Workers Remittances and Economic Development in Sub-Saharan African Countries

International Research Journal of Finance and Economics ISSN 1450-2887 Issue 88 (2012) EuroJournals Publishing, Inc. 2012 http://www.internationalresearchjournaloffinanceandeconomics.com Workers Remittances and Economic Development in Sub-Saharan African Countries Tchamanbé Djiné Louise University of Yaoundé II, Soa, Cameroon E-mail: ltchamanbe@yahoo.fr Tel: (237) 97 18 07 03 Miamo Wendji Clovis University of Dschang, Cameroon E-mail: c_miamo_w@yahoo.fr Tel: (237) 99 32 97 06 Abstract The purpose of this paper is to analyze the relationship Workers Remittances (WRs), poverty, and economic growth in Sub-Saharan African (SSA) countries. According to Banque Mondiale (2010), this external source of finance seems to affect the economic development of recipient countries in the same way as official development aid (ODA) and foreign direct investment (FDI). To achieve this objective, we use the fixed effects panel regression models proposed by Dollar and Kraay (2002), Berg and Krueger (2003), and Jongwanich (2007), as well as a sample of SSA countries to determine the link between WRs and poverty on the one hand, WRs and growth on the other. Considering the specificities of each of the models retained in the present study, recourse to the instrumental variables method turned out to be better for the estimation of the poverty model, while the standard error correction method proved to be well suited for the estimation of the growth model. The empirical results of the study enable us to confirm the ambiguity observed in the literature about the capacity of WRs to give impetus to economic development. Beyond the fact that the coefficient associated with the WRs variable in the poverty model is not significant, the study results indicate that WRs do not contribute significantly to the reduction of poverty. However, it is important to note that WRs seem to be a key element in boosting economic growth in the same manner as investment. Thus, if decision makers allocate a substantial percentage of the fruits derived from this growth to investment in the development of human capital, physical and basic social infrastructures, it is very likely that poverty will be reduced in the long term. Keywords: Workers Remittances (WRs), Poverty, Economic Growth, Development. JEL Classification Codes: D63, F22, F24, F35, I31, I32 1. Introduction Up to the end of the 1970s, Official Development Aid (ODA) was the main source of external financing for a large number of developing countries. Then in the 1980s and the 1990s, the globalization phenomenon witnessed the development of Foreign Direct Investment (FDI) which has proved to be an important source of development financing by contributing significantly to gross fixed

International Research Journal of Finance and Economics - Issue 88 (2012) 25 capital formation (GFCF) in many poor countries 1. At the same time as FDI, the importance of migratory flows was accompanied with the concomitant development of Workers Remittances (WRs) or cash remittances from abroad in the direction of their countries of origin. Moreover, these WRs have in recent years witnessed renewed interest within development institutions such as the World Bank, the IMF, the OECD and the UN, since WRs have been found to contribute significantly to poverty reduction and economic development in some recipient countries (Banque Mondiale, 2010). A comparative analysis of these three sources of development finance arrives at the conclusion of a continual decrease in ODA since the beginning of the 1990s 2, concurrently with a rise in FDI as well in WRs from that period onwards (Banque mondiale, 2005). Owing to their relative stability, these cash remittances are also considered as a source of financing which helps attenuate the effects of the crisis 3. WRs constitute an important mechanism for redistributive policies at the global level, and for the promotion of growth in poor countries. They have established themselves as a major instrument which poor countries should rely on to implement their development policies. The efficient and dynamic way to carry such policies through is to make sure that WRs participate effectively in poverty reduction and the promotion of economic growth in Third world countries 4. The increase in the volume of WRs throughout the world accompanies the fact that this resource is no longer essentially allocated to household consumption. It increasingly participates in financing investment which is the engine of economic growth. As the poorest continent of the planet, Africa is not always the first beneficiary of WRs. Even if some of the countries of this continent such as Lesotho achieve acceptable performances as far as WRs are concerned, other countries such as the Cape Verde islands lag behind the world s average 5. These countries which lag behind have difficulties in gathering additional resources to finance their development. Questions then arise as to the capacity of WRs to reduce poverty or to give some impetus to, and sustain growth in African countries, particularly in countries located south of the Sahara, among which appear some countries ranked as the poorest on the planet 6. In other words, do WRs, which from the start were intended to rush to poor populations aid in recipient countries, effectively fulfil this mission? With ever increasing transfers, can these funds have a positive impact on growth in the poorest African countries? The answers to these questions necessarily require an empirical analysis on the capacity of WRs to reduce poverty and stimulate economic growth in Sub-Saharan Africa. The object of the study then boils down to analyzing the effects of WRs on the development of Sub-Saharan African (SSA) countries through their impact on the general improvement in the living standards of populations and an increase in the wealth of these countries. To achieve this, we will present some statistics on the evolution of the three main external sources of development finance in these countries, namely ODA, FDI, and WRs. Moreover, we will present a review of the literature on the impact of WRs on poverty and economic growth. A description of the econometric models used to capture these impacts will 1 The World Bank Report «Global Development and Finance (2003)» underlines the fact that FDI was of considerable importance for small poor countries such as Lesotho, Mauritania, Moldavia, and Mozambique. The same report adds that from an annual average amount of US$ 0.5 billion in the 1980s, FDI increased to US$ 3.5 billion in 1990 and US $ 8.4 billion in 2004. 2 Report Global Development and Finance (2003). 3 Ratha (2008), notes that cash remittances account for a large share in the GDPs of countries such as Tadjikistan (45 %), Moldavia (38 %), Tonga (35 %), Lesotho (29 %), and Honduras (25 %). 4 Since it is recognized that a rise in per capita income in poor countries make it possible to improve overall welfare through consumption, access to education and healthcare, life expectancy, the quality of life, etc. 5 The ranking of the regions of the world (East Asia and the Pacific, Europe and central Asia, Latin America and the Caribbean islands, the Middle East and North Africa, South Asia, and Sub-Saharan Africa) show that Sub-Saharan Africa ranks last behind East Asia and the Pacific, South Asia, and Latin America, which are the champions of WRs. Since these remittances of cash are a contributory factor in the reduction of poverty even in countries where socio-political crises are taking place, it would be interesting for African countries to extend the impact analysis of this financing source to other considerations. 6 We may mention, among others: Niger, Mali, Chad, and Mozambique (Banque Mondiale, 2005).

26 International Research Journal of Finance and Economics - Issue 88 (2012) enable us to describe the panel of countries under study, and to present and make comments on the obervable trends in this part of the continent. 2. Highlighting the Diversity of Sources of Development Financing The African Development Bank (ADB) gives high priority to the promotion of the cash remittances made by migrant workers, on the one hand, and to their impact on development in Africa on the other. In effect, among the categories of funds transferred in the direction of Africa (ODA, FDI, WRs), various studies indicate that WRs are taking on increasing importance in the economies of countries with strong migration. However, in spite of their weight, the financial flows generated in the direction of African countries are not well known. What is more, many shady areas remain concerning: (i) the real use made of all these resources, (ii) their capacity to improve living conditions in recipient countries, and (iii) their capacity to achieve the structural transformation of the economies of these countries in the long term. From the comparative analysis of the main sources of development finance in the specific case of Africa, has emerged a controversy about ODA and FDI which has given rise to renewed interest in the endogenous sources of development financing, as in the case of the debate on WRs. Both the above conttroversy and the comparison between ODA, FDI, and WRs are presented below. 2.1. Controversy about ODA and FDI in Africa The primary objective of ODA is to reduce multidimensional poverty. One of the ways to achieve this is to boost economic growth in developing countries. In effect, aid helps not only to finance investment in poor countries, but also to increase the capital stock, given that production capacity is proportional to the capital stock (Domar, 1946). For this reason, it participates in economic growth as an key development factor 7. Paradoxically, SSA countries which have benefited from international aid since their independence have yet to practically evolve from their initial situation after half a century of development aid. Many of the economies of these countries have even been in decline since the middle and late 1980s 8. The increase in poverty at the same time as development aid, as well as the indebtedness problems of some aid-recipient countries led development aid into an unprecedented legitimacy crisis in the mid-1990s. The debate on aid, its effectiveness, its conditions, and indebtedness has emerged as burning issue since that time, and SSA countries where development statistics are more pessimistic are placed at the center of this debate. With the advent of globalization and the acceleration of international migrations, FDI and WRs increased significantly. However, in the African environment where extreme poverty is rife and where there is a lack of human capital and infrastructures, investors are less motivated to make income transfers in the form of FDI, given the existence of high risks. And we observe next to this speculative behavior, a significant inflow of financial transfers by migrants to their countries of origin. The ever increasing volume of WRs has helped the latter to become a non negligible source of development finance. 7 Let s consider the example of South Korea. This country largely developed within a period of thirty years thanks to international financing. Between 1945 and 1961, South Korea did not receive any private foreign direct investment. It benefited from 3.1 billion US dollars of aid from the United States in the form of grants. With this aid, South Korea developed its road infrastructure, and its health and educational equipment which was desperately lacking. 8 In contrast to the stunning successes of international aid in Asia, we observe repeated failures in many SSA countries such as Zambia, the Democratic Republic of Congo (DRC, the Central African Republic (CAR), the Congo (Brazzaville), Mauritania, Togo, and Guinea-Bissau. In these countries, the hoped-for economic take-off driven by external financing has not come about, and international development aid has left but heavy indebtedness in its wake. Most African countries today are weighed down by enormous debts. For instance, in 2000, Senegal s debt amounted to about 80% of its GDP (Banque Mondiale, 2006).

International Research Journal of Finance and Economics - Issue 88 (2012) 27 2.2. Workers Remittances and other Sources of External Financing The development of migratory movements in recent years was accompanied by the ever increasing flows of cash remittances sent by migrants 9 for the benefit of their families in their countries of origin. Record volumes of WRs have even been achieved since the onset of the second millennium. These transfers increasingly position themselves as a non negligible external source of development financing 10. WRs are remittances of cash made by migrants to their countries of origin. The IMF distinguishes three components in WRs. Firstly, the transfers of immigrant workers corresponding to the monetary value of the cash remittances made by workers who have been residing abroad for more than a year. Secondly, the salaries earned by the employees of embassies and consular services. Lastly, the transfers of migrants corresponding to the values of flows of interests, profits, and dividends between countries, which occur when the assets held by the residents of one country in other countries, are transferred through international accounts 11. In other words, we are dealing here with the bank transfers of the gains of one country to another. In the context of the present study, the WRs retained in this study are the sum of immigrant workers transfers and the remunerations of employees such as they emerge from the World Bank s data which we use. The diversity of actors who make remittances of cash has given rise to such a surge in financial flows transferred by migrants to their countries of origin, that these remittances increasingly follow the lead of other sources of development financing. A recent report of Banque mondiale (2005) 12 reveals that in 2004, WRs inflows in poor countries reached 15.9 billion US dollars, or on the average 5.1% of GDP in 2002/2003 versus 2.8% in 1990/1991. A comparison between WRs and other sources of development finance shows a sharp fall in Official development aid (ODA) starting in the 1990s, an increase in FDI and in the WRs which rank second behind FDI. According official statistics, cash remittances by emigrants to all of Africa in 2007 amounted to 36.9 billion US dollars. This amount is expected to exceed 45 billion US dollars in 2012. Although the data on this subject is not available to us, and since most of the data on migrations and cash remittances are not always available for all countries, some research studies suggest that the cash remittances dispatched by emigrants contribute tremendously to poverty reduction in Africa. Some studies also reveal that these remittances increase the income of poor households, and indirectly boost economic growth (Nagarajan, 2009). Today, WRs inflows hold considerable importance as a source of external development finance. However, it remains nonetheless true that the interaction between migrations, the financial transfers of migrants, and development has become a subject of controversies between researchers and decision makers. In effect, for many years researchers have analyzed the determinants of migratory movements regardless of their effects and vice - versa. 3. Workers Remittances and Development in the Countries of Destination The literature review on the impact of WRs in countries of origin make it possible to distinguish an opposition between two trends: the first trend considers WRs as being an obstacle to economic 9 The United nations define a migrant as any person who has been living outside his or her country of birth for more than a year. According to the UN, the number of migrants in the world has increased by more than 20 millions of persons over ten years, and presently account for 3% of the world s population. In terms of composition, the migrant s socio-economic profile has also changed. From the single man and unskilled worker of the 1970s, it seems that the contemporary flows of migration are characterized by an increasing number of illegal immigrants of students, refugees seeking asylum, highly qualified migrants, and women. 10 Workers Remittances to have reached 100 billion of US dollars in 2003 (Banque mondiale, 2006). 11 According to the IMF, during the last 15 years, the transfers of immigrant workers have accounted for 2/3 of the world s total of WRs, versus 25% for the compensations of employees, and 10% for WRs. 12 Banque mondiale (2005).

28 International Research Journal of Finance and Economics - Issue 88 (2012) development in the migrants countries of origin, and the second trend considers this source of finance rather as a manna from heaven which contributes to the development of recipient countries through the reduction of poverty and the impetus it gives to economic growth. 3.1. Workers Remittances as an Obstacle to Economic Development The perception that WRs have no impact on the development of labour-exporting countries has been analyzed by Todaro (1969) as well as Harris and Todaro (1970). Moreover, some studies even suggest that WRs benefit the poor. In particular, Adams (1991) shows that cash remittances have increased income inequality in Egypt. One of the main reasons behind this fact is that wealthy families can afford the expenses linked to international migrations more than other households. International migration being costly, it usually concerns the wealthiest households that are able to send their members abroad. Poor households may therefore be likely to benefit from these inflows. This situation tends to generate inequalities and to widen the gap between the rich and the poor in the long term (Koechlin et Leon, 2007; Chauvet and Mesplé-Somps, 2007). In a recent study on Vietnam, Viet (2008) arrives at the conclusion that remittances of cash from abroad noticeably increase household income and consumption, but that these remittances only have a limited impact on poverty reduction. In the context of his study on Mexico, Fuente (2008) has also found a negative and significant relationship between foreign cash remittances an rural household poverty. As regards growth, the skepticism about the capacity of WRs to give a positive and significant impetus to an increase in wealth has also given rise to debates in some studies whose purposes were to link remittances of cash to economic growth. These studies notably indicate that cash remittances are generally spent on financing consumption, housing construction, debt repayments, and the financing of future migrations (Stark et Levhari, 1982; Ahlburg, 1991). According to these authors, cash remittances have made it possible to increase the consumption level without necessarily creating a strong basis within the national economy, since only a small proportion of households benefit from them. The lack of investment in productive activities consequently casts some doubt on the capacity of cash remittances to give impetus to economic growth 13, thus strengthening the point of view of authors who think of WRs as a poverty reduction instrument. These studies contrast however with those which consider WRs as an instrument likely to facilitate development in the migrant workers countries of origin. 3.2. Workers Remittances as an Economic Development Asset in Recipient Countries The first studies on the remittances of cash are primarily those focusing on their capacity to reduce poverty, for these cash inflows directly help to reduce poverty by increasing the incomes of the beneficiaries (Adams and Page, 2003). These increased incomes play a significant role in raising the level of consumption of the poor. Thus, regardless of its impact on growth, such an increase in consumption enables poor households to raise their living standards (Adams and Page, 2005; Gupta et al., 2009) 14. The ever increasing WRs inflows in the countries of destination help to observe the new orientations of funds toward activities which have a direct impact on growth. Brown (1994) examines 13 In addition, cash remittances may affect the supply of labor indirectly by encouraging certain households who benefit from them to work less. This may reduce the labor supply, and hence economic growth. 14 In the same vein, Adams and Page (2005) examine the impacts of transfers on poverty in 71 developing countries. Their study results show that international migrations and cash remittances can reduce the incidence, depth, and severity of poverty significantly in these countries. In fact, when the proportion of migrants increases by 10% on the average the numerical poverty index decreases by 1.6%. Moreover, a rise of 10% in WRs in a country s GDP leads to a fall of 1.2% in the proportion of the population living with less than one dollar per day, and to a reduction of 2% in the intensity or gravity of poverty in the countries considered.

International Research Journal of Finance and Economics - Issue 88 (2012) 29 the relationship between cash remittances, savings, and investment in the Tonga and Samoa islands using micro-analysis to determine how households spend their cash remittances. It emerges from this study that these financial flows significantly contribute to savings and investment in these island economies. In addition, Stark and Lucas (1988) show that migration is an integral part of household income diversification strategies which are increasingly integrated into the development policies of countries to which financial flows are dispatched. According to Russel (1992), WRs increase household income and sustain local demand. This effect is all the more profitable when demand applies to products manufactured in the country of origin or if this demand occurs in a weakly competitive market, for fear that excess demand comes with inflationary tensions, thus cancelling the demand effect. The Remittances inflows, which are less volatile than other external sources of financing (Chami et al., 2008), may have stabilizing effects if they are countercyclical, for instance (Spatafora, 2005; Fajnzylber et al., 2007; Bugamelli and Paterno, 2008; Chami et al., 2009). According to Ratha (2003, 2007), WRs attenuate shocks insofar as economic recessions encourage other workers to emigrate, while those who are already working abroad increase the WRs volume dispatched to their families in the countries of origin, and while speculative capital conversely flee from countries of origin. In addition, WRs do not need counterparts from public authorities, for they are not subjected to contagion, panics, bubbles, and political pressures (Kapur and Singer, 2006) 15. More recently, cash remittances have increasingly been helping households to set up small commercial activities or small production units (Yang, 2004; Woodruff and Zenteno, 2001). Since WRs are not subjected to the credit constraint, more particularly to that related to guarantees, an increase in these resources may facilitate business activity and help increase private investment. Beyond physical investment, cash remittances may also help to finance education and health which also play a key role in the promotion of long-term economic growth. For this reason, cash remittances can increase both physical investment and human capital, thus reducing the level of multidimensional poverty in the long term. Consequently, cash remittances may generate positive effects on economic growth through multiplier effect mechanisms. Since there are upstream and downstream relationships in investment activities, an increase in the investment of one household may generate a rise in the income of other households. In the context of increasing returns, for instance, the expansion of one sector may increase the size of other sectors. The relationship between FTMs, poverty, and growth in the African context will be determined with the use of econometric models and panel data. 4. Choice of Models and Description of the Countries under Study Beyond the controversy mentioned above, a relevant analysis of the relationship between WRs, poverty, and economic growth in SSA countries requires the use of regression models. To do this, panel models are retained in this study, considering the structure of the data. 4.1. Presentation of the Models The choice of models to be used in the determination of the impact of WRs on economic development through poverty and growth constitute a key element for ensuring the relevance of the study s results. By drawing our inspiration from recent studies by Dollar and Kraay (2002), Berg and Krueger (2003), as well as Jongwanich (2007), we have adopted the poverty model given by the following equation: Pov it = β o + β 1 Grow i,t + β 2 In it + β 3 WRs it + β 4 CH + β 5 C it + α i + ε it (1) Where, Pov: is the poverty measure; 15 However, WRs may also be pro cyclical when, instead of being motivated exclusively by altruism, they are motivated by the search for personal interests by migrants who want to diversify their portfolios. The risk for a country receiving pro cyclical WRs is linked to the fact that the latter may further exacerbate the economic cycle (Ebeke, 2010).

30 International Research Journal of Finance and Economics - Issue 88 (2012) Grow: is economic growth. Economic growth changes in both directions relative to poverty. When growth is pro poor, the sign of β 1 is positive or negative if this shouldn t be the case. A negative sign for the coefficient β 1 is expected. In: inequalities. Inequalities in the distribution of income tend to have a negative impact on poverty reduction. WRs: represents the Workers Remittances. As we have gathered from the literature, WRs may have a positive or negative impact on poverty. CH: human capital. An increase in human capital may increase the capacity of poor people to generate income; the coefficients associated with these variables should be positive. C 16 : is the control variable. The control variable used in this model depends on inflation (Inf) and the degree of openness (DO). α i is the unobserved specific effect of the country, and ε is the error term. The model used to analyze the relationship between WRs and economic growth is based on the studies of Barro (1996) and Jongwanich (2007). It is given by the following equation: Grow it = c 0 + c 1 PIB i,t-1 + c 2 I it + c 3 WRs it + c 4 Inf it + c 5 DO it + c 6 DP it + α i + ε it (2) Where, Grow is the average per capita GDP growth rate of country i at time t (Burnside and Dollar, 2000; Easterly et al., 2003; Clemens et al., 2004). It represents the dependent variable. PIB i,t-1 is the initial per capita income expressed as a logarithm. This variable makes it possible to take account of initial endowments, with the aim of capturing the effect of convergence between economies 17. This variable is transformed into a logarithm to minimize the effects of large income gaps on the GDP growth rate (Burnside and Dollar, 2000; Easterly, 2003). I is the investment rate of economy i at time t expressed as a percentage of GDP. Its impact is analyzed with a lag of one period. WRs represents the Workers Remittances. As mentioned in the preceding section, the (sign of the coefficient associated with cash remittances also remains ambiguous in this model. α i is the unobserved specific effect of the country, and ε the error term. Based on the specifications of Barro (1996), Barro and Sala-i-Martin (1995), as well as Giuliano and Ruiz-Arranz (2005), we introduce government consumption (CGouv), the degree of openness (DO), and inflation (INF) 18 into the model. Moreover, following Hall and Jones (1999), we asked ourselves the question as to whether in our models there exist significant causality relationships between the endogenous variables (i.e. poverty and growth) and their respective explanatory variables. The answer to this question requires a good grasp of the populations and characteristics of SSA countries. 16 A rise in inflation tends to have a negative impact on poverty reduction. The ccoefficient associated with poverty reduction through trade liberalization is ambiguous (Berg and Krueger, 2003). The poor may benefit from trade liberalization at least as much as the average individual in the population. Trade liberalization may increase inequality in the distribution of income, particularly among unskilled workers in a situation of increased foreign competition, thus making the poor poorer. In this case, trade liberalization can have a negative impact on poverty reduction in spite of its positive effects on overall growth. 17 The coefficient of initial GDP should be negative, a situation which represents a rate of conditional convergence. Under decreasing returns, we expect the poorest countries to grow more rapidly than rich ones. 18 As indicated in the literature, openness is desirable to promote economic growth. It helps to achieve an efficient allocation of resources, to stimulate the business activity resulting from competition and access to big markets, to reduce the private income of activities inspired by the restriction of trade. We therefore expect to obtain a positive relationship between trade openness and economic growth. On the other hand, we expect to have negative coefficients in the case of the government consumption and inflation variables. Government consumption is the approximate measure of unproductive government expenditures. An increase in this variable tends to generate negative impacts on economic growth. A rise in inflation tends to reduce cash balances, thus subjecting private economic agents to higher transaction costs. A rise in inflation tends to have a negative impact on growth (Amewoa, 2008).

International Research Journal of Finance and Economics - Issue 88 (2012) 31 4.2. Structure of the Panel of Countries under Study Sub-Saharan Africa, commonly known as Black Africa, stretches over an area of 24.3 million square kilometers. The region numbers 49 countries 19. All the countries of the region benefit from WRs to various degrees. However, the economic situation of Sub-Saharan African (SSA) countries has not ceased to deteriorate since the end of the 1970s 20. What is more, the countries of this region were thrown into a phase of economic decadence from which they do not seem to recover up to now, and this despite the continual inflows of foreign capital (ODA, FDI, and WRs). The recovery of growth for most of these countries in the mid-1990s marks the decisive phase of these countries future. To determine the relationship between WRs, poverty, and growth in these countries, we consider a panel of countries over the period 1994-2009 (see Tables 1 and 2 below). Table 1: Countries Retained in the SSA Region Benin Ghana Niger Botswana Guinea Conakry Nigeria Cameroon Kenya Senegal Cape Verde Lesotho Sierra Leone Ethiopia Malawi Togo Gabon Mali Source: By the authors from data available on FTMs The statistics on the three major forms of external financing in this area (see Table 2) enables us to describe the first trends: over the period 1994-2009, SSA countries benefited from increasing WRs inflows. The latter increased from about 2.5% of GDP in 1994 to about 5.6% in 2006. And yet, the rate of economic growth witnessed upward and downward changes during the period 1994-2005 and downward changes during the period 2006-2009. As estimated at 2006, this growth rate progressively decreased up to 2.8 at 2009. It is paradoxically at this time that WRs witnessed an explosion in terms of rate of increase. In effect, starting in 2005 the WRs growth rate became exponential unlike those of other forms of financing (ODA, FDI) whose evolution seemed to be slow. Over the period 2005-2009, we may note that average growth in the region seemed to be positive, and then it witnessed a constant decrease (see Figures 1 and 2 below). Table 2: Data on External Financing in Countries Retained in the Sample Country Data 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 WRs# 1.7 1.7 1.8 3.2 3.2 3.3 2.8 3.06 3.5 4.08 6.1 18.3 21.3 23.9 27.4 25.2 FDI# 2.1 2.2 2.4 2.2 1.9 1.7 1.9 2.1 2.5 2.5 2.7 2.6 2.5 2.7 2.4 3.2 ODA# 17.1 17.4 14.6 12.9 12.8 11.7 11.6 12.8 17.2 22.5 24 30.2 37.9 31.9 35.2 38.4 (TFM/GDP)% 2.3 2.4 2.2 2.1 2.1 2.1 2.0 2.1 2.3 2.6 2.7 3.4 3.6 3.7 3.4 3.5 (FDI/GDP)% 1.1 2.8 2.8 2.8 3.1 2.6 2.8 1.7 3.1 2.7 2.9 2.9 3.0 4.0 4.0 1.9 (ODA/GDP)% 15.9 13.1 11.0 9.3 9.7 8.3 9.4 11.0 10.0 11.0 11.0 9.7 9.5 9.6 9.2 8.9 GDP growth rate 2.5 4.6 5.2 3.5 4.2 2.5 3.1 4.8 4.5 5.0 4.5 4.8 5.6 5.3 5.2 2.8 94-95 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 05-06 06-07 07-08 08-09 Increase in FTMs 0.04-0.1-0.1 0.02 0.01-0.1 0.1 0.1 0.1 0 0.33 0.02 0-0.1 0.05 Increase in FDI 1.59 0 0 0.26-0.4 0.23-1 1.3-0.4 0.2 0.06 0.08 0.9 0.1-2.0 Increase in ODA 0.8 0.2-0.7 0.26-0.7 0.28 0.7-0.1 0.2 0 0.12 0.31-0.1 0-0.9 Source: Calculations by the authors from the data of the WDI (Banque mondiale, 2010). #In billion of US Dollar. 19 From the start, we have numbered 49 countries. South Africa was removed from the sample because its characteristics are different from those of other countries. Most of the data of countries such as Angola, Burkina Faso, Burundi, the Central African Republic (CAR), the Comoros, the Democratic Republic of the Congo (DRC), the Republic of the Congo, Côte d Ivoire, Equatorial Guinea, Eritrea, Gambia, Guinea-Bissau, Liberia, Madagascar, Mauritania, Zambia, Mozambique, Namibia, Rwanda, Sao Tome et Principe, the Seychelles, Somalia, Sudan, Swaziland, Tanzania, Uganda, Mayotte, and Zimbabwe are not available for various reasons specific to the history of each country. 20 The fall in the prices of raw materials and commodities toward the end of the 1970s aggravated the economic situation of the countries of this region.

32 International Research Journal of Finance and Economics - Issue 88 (2012) Figure 1: ODA, FDI, WRs as a percent of GDP. Figure 2: ODA, FDI, and WRs in absolute value. Source: Calculations by the authors from data of the WDI (Banque mondiale, 2010). 5. Estimations of the Models and Results We carried out different estimations for each of the models. With respect to poverty model (1), the estimator of instrumental variables (IVs) is used to overcome the problems linked to standard estimators 21. In effect, the IVs estimator is well suited to cross-section or panel data studies such as those encountered in the literature which examine a large number of relations between countries. In addition, model specification and estimation using IVs makes it possible to correct for the eventual specification biases caused by missing explanatory variables over the whole study period. The individual specific constant represents non random and invariant characteristics over time, specific to each country (Kristensen and Wawro, 2002). From our regression estimates we note that, all things being equal, a rise of 10% in WRs leads to an increase in poverty by the same proportions (see Table 3, Column 2 in the Appendix). This result seems to suggest that WRs inflows into SSA countries do not benefit the poorest populations. WRs rather seem to favor an increase in inequalities. This result can be justified, among others things, by the fact that the poverty levels being very high in the different countries, international migrations concern a minority of non poor households which can afford to finance these costly adventures. The members of these households, who are more capable of engaging in these migrations, seem to be the main beneficiaries of WRs. Meanwhile, poor households do not directly profit from this type of transfers which, for that reason, have the tendency to increase inequality and eventually poverty. The data in Table 4 in the Appendix reveal the existence of high levels of poverty and inequalities in most of the countries included in the panel. However, we note that human capital exerts a strong influence on the reduction of poverty. A rise of 1 percentage point in human capital leads to a reduction of a little more than 142 percentage points in poverty (see Table 3, Column 2). Table 3: Poverty, Growth, and WRs during the period 1994-2009 22 Regression with Instrumental Variables Poverty Model Investment 1.330746 (1.229034) WRs.1018474 (.6275169) Standard Error Correction Method Growth Model.0563807*** (.030582).0866241*** (.0463163) 21 For example, when applied to this model the least squares dummy variables (LSDVs) method or the within group method yield biased and inconsistent coefficient estimates. Baltagi (1995) recommends the use of instrumental Variables to solve these problems. 22 Standard deviations are between parentheses. *, **, *** these stars respectively indicate that the results are significant at the 1%, 5% and 10% significance levels.

International Research Journal of Finance and Economics - Issue 88 (2012) 33 Table 3: Poverty, Growth, and WRs during the period 1994-2009 22 - continued Inequality 1.713251* (.3209033) Human Capital -142.6494* (47.27475) PIB t-1.314835* (.0576933) Public spending -.2538485* (.0740403) Degree of openness -.0220104* (.0085811) Inflation.0021844 (.0024779) Constant 21.98902 (24.20322) 7.312982* (1.614318) Instrumented: PIB per head Autocorrelation: no autocorrelation Instruments: Degree of Inflation Estimated openness autocorrelations = 0 F test that all u_i=0: F(16,47) = 8.50 Prob > F = 0.0000 Estimated covariances= 153 sigma_u 13.340133 sigma_e rho.77742412 (fraction Number of obs = 68 Number of obs = 68 7.1378958 of variance due to u_i) Source: By the authors from the data of the World Bank R-squared = 0.5193 Wald chi2 =128.24 Prob > chi2 = 0.0000 The standard panel error correction method developed by Beck and Katz (1996), is used to estimate the growth model. The use of this method is justified, among others, by the fact that it is efficient in the estimation of panel models with fixed effects which take account of lagged endogenous variables 23. Moreover, the use of the standard panel error correction method improves the inferences made, given the complexity of the error processes 24. Estimation through the standard error correction method yields standard deviations which are more reliable than those of the other methods recommended in the literature 25. Furthermore, this method helps to obtain quality estimators when it is applied to small samples (Beck et Katz 1996). From the estimation of our growth model, we note that the coefficient associated with WRs is positive and statistically significant. An increase of 10% in WRs corresponds, all things being equal, to an rise of 8.6% in wealth (in terms of per capita GDP growth). According to these results, the impact of WRs on economic growth in SSA countries is predominant 26. These results show that WRs can relax the credit constraint for recipient households, thus making possible an increase in private investment. In addition, according to the literature, when WRs are used for private investment, they are extensively allocated to education and health which help to improve human capital. A proposition can be made at 22 Standard deviations are between parentheses. *, **, *** these stars respectively indicate that the results are significant at the 1%, 5% and 10% significance levels. 23 The introduction of the lagged endogenous variable is useful in the use of the standard panel error correction method for a couple of reasons. This method is appropriate only if serial correlation is not present in the data. A standard method (undoubtedly the preferred method) used to remove the serial correlation which often occurs in panel data, is to include a lagged dependent variable in the specification of the model (Kristensen and Wawro, 2003), which we have done in this to correct for serial correlation. 24 The errors in panel models may be associated with variances which differ according to transversal units owing to their specific characteristics (panel heteroscedasticity) or the serial correlation of errors. Beck and Katz (1995) say that the original way to go about things when manipulating complex error structures in the analysis of panels, is to estimate the coefficients with OLS, and then to implement the standard panel error correction method. 25 Within- group Estimators, LSDV 26 Contrary to the results obtained in a similar study on the countries of Asia and the Pacific where a rise in WRs of 1% leads to a 0.03% increase in growth (Jongwanich, 2007).

34 International Research Journal of Finance and Economics - Issue 88 (2012) this point in accordance with the literature which suggests the allocation of this resource to investment in physical capital and human capital. Since an improvement in the human Capital (HC) seems favorable to poverty reduction. Moreover, SSA States must give particular attention to WRs since this source of financing constitutes, according to our results, a key element in economic growth in the same way investment. We can thus note that the other coefficients of the variables associated with the growth model have the expected signs as in the literature, and most of them are statistically significant 27. The negative and significant sign of the coefficient associated with the degree of openness highlights the fragility of SSA countries to any trade liberalization policy linked to economic growth. Beyond these results, it is important to note that WRs may have an indirect impact on poverty reduction, since they positively affect economic growth, which is a key element in poverty reduction. As already noted in Section 3 above, WRs may generate income even in the families which do not receive these transfers through the multiplier effect linked to the increase in expenditures. Given that the families of migrants will increase their consumption of goods and services produced in the sectors with excess capacity, the additional demand may facilitate job creation for the other families which create additional demands through their consumption. This type of multiplier effect may favor poverty reduction, even for poor families which do not directly receive WRs. 6. Conclusion The objective of this study was to evaluate the impact of WRs on poverty and economic growth in SSA countries over the period 1994-2009. The developments carried out in this study have enabled us to note that, beyond the fact that the literature analyzing the impact of WRs on economic development is abundant and varied, it also comprises many debates. While some authors consider WRs as not having any implication on poverty reduction, on the one hand, and on the recovery of economic growth on the other, some other authors rather consider them as being essential for the development of poor countries. The ever increasing financing flows stemming from this source, even during a crisis period, strengthen this position. The methodology used in this study to find an answer to the question as to whether WRs participate in the development of recipient countries, rest on two models with different approaches to estimation. The first approach uses estimation with panel instrumental variables for the poverty model, and the second approach uses the standard panel error correction method to estimate the growth model. From the results of these estimations, we can generally conclude that, on the one hand, WRs do not contribute directly to the reduction of poverty but, on the other hand, they contribute significantly to economic growth. However, the fact that the coefficient associated with WRs is not significant despite the importance of this financing source as a vector of growth in SSA countries opens the way to various interpretations as to the possibility for WRs to reduce poverty in the long term. As a policy recommendation for the region, it is important for SSA States to consider WRs as a key element in economic growth in the same way as investment. This being so, policymakers must envisage the way WRs should be allocated to give impetus to development, notably through human capital formation and basic infrastructure development. Beyond all these implications, it is important for the governments to have better quality databases at their disposal on the evolution of transfers, since data on WRs are not always available. In some countries WRs are even assimilated to export receipts, to the deposits of non residents, or again to FDI. In addition, a large share of transfers is dispatched home through informal circuits, which keeps decision makers from realizing the real contribution of WRs to development in SSA countries. It 27 With the exception of inflation whose coefficient is not significant. However, this can be explained by the fact that most of these countries belong to a currency area (Benin, Cameroon, Gabon, Mali, Niger, Senegal, Togo, among others) in which monetary policy is almost inexistent.

International Research Journal of Finance and Economics - Issue 88 (2012) 35 is only at this price that it is possible to envisage WRs as an essential and very effective lever for the development of Sub-Saharan Africa. Bibliography [1] Adams, R.H., Jr. (1991), The Effects of International Remittances on Poverty, Inequality and Development in Rural Egypt, Research Report no 96, International Food Policy Research Institute. [2] Adams, R and J. Page (2005), Do International Migration and Remittances Reduce Poverty in Developing Countries?, World Development, Vol. 33 (10): 1645-69. [3] Adams R. and J. Page (2003), International Migration, Remittances and Poverty in Developing Countries, [4] Ahlburg, D. (1991), Remittances and Their Impact, A study of Tonga and Western Samoa, Pacific Policy Paper No.7, The Australian National University, Canberra. [5] Amewoa A., K, (2008) l Aide au Développement aide-t-elle le développement? Le cas de l Afrique Sub-saharienne. Thèse de doctorat en Sciences économiques, Université de Limoge. [6] Baltagi, Badi H. 1995. Econometric Analysis of Panel Data. New York: Wiley. [7] Banque mondiale (2003), World Development Indicators 2003, database. [8] Banque mondiale (2005), World Development Report, Washington DC, World Bank. [9] Banque mondiale (2006), Global Economic Prospects, World Bank, Washington DC. [10] Banque mondiale (2010), World Development Indicators 1999-2009, database. [11] Barro, R. (1996), Determinants of Economic Growth: A Cross-Country Empirical Study, NBER Working Paper 5698, National Bureau of Economic Research. [12] Barro, R. and X. Sala-i-Martin (1995), Economic Growth, McGraw-Hill, Cambridge, MA. [13] Beck, Nathaniel and Jonathan N. Katz. (1995). What To Do (and Not To Do) with Time- Series Cross-Section Data in Comparative Politics. American Political Science Review 89: 634 647. [14] Beck, Nathaniel and Jonathan N. Katz. (1996). Nuisance vs. Substance: Specifying and Estimating Time-Series-Cross-Section Models. Political Analysis 6: 1 36. [15] Berg A. and A. Krueger (2003), Trade, Growth, and Poverty: A Selective Survey, IMF Working Paper 03/30, International Monetary Fund, Washington DC. [16] Brown, R.P.C (1994), Migrants Remittances, Savings and Investment in the South Pacific, International Labour Review, Vol. 133 (3): 357-67. [17] Bugamelli M., and Paterno F., (2008) Output Growth Volatility and Remittances, Bank of Italy Temi di Discussione (Working Paper) No. 673. [18] Burnside Craig & David Dollar (2000), Aid, Policies, and Growth. American Economic Review, September 90 (4) pp 847-68. [19] Chami R., Barajas A., Cosimano T., Fullenkamp C., Gapen M., et Montiel P., (2008) Macroeconomic Consequences of Remittances, IMF Washington DC, Occasional Paper 259 [20] Chami R, Hakura D and Montiel P (2009) Remittances: An Automatic Output Stabilizer? IMF Working Papers: 91/2009. [21] Chauvet, L., Mesplé-Somps (2007) L aide publique au développement et les investissements directs étrangers en Afrique sont-ils complémentaires? Revue Tiers-Monde, n 192, pp. 815-832. [22] Clemens M., Radelet S., and Rikhil B., (2004), Aid and Growth : The Current Debate and Some New Evidence, Center for Global Development. [23] Collier Paul & Dollar David (2002), Aid Allocation and Poverty Reduction, European Economic Review, 46 (8), 1475-1500, World Bank. [24] De la Fuente (2008) Remittances and Vulnerability to Poverty in Rural Mexico World Institute Of Development Economic Research, Research Paper No. 2008/17

36 International Research Journal of Finance and Economics - Issue 88 (2012) [25] Dollar, D. and A. Kraay (2002), Growth is Good for the Poor, Journal of Economic Growth, Vol. 7 (3): 195-225. [26] Easterly William (2003), The Cartel of Good Intentions: The problem of Bureaucracy in Foreign Aid, Department of Economics, New York University and Center for Global Development, Policy Reform, Vol.00 pp 1-28. [27] Ebeke C., (2010). "Transferts des migrants, ouverture sur l'extérieur et dépenses publiques dans les pays en développement," Working Papers halshs-00552983 CERDI [28] Domar E., (1946) "Capital Expansion, Rate of Growth and Employment", Econometrica. [29] Fajnzylber, P., and H. Lopez. (2007). Close to Home: The Development Impact of Remittances in Latin America. The World Bank, Washington DC. [30] Giuliano, P., and M. Ruiz-Arranz. (2005). Remittances, Financial Development and Growth. IMF Working Paper Series, WP/05/234, IMF. [31] Global Development and Finance (2003). Workers' Remittances: An Important and Stable Source of External Development, World Bank GDF2003 [32] Gupta S, Pattillo A and Wagh S (2009) Effect of Remittances on Poverty and Financial Development in Sub-Saharan Africa, World Development, 37, 104-115. [33] Hall Robert and Chad Jones (1999), Why Do Countries Produce So More Output per Worker than Others?, Quarterly Journal of Economics, 114, 83-116. [34] Harris J.R et Todaro M.P. (1970), Migration, Unemployment and Development: A total sector analysis, AER, 60, 1 [35] Kapur, Devesh and Singer, David Adrew (2006) Remittances and Government Spending, and the Global Economy, Annual Meeting, San Diego, CA, March 22-25, 2006. [36] Keith, Linda Camp. (1999). The United Nations International Covenant on Civil and Political Rights: Does It Make a Difference in Human Rights Behavior? Journal of Peace Research 36 (1): 95 118. [37] Koechlin, V., and G. Leon. (2007). International Remittances and Income Inequality: An Empirical Investigation. Journal of Policy Review, Vol. 10, No. 2. [38] Kristensen I. P et Wawro G., (2003) Lagging the Dog?: The Robustness of Panel Corrected Standard Errors in the Presence of Serial Correlation and Observation Specific Effects, Working paper [39] Lucas, R.E.B. (1988) Emigration to South Africa s Mines. American Economic Review, Vol. 77, No. 3, pp. 313-330. Nagarajan 2009 [40] Ratha, D. (2005). Remittances: A Lifeline for Development. Finance and Development, vol. 42, no. 4, 42. [41] Ratha and al. (2007), Remittance Trends 2007, Migration and Development Brief 3, Development Prospects Group, Migration and Remittances Team, November [42] Russell et Stanton (1992), Migrants Remittances and Development, International Migration, Quarterly Review 30 (4/4); 267-87). [43] Ratha, D, (2003), Workers Remittances: An important and Stable Source of External Development Finance, Global Development Finance, 157 172. [44] Spatafora, N. (2005). Two current issues facing developing countries, World Economic Outlook, International Monetary Fund, Washington, DC. [45] Stark, O and D. Levhari (1982), On Migration and Risk in LDCs, Economic Development and Cultural Change, Vol. 31 (1): 191-96. [46] Todaro M. P. (1969), A model of labor migration and urban unemployment in less development countries, AEZ, 59 [47] Viet, C. N. (2008), Do foreign Remittances Matter to Poverty and Inequality?: Evidence from Vietnam, Economics Bulletin, Wageningen University. [48] Woodruff, C., & B. Zenteno. 2001. Remittances and Micro Enterprise in Mexico. UCSD Graduate School of International Relations and Pacific Studies Working Paper, University of California San Diego.