing the variable (tot); which is a measure of openness of the economy to capture the impact of trade or economic openness on economic growth; we obtain firstly coefficient with it a positive sign and statistically significant at 1% (column (2)). This confirms what was found by Sachs and Warner (1995). Thus, an increase of Variable (tot) of 1% will increase 39.67% of the annual growth rate of GDP in the developing countries of the MENA region. By introducing the variable log of the rate of gross fixed capital formation as a ercentage of GDP (gcf) (column (3) and (5)), we notice that the impact of this variable is positive and statistically significant at the 1% level. An increase of GDP of 1% will increase annual growth rate of GDP by 53%. Thus, in accordance with the findings of Solow (1956), Barro (1990) and Temple (1990), we find that investment in physical capital, (GCF), measured as gross fixed capital as a percentage of GDP has a significant positive effect on economic growth. By introducing the variable secondary school enrollment rate (column (4)), we notice that the impact of this variable is positive and significant at the level of 10%. Indeed, an increase of 1% of capital investment human lead to an increase of 38.25% annual growth rate of GDP. And finally, making a mix between the traditional variables of economic growth and external capital flows variables; the results of estimating the overall model (column (5)) confirm the results found previously. Concerning the effect of the aid variable, its impact is positive but not significant this is explained by the presence of several factors affecting its effectiveness in promoting economic growth. 1250