THE EFFECTS OF REMITTANCES ON ECONOMIC GROWTH IN SUB-SAHARAN AFRICA LEARNMORE MUCHEMWA SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS OF THE

Transcription

1 THE EFFECTS OF REMITTANCES ON ECONOMIC GROWTH IN SUB-SAHARAN AFRICA BY LEARNMORE MUCHEMWA SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS OF THE DEGREE OF MASTER OF COMMERCE IN ECONOMIC DEVELOPMENT AND POLICY ISSUES IN THE DEPARTMENT OF ECONOMICS AND ECONOMETRICS AT THE UNIVERSITY OF JOHANNESBURG Supervisor: Prof P.F. Blaauw Co-Supervisor: Prof F. Tregenna

2 DECLARATION I LEARNMORE MUCHEMWA Declare that THE EFFECTS OF REMITTANCES ON ECONOMIC GROWTH IN SUB-SAHARAN AFRICA Is my own work, that all sources used or quoted have been indicated and acknowledged by means of complete references, and that this research was not previously submitted by me for a degree at another University. Learnmore Muchemwa 2

3 ABSTRACT The subject of the growth effects of remittances is characterised by different and conflicting perspectives. While migration optimists believe in positive growth effects of remittances, migration pessimists, on the other hand, challenge this position and claim that remittances have either a negative or statistically insignificant effect on economic growth. Those for remittances argue that remittances have a positive effect on economic growth mainly through subsequent increases in investment capital and human capital. Migration pessimists, however, stress that remittances negatively impact economic growth, mainly, because of inflationary pressures and moral hazards that result in reduced labour supply. Given such contrasting literature, this study makes an attempt to contribute to the existing literature by assessing the growth-effects of remittances in twenty-nine Sub-Saharan Africa countries over the period The Arellano-Bover/Blundell-Bond GMM one-step estimator is used in the assessment. Empirical results from the study reveal evidence supporting for statistically significant positive growth effects of remittances in Sub-Saharan Africa. The study further reveals that these positive growth effects of remittances in Sub-Saharan Africa happen through the human capital channel. Even when heterogeneity of sub-regions is taken into account, there is still evidence showing positive growth effects of remittances in Sub-Saharan Africa. Results, however, reveal that in West Africa, remittances have a low positive effect on economic growth. 3

4 ACKNOWLEDGEMENTS First and foremost, I thank the Almighty God for equipping me with the strength to carry out this study. I would also like to appreciate and thank my two academic supervisors, Professor Derick Blaauw and Professor Fiona Tregenna for their knowledge input, support and guidance throughout the stages of this study. This academic piece is what it is today because of you. My sincere gratitude also goes to my other lecturers, Mr Arnold Wentzel, Professor Alain Kabundi and Professor Stephen Gelb for equipping me with the knowledge that has been a solid base for this academic piece. I also want to thank my family for the financial and emotional support. Last but not least, I would like to thank all my colleagues for all the constructive comments and support. 4

5 TABLE OF CONTENTS CHAPTER ONE INTRODUCTORY BACKGROUND Background Research problem/question Hypotheses of the study Significance of the research Structure of the Study CHAPTER TWO LITERATURE REVIEW Introduction Background literature on remittances Theories and behavioural patterns of remittances Relating globalisation to labour migration and remittances Growth effects of conventional external capital flows Growth effects of remittances Growth effects of remittances through domestic savings and investment Growth effects of remittances through consumption Growth effects of remittances through the Human Capital Investments channel Growth effects of remittances through labour supply Growth effects of remittances through exchange rate and export performance Growth effects of remittances through extra demand The growth effects of remittances through financial development Conflicting literature on the growth effects of remittances Empirical evidence from earlier studies Conclusion CHAPTER THREE CONTEXTUAL BACKGROUND Introduction The Sub-Saharan African context SADC and the imbalanced labour migration flows Aggregate trends of remittance inflows in Sub-Saharan Africa Growth patterns in developing regions Conclusion CHAPTER

6 RESEARCH METHODOLOGY Introduction Research design Model specification A priori expectations Data sources Comparative analysis Methods of estimation System dynamic panel data estimation Diagnostic tests Conclusion CHAPTER EMPIRICAL ANALYSIS Introduction Data description Model results Primary model estimations Baseline model results Supplementary regressions Sub-regional analysis Diagnostic tests results Conclusion CHAPTER CONCLUSION Introduction Research findings Policy recommendations BIBLIOGRAPHY APPENDIX Appendix 1: List of Sub-Saharan African countries used in the study Appendix 2: Estimations for the entire panel (n=29, t=29) Appendix 3: Baseline model estimations (n=29, t=6) Appendix 4: Growth model estimations using remittances as a share of GDP (n=29, t=6) Appendix 5: Growth model estimations: without outliers (n=24, t=6) Appendix 6: Growth model estimations without the investment variable (n=29, t=6) Appendix 7: Growth model estimations without the enrolment variable (n=29, t=6)

7 Appendix 9: The Sargan test of over identifying restrictions Appendix 10: Test for autocorrelation

8 LIST OF TABLES Table 2.1 Channels under which remittances effect economic growth Table 2.2 Summary of contrasting literature on growth effects of remittances Table 2.3 Summary of previous empirical findings Table 3.1 Income groups of different countries in Sub-Saharan Africa Table 5.1 Descriptive statistics summary Table 5.2 Correlation matrix Table 5.3 Estimations for the entire panel (n=29, t=29) Table 5.4 Baseline model estimations (n=29, t=6) Table 5.5 Growth model estimations using remittances as a share of GDP (n=29, t=6) Table 5.6 Growth model estimations: without outliers (n=24, t=6) Table 5.7 Growth model estimations without the investment variable (n=29, t=6) Table 5.8 Growth model estimations without the enrolment variable (n=29, t=6) Table 5.9 Growth model estimations with dummy variables for West and Southern Africa..74 LIST OF FIGURES Figure 3.1 Remittances, Foreign Direct Investment and Official Development Assistance inflows to Developing Economies Figure 3.2 Distribution of Remittance, ODI and FDI inflows in Sub-Saharan Africa Figure 3.3 Growth patterns in developing regions Figure 5.1 Relationship between remittances and growth Figure 5.2 Scatter plot illustrating the relationship between growth and remittances Figure 5.3 Correlation between averaged remittances and averaged growth

9 ABBREVIATIONS 2 SLS Two Stage Least Squares FDI FE GDP GMM GNP IMF LSDV NELM ODA OECD OLS RE Foreign Direct Investment Fixed Effects Gross Domestic Product Generalised Methods of Moments Gross National Product International Monetary Fund Least Squares Dummy Variable New Economics of Labour Migration Official Development Assistance Organisation for Economic Cooperation and Development Ordinary Least Squares Random Effects 9

10 CHAPTER ONE INTRODUCTORY BACKGROUND 1.1 Background Low income levels are one of the greatest economic challenges facing a majority of countries in Sub-Saharan Africa (SSA). Data from the World Bank Development Indicators indicate that close to 85 per cent of countries within Sub-Saharan Africa are classified under the low and lower middle income country categories (World Bank, 2010). Such low level income classifications of the majority of countries in Sub-Saharan Africa reveal the challenge of limited economic development currently holding in the region at large. Supporting the notion of limited economic development within the region, Garner (2006:3) describes Sub-Saharan Africa as a region popular for underdevelopment and...non-existent economic growth. Sub-Saharan Africa as a region has failed to move at par with some other developing regions regarding overall economic growth levels. In fact, in Sub-Saharan Africa, low levels of GDP can be traced as one of the lead causes of poverty. The United Nations (2009:15) reports how the share of people living in poverty in Sub-Saharan Africa as a whole more than doubled between 1981 and Because of poor economic performance and extensive poverty coupled with other factors like excessive population growth and unstable political factors, many people in Sub-Saharan Africa have resorted to international migration (Adepoju, 2008:5). Towards the end of the twentieth century the subject on international migration has attracted much attention from economic scholars. Scholars such as Wong and Yip (1999), Hague and Kim (1995), Vidal (1998), Mountford (1997), and Stark, Helmenstein and Prskawetz (1997) led studies that examined the relationship between migration and economic growth. In most of these studies, the issue of migrant s remittances emerges as one particular area that dominates the migration discourse. Since remittances are a major feature of international migration, it is critical to assess their impact on economic activity in Sub-Saharan Africa, a region dominated by excessive migration and low levels of relative economic growth. One way of doing this is to assess the effects of remittances on economic growth in recipient countries in Sub-Saharan Africa. It remains to be established whether remittances are a solution to the challenge of low levels of economic growth persisting in most countries within Sub-Saharan Africa. 10

11 1.2 Research problem/question The literature on how remittances impact economic growth of recipient countries is conflicting. Some scholars believe that migrant remittances have positive growth effects in recipient economies (Pradhan, Upadhyay & Upadhyaya, 2008; Fayissa & Nsiah, 2010b; de Haas, 2005; Dos Santos & Vinay, 2003) while other scholars highlight the negative growth effects of remittances (Chami, Fullenkamp & Jahjah, 2003; Karagoz, 2009). The latter argue that remittances do not result in positive economic growth since the two variables are negatively correlated. Adding to the debate, there are also scholars who claim that remittances have no impact on economic growth of recipient countries (Barajas, Chami, Fullenkamp, Gapen & Montiel, 2009; Rao & Hassan, 2011). For these scholars, there is no causal relationship between remittances and economic growth of developing economies. All these conflicting empirical findings on the growth effects of remittances are, to some extent, informed by the available theoretical literature conversing on the channels through which remittances impact economic growth. There are a lot of different perspectives in both the empirical and theoretical literature on the growth effects of remittances. Available theoretical literature on the growth effects of remittances can be categorised into two main schools of thought. The two schools of thought include the migration optimists and the migration pessimists. Migration optimists argue for positive growth effects of remittances. They demonstrate the positive indirect growth effects of remittances through economic channels such as increased savings, investment capital, human capital investments, extra employment and the overall multiplier effects of consumption on aggregate demand and output (Adenutsi, 2010:34; Balde, 2010:17). Unlike the migration optimists, migration pessimists argue against the positive growth effects of remittances. According to the migration pessimists, remittances have either negative growth effects or zero impact on economic growth. They argue that remittances are mostly used for consumption instead of productive investments as argued by the migration optimists (de Haas, 2007:5). They also argue that remittances create moral hazard problems which reduce labour supply in recipient economies (Chami et al., 2003:5). Migration pessimists also indicate that remittances have negative growth effects also as a result of reduced human capital investments (Chami et al., 2003:5) and inflationary pressures (Amuedo-Dorantes & Pozo, 2004:1408). Both the two schools of thought use the same channels namely; 11

12 consumption, human capital investment and labour supply, to highlight their contrasting evidence on the growth effects of remittances. As a result of the contested literature, it is difficult for one to conclude on the growth effects of remittances in a region like Sub-Saharan Africa. There is a need to examine the growth effects of remittances and answer the research question that follows: How do remittances impact economic growth of recipient countries in Sub-Saharan Africa? 1.3 Hypotheses of the study A negative and statistically significant co-efficient of remittances indicates that remittances have a negative impact on economic growth. An increase in remittances will result in lower economic growth. A positive and statistically significant co-efficient of remittances implies that remittances have positive growth effects. An increase in the volumes of remittances will result in increased economic growth. A statistically insignificant co-efficient of remittances implies that remittances do not have significant direct growth effects. In other words, either the positive or negative impact of remittances on economic growth will be low. 1.4 Significance of the research Over the past years there has been an increase in the number of studies conversing on the subject of remittances. Most of the studies focused entirely on the growth effects of remittances in developing economies in general. Very little has been done on remittances in Sub-Saharan Africa or its sub-regions in particular. Campbell and Kandala (2011:130) note the scarcity of studies examining the macroeconomic effects of remittances in Sub-Saharan Africa. There is a need for a study that distinguishes Sub-Saharan Africa from the rest of the developing world in assessing the growth effects of remittances. Sub-Saharan Africa is a region that is unique in its migration and remittance inflow patterns hence the need for a 12

13 study that takes such heterogeneity into consideration (Gupta, Pattillo & Wagh, 2008:105; Adenutsi, 2010:33). Most of the available literature on remittances in Sub-Saharan Africa is not entirely focused on the growth effects of remittances. Gupta et al. (2008) led a study that assessed the impact of remittances on poverty and financial development in Sub-Saharan Africa. Anyanwu and Erhijakpor (2010) analysed the overall effects of remittances on poverty in Africa while Fayissa and Nsiah (2010b) assessed the impact of remittances on economic growth and development in Africa. Salisu (2005) looked at the determinants of remittances in Sub-Saharan Africa. Singh, Haacker, Lee and Goff (2010) not only looked at the determinants of remittances, instead, they went on further to establish the macroeconomic role of remittances in Sub-Saharan Africa. This study comes in to complement the fewavailable research studies on the growth effects of remittances in Sub-Saharan Africa. In examining the relationship between remittances and economic growth in Sub-Saharan Africa, the study also incorporates results from two Sub-Saharan Africa sub-regions. This methodology enables an analysis of the growth effects of remittances in both Sub-Saharan Africa, and two of its sub-regions are utilised in order to cover some of the prevailing literature gaps on the subject of remittances and economic growth in the region. Besides being specific to the Sub-Saharan Africa context, the study employs a macroeconomic approach to the assessment of remittances. Focal interest is on the relationship between remittances and economic growth of twenty-nine Sub-Saharan African countries. Some studies such as the one by Lucas and Stark (1985) are micro-economic in approach. They seek to understand the behavioural patterns of remittances at household level. Much attention is placed on the motivations and spending patterns around remittances. Understanding the mechanics of remittances is an important aspect that allows for informed assessments of the macro-economic effects of such remittances. This research, therefore, complements all the other studies on remittances that are micro-economic in focus. It helps to shed light on how remittances affect macro-economic variables such as economic growth. In order to try and come up with justified and unbiased results on the growth effects of remittances in recipient countries in Sub-Saharan Africa, this study employs modern panel data estimation methods. Such modern panel data estimation tools and methodology distinguish it from prior studies that had challenges in handling endogeneity between remittances and economic growth, a feature which resulted in most of the studies coming up with questionable results due to misspecified models. 13

14 1.5 Structure of the study Having introduced the research, the rest of the study is presented as follows: Chapter 2 outlines the literature on the growth effects of remittances; Chapter 3 presents the contextual background information on remittances in Sub-Saharan Africa. Chapter 4 gives a presentation of the research methodology and model specification while Chapter 5 reports on the results and Chapter 6 sums up the study with the conclusion and policy recommendations. 14

15 CHAPTER TWO LITERATURE REVIEW 2.1 Introduction This chapter outlines the different theoretical and empirical discussions on remittances covered in the migration literature. The subject on remittances remains one of the most researched about areas in migration economics. In many of the studies, the motivation has been that of trying to understand remittances as a welfare-enhancing tool, that is, whether remittances as a private source of capital can be of significance in improving the lives of individuals in different parts of the world. It is against this background that this study explores how the available literature can be of significance in helping us assess the contribution of remittances in enhancing human welfare. In order for one to be able to model a comprehensive assessment of how remittances can impact economic growth, it is critical that the study looks into the theory of growth as presented in the literature. The relationships between capital, investment, savings, consumption and growth will constitute a vital framework for this study. As will be elaborated in the forthcoming subsections, there are diverse opinions within migration literature when it comes to the question of whether remittances are mainly invested or consumed. The conventional approach presents remittances as positively impacting economic growth only if they are invested. If remittances are only used for consumption then their potential to positively impact economic growth becomes questionable. There are circles however that argue that remittances can still boost economic growth even if they are used for consumption. Such an argument signifies how the literature is highly contested. It is against this background of contrasting literature that this chapter will scrutinise the available literature and relate it to the Sub-Saharan African context in an effort to fill the gaps on how remittances and economic growth are related. The chapter begins by presenting the different definitions available for remittances. The chapter goes on to discuss the theories explaining for the different behavioural patterns of remittances. The discussion on the behavioural patterns of remittances is then followed by another discussion relating globalisation to labour migration and remittances. This is then followed by a brief theoretical outline of the growth effects of other external sources of 15

16 capital such as Foreign Direct Investment (FDI) and foreign aid. The rest of the chapter is predominantly focused on the growth effects of remittances. 2.2 Background literature on remittances Migration literature defines remittances in two different ways. According to Rajan (2006:3) the narrowest definition of remittances is restricted to the transfer of funds made by labour migrants to their countries of origin (Karagoz, 2009:1892; Jongwanich, 2007:2). The funds are normally presented in the migration literature as being sent to families or relatives back home. There is however a more broad or formal definition of remittances derived from the IMF Balance of Payments Yearbook that incorporates compensation of employees and migrants transfers to workers remittances (Rajan, 2006:3; Jongwanich, 2007:2; Salomone, 2006:2). In the IMF statistics, compensation of employees is accounted for in the income component. Migrants transfers are included in the capital transfers while workers remittances form part of the current transfers (Jongwanich, 2007:2; Salomone, 2006:2). This study makes use of the much broader definition derived from the IMF Balance of Payments Yearbook. As noted by Jongwanich (2007:2), there are quite a number of limitations present in accounting for remittances in different countries around the world. The derivations and definitions of remittances tend to differ amongst countries. There are different channels that migrants may make use of when sending remittances to their countries of origin. Large amounts of remittances are not captured simply because they are sent through informal channels while those sent through the formal channels might tend to be understated as a result of some weaknesses in the data collection systems (Jongwanich, 2007:2). These limitations can to some extent distort the conclusions normally drawn in most remittances studies especially those with a macro-economic orientation. Having looked at how remittances are defined and some of the limitations in the data computations, it is critical that we look at the determinants of the levels of migrants remittance flows. The OECD (2006:145) mentions two important factors explaining for the level of migrants remittance flows. The first factor is the individual migrant s income and savings which influence the amount of money he/she might be able to remit back home. The other factor involves the residence status of the migrant in the host country and his/her 16

17 intentions with regard to whether he/she is going to be in the host country temporarily or permanently. The OECD (2006:145) highlights that network effects and the welfare of the migrant s family left in the home country are other significant factors that explain for the level of remittances flow. One of the factors responsible for the level of migrants remittance flows is the issue of motivation to remit (OECD, 2006:145; de Haas, 2007:7; Vargas-Silva, 2008:292; Brown, 2006:62; Fayissa & Nsiah, 2010b:94; Chami et al., 2003:3; Schiopu & Siegfried, 2006:8). What motivates migrants to remit is an important subject especially when trying to assess how remittances impact economic growth. In order to answer the question on the relationship between economic growth and remittances it is important that one looks at the objectives of the migrants who are remitting the funds back home. Chami et al. (2003:3) indicate that, in order for remittances to positively impact development, there is a need for people to understand the behavioural patterns of such remittances. Such behavioural patterns are drawn from what motivates migrants to remit funds back home. These motives have been turned into microeconomic theories which help explain for the flow of remittances Theories and behavioural patterns of remittances The literature on remittances identifies three theories to explain the flow of remittances. These are; Pure Altruism, Pure Self Interest and Tempered Altruism which is also referred to as Enlightened Self Interest. Most discussions in the literature are centred on the first two of them. These theories illustrate that remittances are sent mainly as a result of pure altruistic and self-interest motives (de Haas, 2007:7; Schiopu & Siegfried, 2006:8; Hagen-Zanker & Siegel, 2007:4; Lucas & Stark, 1985:902). The literature on remittances is however conflicting when it comes to the question of determining which of the two motives listed above better explains the increased flow of remittances. Both altruistic and self-interest motives have been argued to be factors responsible for the increased or decreased flow of remittances. Sayan (2006:5) challenges the idea of viewing altruism as the only motive for remitting. He asserts that remitting is a multifaceted behaviour hence it involves many other explanations besides altruistic and self-interest motives. This study will only discuss the two main theories used to explain for remittances. These two theories are crucial in providing the framework of knowledge explaining for the different growth-remittance inflow patterns in Sub-Saharan African countries. 17

18 The Pure Altruism theory highlights that migrants remit money back home in concern of the welfare of the remaining family members (Hagen-Zanker & Siegel, 2007:5; OECD, 2006:145). Chami et al. (2003:4) report that in this model, the migrant s utility is derived from that of his/her family back home. The migrant is rather satisfied when the welfare of his family back home is better off (OECD, 2006:145). This implies that the migrant is motivated to remit more funds to his family when there are unfavourable economic conditions holding in the home country. The theory observes that remittances are compensatory transfers since they increase when the migrant s home country is faced with economic disruptions such as droughts and a financial crisis (Chami et al., 2003:4). In order for the migrant to remit more funds, the economic disruptions or bad luck, a term used by Chami et al. (2003:4), must be creating a shortfall for the remaining family. As a result, the compensatory nature of remittances under the Pure Altruism model implies that remittances are countercyclical, that is, they increase during times when there is deterioration in economic conditions in the business cycle (Vargas-Silva, 2008:292; Chami et al., 2003:4). The Bank of Uganda (2007) emphasises that altruistic remittances can be countercyclical to GDP patterns possibly because migrants tend to remit more during periods of economic disturbances in order for their families in the home country to smoothen their consumption. Also commenting on behavioural patterns of remittances under a Pure Altruism model, Brown (2006:63) suggests that there is an inverse relationship between the volumes of remittances and economic conditions holding in the home country. Under this model, favourable economic conditions in the home country would imply a reduction in the volume of remittance inflows. The Pure Self Interest theory is modelled around the argument that remittances are not always countercyclical. There are some instances or contexts where volumes of remittances reduce following poor economic conditions in the recipient country. In such a case, there is no inverse relationship between volumes of remittances and the economic performance of the home country as postulated by Brown (2006:63). In fact, there might be a positive correlation between volumes of remittances and economic performance of the home country where bad economic conditions may result in low volumes of remittances. Such behavioural patterns have led to the formulation of the Pure Self Interest theory. Lucas and Stark (1985:904) claim that migrants self-interest can be one other motive for remittances. In this context, migrants remit money in order for them to invest or inherit in assets back home and also for them to return home with dignity (Hagen-Zanker & Siegel, 2007:5; OECD, 2006:146). When there is deterioration in economic performance of the home country, migrants are most likely to remit 18

19 less since the situation will have a negative impact on both investible and inheritable assets. There is most likely to be an increase in the volumes of remittances if the home economy is undergoing a favourable spell. It is however significant to note that both the Pure Altruism and Pure Self-Interest models do not account for all the reasons that motivate migrants to remit to their home countries (Lucas & Stark, 1985:904). There are some other different behavioural patterns of remittances which can be influenced by other factors not explicitly highlighted in the literature. Factors such as the relationship between the migrant and the remittances recipient, distance between host and recipient country and the costs involved in remitting income back home can influence the patterns of remittance inflows. Channels through which migrants remit their income back home have some influence on the overall cost of remittance transmission Relating globalisation to labour migration and remittances Globalisation has been widely used to explain increased global volumes in the flow of labour migration (Trimikliniotis, Gordon & Zondo, 2008:1324). A clear definition of globalisation can help one to derive the relationship between globalisation and labour migration. Martens and Raza (2010:280) define globalisation as the increased international movement of goods and services, financial capital, information and people. From this definition, the increased international movement of people can to some extent be equated to labour migration. It can thus be noted that labour migration is a function of globalisation since it results in an increased demand for skilled labour internationally. The need for skilled labour results in many countries importing labour. There is indeed a causal link that exists between globalisation and labour migration. The increased flow of goods and capital around the world as a result of globalisation has also resulted in increased volumes of labour migration (Molina, 2007:1). Molina (2007:1) furthermore highlights that the increased labour migration volumes coming out of globalisation have resulted in increased remittance flows around the world. Brown (2006:55) reaffirms this argument by asserting how remittances have increasingly become a reference point of globalisation. To date, remittances have become a vital source of capital for many developing countries. Acosta, Lartey and Mandelman (2009:102) claim that over the past years the volumes of remittance flows into developing countries have surpassed those of other external capital flows like official aid. 19

20 Not only is globalisation resulting in increased remittance inflows, one can also argue that external capital flows like aid and foreign direct investment are a result of globalisation. It is imperative to briefly assess the growth effects of foreign direct investment (FDI) and foreign aid since these are the more common sources of external capital for developing countries Growth effects of conventional external capital flows FDI can positively impact economic growth of recipient countries (Yasin, 2005:23; Borensztein, De Gregorio & Lee, 1998:115). It is believed that FDI, a source of external capital which is mainly associated with multinational corporations, results in capital accumulation and the transfers of technological advancement coupled with managerial expertise (Yasin, 2005:24; Alfaro, 2003:2; Borensztein et al., 1998:115). Recent growth literature argues that technological diffusion is an important element explaining economic growth and development in modern day economies (Borensztein et al., 1998:115). If developing economies are to grow, there is a need for them to catch-up with developed countries in terms of technological advancement. It is then through FDI that developing nations have access to such technology. While de Mello (1999:135) highlights that the main channel under which FDI contributes to positive economic growth is that of increment in physical capital in developing nations, Borensztein et al. (1998:118) dispute this notion and emphasise that stimulation of technological progress happens to be the core channel. It is also believed that FDI has some positive indirect effects on human capital development which also bear some overall positive impact on the economic development of developing nations (Alfaro, 2003:2; de Mello, 1999:135). Foreign aid, portfolio investment and private debt are some other important conventional sources of external capital mostly used by developing countries. Traditional growth theories like the 2-gap model argued that the savings constraint is one of the main barriers to economic growth (Iqbal, 1995:1119). Foreign aid can thus be used to offset this constraint and supply much needed capital in developing nations (Ali & Isse, 2006:242). Molina (2007:69) argues that foreign aid is normally disbursed to countries with limited gross national savings. By so doing, foreign aid can trigger long-run economic growth as observed by Minoiu and Reddy (2010:27). Asteriou (2008) supports the argument of a positive relationship between foreign aid and long-run economic growth. The main attraction in the foreign aid-economic growth literature is that aid helps increase stock of physical and human 20

21 capital in developing countries (Burke & Esfahani, 2006:351). It can thus be employed as a liquidity mechanism that allows for developing nations to have operating capital that calibrates huge developmental projects in developing economies. Having reviewed the growth effects of foreign direct investment and foreign aid, it is now important for the study to focus more on growth effects of remittances as they are presented in the literature. 2.3 Growth effects of remittances Having looked at the various channels through which other external sources of capital like foreign direct investment and official development assistance impact economic growth, it is now necessary to consider the different macro-economic growth effects of remittances as discussed in the literature. The literature on remittances is filled with conflicting views when it comes to the question of how remittances directly or indirectly impact economic growth of developing economies. A lot has been said and argued about the contribution of remittances to economic growth of developing countries but to date there isn t any consensus in the literature regarding this discussion 1. First, the study outlines the main channels under which remittances effect economic growth as presented in the literature. These can be presented as follows: 1 Refer table 2.2 and table 2.3 of this study. 21

22 Table 2.1 Channels under which remittances effect economic growth Channel Literature Domestic savings and investment Brown (2006:61); Gupta et al. (2008:105); Adenutsi (2010:34); Catrinescu, Leon- Consumption Ledesma, Piracha and Quillin (2009:81); Balde (2010:17); de Haas (2007:14); Ratha (2003:157); Karagoz (2009:1898); Drinkwater, Levine and Lotti (2003:1) Human capital investments Barajas et al. (2009:6); de Haas (2007:6); Brown (2006:65) Labour supply Brown (2006:67); Barajas et al. (2009:6); Adenutsi (2010:36); Acosta, Lartey & Mandelman (2009:104); Pradhan et al. (2008:498); Chami et al. (2003:5) Exchange rate and export performance Barajas et al. (2009:8); Gupta et al. (2008:105); Acosta, Lartey & Mandelman (2009:114); Catrinescu et al. (2009:81); Acosta, Baerg and Mandelman (2009:2); Amuedo-Dorantes and Pozo (2004:1408); Pradhan et al. (2008:498); Karagoz (2009:1899); OECD (2006:156) Extra demand OECD (2006:156); Catrinescu et al. (2009:81) Financial development Guiliano and Ruiz-Arranz (2009:147); Bettin and Zazzaro (2009:2); Gupta et al. (2008:104); Rao and Hassan (2011:701); Guiliano and Ruiz-Arranz (2009:144); Acosta, Baerg and Mandelman (2009:3) Migration literature is filled with many conflicting arguments when it comes to the subject of how the channels listed in Table 2.1 influence the growth effects of remittances. The literature is also not clear how the mentioned channels contribute to the growth effects of remittances. The following sub-sections present a detailed discussion on how each of the identified channels directly or indirectly influences the growth-effects of remittances in developing countries. 22

23 2.3.1 Growth effects of remittances through domestic savings and investment Like most private capital flows, remittances can have significant macroeconomic effects on growth in developing economies (Fayissa & Nsiah, 2010b). Investment is one of the channels through which remittances directly or indirectly impact economic growth. Some economists believe that remittances play a similar role like other external sources of capital such as foreign aid and foreign direct investment in boosting national savings in developing economies (Molina, 2007:69; Barajas et al., 2009:4). Adenutsi (2010:34) stresses that developmental migration optimists argue that remittances can have a positive impact on economic growth of developing economies as a result of investment capital coming out from the remitted funds. In other words, remittances are believed to provide the much needed capital for households which they can use for savings and also to finance investments (Catrinescu et al., 2009:81). Ratha (2003:157) observes that when remittances are channelled to countries that have healthy and sound economic policies, most of them can be invested in productive projects. When remittances are saved or productively invested, they can have an indirect positive impact on economic growth of the recipient economy (Balde, 2010:17). In his argument, Balde (2010:17) observes that remittances alone do not have a direct positive effect on economic growth but they do have an indirect positive effect that is channelled through the generation of savings and investments. Ratha (2003:164) notes that when remittances are invested in developing economies, a higher output growth can be realised in that particular economy. Karagoz (2009:1898) supports the argument that the most significant channel under which remittances can have a positive effect on economic growth in developing countries is through savings and investment. Drinkwater et al. (2003:2) further discuss the manner in which the growth effects of remittances operate. It is believed that remittances help ease credit constraints that happen to be the most common challenge associated with private enterprises in many developing countries. Drinkwater et al. (2003:2) argue that remittances can have similar growth effects like those of foreign direct investment when it comes to easing such financial constraints holding at national level. Many private enterprises in developing economies operate at limited production levels due to the investment challenges they encounter. Such challenges can be addressed by remittance flows provided the country s financial system is effective in allocating financial resources to where they are needed most. 23

24 2.3.2 Growth effects of remittances through consumption While developmental migration optimists argue that remittances can have a positive growth effect through increased savings and investments, migration pessimists refute such a notion and argue that remittances are rarely used for productive investments but are rather usually allocated for consumption (de Haas, 2007:5). Brown (2006:61) supports the claim that remittances are usually used for consumption by reporting how huge shares of remitted income are usually used for food. Barajas et al. (2009:6) make use of two arguments to elaborate on why a larger chunk of remitted income is used for consumption instead of being saved or invested as argued by developmental migration optimists. The first argument claims that, since remittances are compensatory in nature, they are more likely to be channelled or directed towards families with a higher propensity to consume as compared to families with a higher propensity to invest or save. Secondly, they maintain that due to the perceived permanency of remittances by receiving households, there is likely to be a moral hazard problem where the receiving households are more likely to use the income from remittances for consumption and not for productive investments. As a result, a larger share of remittances might then be used for consumption alone, a situation which is argued as having a negative effect on economic growth of developing countries (de Haas, 2007:14). Also in support of the argument that a greater share of remittances is mainly used to finance consumption is the OECD (2006:154) which claims that remittances like any other sources of income are likely to be spent basing on the hierarchy of needs. According to the hierarchy of needs, consumption is at the initial stages and dominates a bigger share hence most households are likely to spend their income on consumption as compared to investment. While it has been widely claimed that remittances are mainly used for consumption and not for productive investments, recent evidence suggests that even when used for consumption, remittances can still result in some positive growth effects due to the multiplier effects on aggregate demand and output in the entire economy (Brown, 2006:65; Gupta et al., 2008:105; Pradhan et al., 2008:498; Ratha, 2003:164; Ajayi, Ijaiya, Ijaiya, Bello, Ijaiya & Adeyemi, 2009:79). This evidence further complicates the subject of growth effects of remittances especially when related to the issue of savings and consumption. The argument that remittances will always positively impact economic growth of a recipient country even when they are used to finance consumption is mainly used by the 24

25 developmental migration optimists to refute the migration pessimists argument that concludes on negative growth effects of remittances due to consumption. Brown (2006:65) makes use of the Keynesian multiplier effects to explain the positive indirect growth effects that can be realised from remittances. It is believed that the increased consumption which implies increased spending may generate an increased demand which results in increased production. Barajas et al. (2009:3) support the above argument by referring to the case study of Pakistan and Mexico where evidence of the existence of such multiplier effects within the two economies was established. In trying to consolidate the whole argument, Ajayi et al. (2009:079) observe that when remittances are invested they result in output growth while if they are used for consumption, they can still still positively impact economic growth of recipient countries through the multiplier effects on aggregate demand and output. Having looked at the relationship between remittances and economic growth that is influenced by consumption, the next sub-section looks at human capital investments. The belief that a bigger percentage of income from remittances is used to help fund education of the remaining recipient family members is common in migration literature. The next subsection explains how human capital investments can be a channel through which remittances indirectly effect economic growth of the recipient economy Growth effects of remittances through the Human Capital Investments channel Migrants remittances might impact economic growth through human capital investments. Remittances boost human capital investments when they are used to finance the education and health of recipients. According to the human capital theory, more human capital investments result in positive economic growth for a country in the long-run (Olaniyan & Okemakinde, 2008:158). Barajas et al. (2009:6) mentions that remittances help boost human capital investments which are critical in economic growth through financing formal schooling of the receiving households. They however observe that positive economic growth would only be possible when those who have received the formal education are employed in the labour markets of that particular country. In other words, remittances can indirectly contribute to positive economic growth in a recipient country when individuals who receive the formal education financed through remittances supply their labour to the local labour markets. 25

26 According to de Haas (2007:6), remittances can either have a promoting or discouraging influence on the education of remaining citizens. Scholars following New Economics of Labour Migration (NELM), a hypothesis developed within the American research context responding to both the migration optimists and pessimists, do believe that remittances impact positively on education levels (de Haas, 2007:6). Their argument is motivated by the assumption that income from migrants will mainly be used to finance the remaining family members education. Remittances are here perceived as a means of financing development through education in many developing countries. Remittances can help boost human capital in recipient countries through financing the health care of recipients. When remittances are used to finance health care in recipient countries as highlighted by Brown (2006:65), this can have positive impacts on human capital development which in the long run leads to positive economic growth. It is however important to determine whether a greater share of remitted income is used for education and health care in developing countries. Not many studies have revealed and supported the argument that remittances are being used to finance education and health care in developing countries Growth effects of remittances through labour supply While it is universally acknowledged that labour plays a vital role in economic growth, it is critical to assess the growth effects of remittances through the labour supply and effort channel. The argument that remittances result in some moral hazard problems was first raised by Chami et al. (2003). Chami et al. (2003:5) argue that due to information asymmetry holding between the migrant and the remittances recipient household, moral hazards through reduced labour supply by the members of the recipient households are likely to negatively impact economic growth of the recipient economy. Many scholars support this claim and make use of different terms to describe the moral hazard problem emanating from remittances. Adenutsi (2010:36) supports the moral hazard argument and refers to the reduced labour supply by the recipient household members as voluntary unemployment resulting from increased remittance flows. Pradhan et al. (2008:498) use the term idleness to explain the moral hazard problem, a situation where recipients of remittances either reduce their labour effort or completely refrain from working. It is believed that the recipient households may use the remittances for increased leisure while reducing their labour supply 26

27 (Karagoz, 2009:1899; Barajas et al., 2009:7). Acosta,Lartey and Mandelman (2009:104) reveal how the moral hazard problem negatively affects economic growth in the recipient economy. They argue that reduced labour supply results in higher wages which results in increased production costs. In the long run, such production costs may have an adverse effect on output. This argument is related with the discussion on the Dutch-disease negative effects of remittances explained through exchange rate appreciation and export performance Growth effects of remittances through exchange rate and export performance According to Gupta et al. (2008:105) the Dutch disease effects of remittances is a subject that is strongly contested in migration literature. Acosta, Baerg and Mandelman (2009:2) identify the Dutch disease as an...upward pressure on the real exchange rate... that is caused by an influx of capital inflows. Such capital inflows may result from external sources of capital such as aid or from huge incomes realised from trade in natural resources (Acosta, Baerg & Mandelman, 2009:2). It is believed that an influx of remittances may result in real exchange rate appreciation which has a negative effect on export performance in the long run (Karagoz, 2009:1899; OECD, 2006:157; Amuendo-Dorantes & Pozo, 2004:1414; Catrinescu et al., 2009:81). Pradhan et al. (2008:498) further argue that low export performance resulting from real exchange appreciation can have some negative effects on economic growth of the remittance recipient economy. According to Catrinescu et al. (2009:81) low export performance resulting from real exchange rate appreciation caused by remittances has an adverse effect on output. Generally the adverse effects of low export performance on output tend to negatively impact employment. Such negative employment effects can result in negative growth effects in an economy Growth effects of remittances through extra demand It is not always the case that remittances will result in negative growth effects due to low export performance which has adverse effects on output. In some cases remittances generate an extra demand in the economy which can either have positive or negative growth effects (OECD, 2006:156). The negative growth effects may be as a result of the inflationary pressures arising from extra demand while positive growth effects happen through positive employment effects. The thesis of positive employment arising from extra demand generated 27