Remittances, financial development and economic growth in Africa

Remittances, financial development and economic growth in Africa

Journal of Economics and Business 64 (2012) 240–260 Contents lists available at SciVerse ScienceDirect Journal of Economics and Business Remittance...

696KB Sizes 133 Downloads 338 Views

Journal of Economics and Business 64 (2012) 240–260

Contents lists available at SciVerse ScienceDirect

Journal of Economics and Business

Remittances, financial development and economic growth in Africa Esman Morekwa Nyamongo a,∗, Roseline N. Misati b,1, Leonard Kipyegon b,2, Lydia Ndirangu a,3 a b

KSMS Research Centre, Central Bank of Kenya, P.O. Box 65041-00168, Nairobi, Kenya Research Department, Central Bank of Kenya, P.O. Box 60000–00200, Nairobi, Kenya

a r t i c l e

i n f o

Article history: Received 2 March 2011 Received in revised form 4 December 2011 Accepted 10 January 2012 Keywords: Remittances Financial development Economic growth Africa

a b s t r a c t This study investigates the role of remittances and financial development on economic growth in a panel of 36 countries in Africa over the period 1980–2009. It uses a panel econometrics framework and the main findings of the study are as follows: (1) Remittances appear to be an important source of growth for these countries in Africa during the period under study. (2) Volatility of remittances appears to have a negative effect on the growth of countries in Africa. (3) Remittances appear to be working as a complement to financial development. (4) However, importance of financial development in boosting economic growth appears weak, at least among the countries under study. © 2012 Elsevier Inc. All rights reserved.

1. Introduction Largely ignored in the past, workers’ remittances-transfers by international migrants to their countries of origin – have grown to become one of the largest sources of financial flows to developing countries, often overshadowing the traditional sources such as official aid and private capital flows (see World Bank, 2003, 2004; Aggarwal, Demirgüc¸-Kunt, & Martínez Pería, 2010; Giulia & Zazzaro, 2011; Giuliano & Ruiz-Arranz, 2009; Rao & Hassan, 2011). Further evidence shows that in 2010, worldwide remittance flows are estimated to have exceeded US $440 billion of which US $325 billion were ∗ Corresponding author. Tel.: +254 020 8646000; fax: +254 020 8018783. E-mail address: [email protected] (E.M. Nyamongo). 1 Tel.: +254 020 2863219. 2 Tel.: +254 020 2863212. 3 Tel.: +254 020 8646000. 0148-6195/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jeconbus.2012.01.001

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

241

transmitted to developing countries, an amount that far exceeded the volume of official aid flows and constituted more than 10 percent of gross domestic product (GDP) in many developing countries. It is also shown that recorded remittances in 2009 were nearly three times the amount of official aid and almost as large as foreign direct investment (FDI) flows to developing countries (World Bank, 2011). And in 2007 alone, over US$300 billion of workers’ remittances were transferred worldwide through official channels, and it is likely that billions more were transferred through unofficial channels (see Barajas, Chami, Fullenkamp, Gapen, & Montiel, 2009). As shown in the literature, remittances are transmitted through official and unofficial channels (see World Bank, 2011). As shown by Nyamongo and Misati, 2011; Aggarwal et al. (2010) remittances channeled through the formal avenues impacts growth of financial sector. This occurs especially when the recipients of such funds open accounts with commercial banks. In addition, when the recipients visit the banks they may gather information regarding existing bank loan products which they may take advantage of. If this effect on financial sector is substantial then we expect higher financial development. But as shown in the literature, financial development is also linked to private investment and economic growth (see Deodat, 2011; Mundaca, 2009; Misati & Nyamongo, 2010, 2011; Sufian & Siridopoulos, 2010). There is also a pool of evidence in the literature that find financial development to be important in driving economic growth. The earliest theoretical arguments in support of the role of financial development are found in Bagehot (1873), Schumpeter (1911) and more recently Hicks (1969). As shown by Schumpeter (1911) the services provided by the financial intermediaries are important for innovation and development. Schumpeter also showed that financial institutions may spur innovation and growth by identifying and funding productive investments. The same view was held by Hicks (1969) who traced the critical historical role played by the financial system in igniting industrialization in England by facilitating the mobilization of capital for “immense works.” In contrast, Joan Robinson (1952) declares that “where enterprise leads, finance follows.” According to this view, economic development creates demands for particular types of financial arrangements, and the financial system responds automatically to these demands. Further evidence (see Aggarwal et al., 2010) shows that where there is higher level of financial development remittances tend to have a lower marginal effect on growth. This is because financial development, as we know it, tend to be associated with producing information about possible investments and allocating capital; monitoring firms and exerting corporate governance; trading, diversification, and management of risk; mobilization and pooling of savings; and easing the exchange of goods and services. These financial functions tend to have a bearing on savings and investment decisions, and technological innovations and ultimately contribute to economic growth (see Misati, 2007; Misati & Nyamongo, 2011; Brown, 1994). However, on the other, a negative channel from remittances to growth also exists in the literature. Proponents of the negative effect of remittances on growth contend that, first, remittances occur in the context of information asymmetry, in which case the remitter lacks control of the usage of transferred funds by the recipient, thus, the recipient may not use the remitted funds for the investment projects or as productively as originally intended. Second, since remittances are largely transferred to households for consumption smoothening besides investment, the recipients may consider the remitted funds as a substitute for labor income and increase their leisure activities, which is likely to negatively affect labor productivity and growth. Third, while remittances enhance foreign exchange flows, the resultant appreciation of the exchange rate may erode the competitiveness for countries depended on the tradable sector (Amuedo-Dorantes & Pozo, 2004; Chami, Fullenkamp, & Jahjah, 2003) From the foregoing, it is obvious that there is a dearth of literature on African countries on this subject. Given the importance attached to remittances in African countries and in the light of ambiguity in terms of its impact on financial development and growth, it is important to examine its linkages to growth so as to facilitate effective policy oversight. This paper therefore seeks to contribute to the existing knowledge by making the following contributions: (1) Most studies in the literature tend to conduct panel studies on a global scale with African countries not being given due prominence. One of the reasons highlighted in these studies is lack of appropriate data for a number of countries in Africa. This particular study will be Africa specific and will therefore seek to demonstrate the role played by remittances on economic growth. (2) Literature covering economic growth within a panel framework is vast, in recent years this has been linked to the financial development. However, not much has been

242

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

done on the role of remittances in enhancing financial development to support economic growth in Africa. (3) In this study we consider the possibility of volatility of remittances impacting economic growth. This possibility has not be explored, at least, in the studies we have seen. This therefore is a novel contribution of this study to the existing pool of knowledge. (4) Unlike existing studies, the current study covers a long period from 1980 to 2009. The rest of the paper is organized as follows: in the next section we endeavor to characterize remittances, offer an account of evolution of remittances to Africa during the period 1980–2009. Then we examine data on remittances during the financial crisis. Section 2, provides literature review. In Section 3, we outline the methodology used in this study. Section 4 reports the results while Section 5 concludes. 1.1. Remittance types and channels Remittances are the cross-border earnings that migrants send to their countries of origin. These transfers may be channeled through official or unofficial channels. Official channels involve transfers using the banking system and money-transfer organizations. Unofficial channels involve sending remittances through mainly cash or in-kind transfers through carriers, such as family members, friends or other carriers; money or goods taken by the migrant on his/her seasonal visits to his/her homeland and; funds remitted through unlicensed money transfer operators using traditional networks such as hawala. According the World Bank (2011) more than 50 percent of the remittances to Sub-Saharan Africa is through the informal channels. Therefore this poses a major challenge in getting a near accurate estimate of the magnitude of remittances. Further evidence shows that even if all the remittances are channeled through the formal channels there are still issues. As shown by Barajas, Chami, Fullenkamp, and Garg (2010), the World Bank maintains one of the most reliable and often-used databases on remittances. The data considered as remittances comprises of three distinct categories of transfers. (1) Workers’ remittances category which records current transfers to nonresidents by migrants who are employed in, and considered a resident of, the countries that host them. (2) Employee compensation category is composed of wages, salaries, and other benefits earned by individuals in countries other than those in which they are residents for work performed for and paid for by residents of those countries. (3) Migrants’ transfers category comprises of contra-entries to the flow of goods and changes in financial items that arise from individuals’ change of residence from one country to another, such as movement of accumulated savings when a migrant returns permanently to the home country. As shown in the literature, informal money transfer systems are attractive to many immigrants because they are: (1) accessible as no bank account needs to be opened, and no complex bureaucratic procedures are needed, (2) anonymous as no proof of identity is required, (3) cheap as the cost of transactions is lower than the official channels, (4) swift and reliable as it is based on established informal networks of relatives and friends, people professing same religious beliefs. However, the informal channel has consequences: (1) it hinders valuable data collection by government on the nature and size of remittances, (2) increases the risk of misuse of remittances for money laundering and financing of illegal activities, among them terrorism. This may contravene the anti-money laundering and financing of terrorism legislations in place in many of the countries in the world, (3) diminishes the financial development impact of remittances as envisaged by the proponents of financial development. 1.2. Trend of remittances in Africa Fig. 1 shows the time profile of remittances to the countries in Africa during the period 1980–2009. From the figure it is evident that remittances to Africa during the early 1980s stood at approximately US $ 5.0 billion. This started to grow gradually at an average annual growth rate of 3.4 percent reaching approximately US $ 9.0 billion. The growth during this period may have been sustained by a 25.0 percent and 30 percent growth rates reported in 1983 and 1987, respectively. During the 1990s the remittances appear to have stabilised at a level slightly above US $ 10 billion, with an average annual growth rate of 5.1 percent. During the 2000s the remittances appear to have gained prominence, it stood at US $ 11.2 billion in 2000 rising rapidly at an average rate of 13.4 percent during this period

Remiances, US $ billions

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

243

50 40 30 20 10 0

Source: World Bank. World Development Indicators Fig. 1. Evolution of remittances in Africa: 1980–2009. Source: World Bank. World Development Indicators.

to stand at US $ 40.9 billion in 2008 before declining slightly by US $ 3.1 billion to stand at US $ 37.8 billion in 2009. The decline witnessed during this period may have been on account of the effects of the financial crisis which may have led to job losses for migrant workers thus occasioning reduced remittances to their mother countries. Overall, remittances to Africa grew at an average rate of 7.4 percent per annum during the period 1980–2009. Fig. 2 compares the behaviour of remittance to Africa and the rest of the world. It also shows remittance to Africa as a share of total remittances. It is evident from the figure that growth of remittances to Africa and the rest of the world are fragile. Further evidence from Fig. 2 shows that the share of remittances to Africa as share world remittances has been on a downward trend. It averaged 15.7 percent in the 1980s with the highest share of 18.7 percent reported in 1983. Coincidentally, this is the period when the growth of remittances to Africa was highest at 25.1 percent. In the 1990s the downward trend was sustained resulting in the average share declining to 11.4 percent, reaching a low of 7.4 percent in 2002. Since then the share appeared to recover but again declined in 2007. While Fig. 1 shows trend of the remittances to Africa in general, Fig. 4 shows the quantum of remittances to the top destinations in Africa. However, Fig. 3 tries to compare the remittances as a share of GDP. In so doing it will be possible to discern the countries that are heavily dependent on remittances. As shown in Fig. 3 it appears, on average, Lesotho is the highest recipient of remittances as a percentage of GDP at 34.2 percent of GDP followed by Cape Verde with a value of 14.5 percent. It is important to note that Lesotho is a small country surrounded in all sides by South Africa and that South Africa is accommodative to the citizens of the Southern African Development Community (SADC), where Lesotho is a member and therefore easier for its citizens to seek education and work 50

20 18

40 16 30

14 12

20

10 10

8 6

0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008

4

-10 2 -20

0 % change Africa

% change rest of World

Africa (%)

Source: World Bank. World Development Indicators Fig. 2. Review of performance of remittances in Africa. Source: World Bank. World Development Indicators.

244

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

Fig. 3. Top destinations – remittances as percent of GDP (1980–2009). Source: World Bank. World Development Indicators.

in South Africa. On the global scale (see World Bank, 2011), as a share of GDP smaller countries were the largest recipients of remittances in 2009: Tajikistan (35 percent), Tonga (28 percent), Lesotho (25 percent), Moldova (31 percent), and Nepal (23 percent). 1.3. Remittances to African countries during the financial crisis 2007–09

Remiances (US dollar millions)

The financial crisis that started in the summer of 2007 in the United States of America resulted in devastating economic and social effects on both developed, emerging and developing countries. This development has elicited policy and research responses in the recent past across the world. At the onset of the financial crisis there was a feeling that Sub-Saharan African countries would be insulated from the crisis as a result of the region’s low level of integration with the rest of the world. However, available evidence shows that these countries have witnessed reduction and changing structure of short-term capital flows and worsening of commodity prices (see Allen & Giovanneti, 2010), and declining equity returns (see Nyamongo & Misati, 2010). Further evidence has pointed out that these economies were partially insulated from the full effects of the financial crisis by resilient remittances. The reduction in remittance inflows during the financial crisis is traced to two factors: (1) reduced ability of the Diaspora to send money home, and (2) return migration as migrants lose jobs and are forced to return to their countries of origin. Until the financial crisis, remittances had proven to be a remarkably dependable source of foreign income for African countries. As shown in Table 1, remittance

12000.0 10000.0 8000.0 6000.0 4000.0 2000.0 0.0

Source: World Bank: African Development Indicators Fig. 4. Top remittance destinations in Africa, 2009. Source: World Bank. World Development Indicators.

Table 1 Remittances to Africa (US dollar, millions): 1980–2009. Crisis period 1985

1990

1995

2000

2005

2006

2007

2008

2009

2008–2009

406.0 77.0 77.3 150.4 29.4 40.1 0.0 1.6 3.2 32.2 0.0 2696.0 12.2 0.0 2.0 0.9 0.0 0.0 27.7 263.2 0.4 0.0 59.4 5.6 0.0 1053.9 53.4 0.0 11.3 22.0 3.4 0.8 77.1 0.0 0.1 67.1

313.0 40.3 31.3 125.9 12.2 20.8 0.0 4.2 0.6 24.0 0.0 3211.7 14.2 0.1 6.0 4.2 0.0 0.0 66.0 223.9 4.6 0.0 67.0 0.8 0.0 972.5 40.8 0.0 9.0 10.0 2.9 0.3 79.4 0.0 0.0 38.7

352.0 101.0 85.6 139.7 23.0 59.1 0.1 9.9 0.0 44.4 0.0 4283.5 5.2 0.9 10.0 6.0 18.0 1.0 139.3 427.9 7.9 0.0 106.9 13.7 0.0 2006.3 70.4 13.5 13.9 10.0 2.6 0.3 142.1 7.5 0.0 135.6

1120.0 100.1 59.2 78.2 11.2 105.9 0.0 12.2 4.2 151.1 11.9 3226.0 27.4 4.5 19.0 17.2 0.7 1.8 298.3 410.5 13.8 0.5 112.1 5.4 132.0 1969.5 59.1 15.6 7.7 804.0 21.1 0.0 146.0 0.5 23.7 105.3

790.0 87.1 26.2 67.3 30.2 86.9 0.0 12.0 10.4 119.3 12.3 2852.0 53.2 6.1 14.0 32.4 1.2 8.0 537.9 252.2 11.3 0.7 73.2 2.0 177.0 2161.0 36.8 9.5 14.4 1391.8 6.6 0.5 233.5 3.2 7.1 343.7

2060.0 172.7 131.0 56.6 77.2 136.6 0.0 12.0 11.4 163.2 25.8 5017.3 173.5 11.0 59.3 99.2 78.0 19.9 805.0 326.6 11.0 1.0 177.2 2.0 215.0 4589.5 56.6 18.1 66.4 3328.7 20.9 1.5 788.8 12.1 2.4 658.0

1610.0 224.0 116.5 67.9 129.9 136.6 0.0 12.0 13.2 166.8 28.5 5329.5 172.2 11.0 63.8 105.3 114.4 25.5 1128.0 361.5 11.0 1.0 211.8 2.0 215.0 5451.4 80.0 16.9 78.1 5435.0 21.2 1.6 925.2 13.3 15.6 734.1

2120.0 281.6 105.0 84.3 167.3 138.9 0.0 12.0 14.8 184.7 28.6 7655.8 357.8 11.0 69.8 117.4 150.7 43.0 1588.0 451.2 11.0 1.0 343.9 2.0 215.0 6730.5 99.4 16.2 79.4 9221.0 51.3 2.0 1191.8 11.2 42.0 833.6

2202.0 251.3 114.3 99.3 162.0 155.1 0.0 12.0 14.8 198.9 30.3 8694.0 386.7 11.0 65.0 126.1 82.2 49.5 1692.0 438.6 11.0 1.0 431.0 2.0 215.0 6895.4 115.7 13.9 93.7 9980.0 67.8 3.0 1476.1 7.8 27.5 822.8

2058.7 242.5 87.9 99.3 147.6 146.2 0.0 11.3 13.7 185.5 32.5 7149.6 261.6 10.3 79.8 114.5 63.7 46.7 1686.2 414.1 10.3 0.9 404.7 1.9 211.2 6269.5 111.1 13.6 89.1 9584.8 92.6 2.0 1364.7 12.5 46.7 902.3

−6.5 −3.5 −23.1 0.0 −8.9 −5.8 −5.5 −7.7 −6.8 7.1 −17.8 −32.3 −6.3 22.8 −9.2 −22.5 −5.6 −0.3 −5.6 −6.4 −12.6 −6.1 −5.9 −1.8 −9.1 −4.0 −2.4 −4.9 −4.0 36.6 −33.3 −7.5 60.6 69.7 9.7

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

Algeria Benin Botswana Burkina Faso Cameroon Cape Verde Central African Republic Comoros Congo, Rep. Cote d’lvoire Djibouti Egypt, Arab Rep. Ethiopia Gabon Gambia Ghana Guinea Guinea–Bissau Kenya Lesotho Madagascar Malawi Mali Mauritania Mauritius Morocco Mozambique Namibia Niger Nigeria Rwanda Sao Tome and Principe Senegal Seychelles Sierra Leone South Africa

% change

1980

245

246

Crisis period 1980 Sudan Swaziland Tanzania Tunisia Uganda Zambia Total

1985

1990

1995

2000

2005

2006

2007

% change 2008

2009

2008–2009

262.2 35.3 0.0 318.6 0.0 0.0

260.8 47.6 0.0 270.8 0.0 0.0

61.9 112.8 0.0 551.0 0.0 0.0

346.2 82.6 0.8 679.9 0.0 0.0

640.8 56.9 8.0 796.0 238.1 0.0

1016.1 95.4 19.4 1392.7 321.8 52.9

1179.1 98.6 15.4 1510.0 411.0 57.7

1769.2 100.5 14.3 1715.8 451.6 59.3

3100.5 89.6 18.6 1977.0 723.5 68.2

2992.7 93.5 23.3 1964.5 749.7 41.3

−3.5 4.3 24.9 −0.6 3.6 −39.5

5789.8

5903.6

8963.1

10185.4

11210.6

22283.9

26301.6

36543.7

40926.3

37834.3

−7.6

Source: World Bank: African Development Indicators.

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

Table 1 (Continued)

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

247

to Africa was on an increasing trend standing at US $ 5.8 billion in 1980 rising to US $ 26.3 billion in 2006, just before the global financial crisis set in. At the onset of the financial crisis in 2007, remittance to Africa totaled US $36.5 billion and grew by 12.1 percent to stand at US $ 40.9 billion in 2008. The effect of the crisis on remittances inflows was felt in 2009 when the total remittances to Africa declined by 7.6 percent to stand at US $37.8 billion. As shown in Fig. 4, the major recipients of remittances in 2009 were the countries in North Africa. The bulk of the remittances went to the Arab countries in North Africa, namely Egypt (US dollar 7.1 billion), Morocco (US dollars 6.3 billion), Algeria (US dollars 2.1 billion) and Tunisia (US dollar 2.0 billion). The larger share of inflows of remittance to the North African countries may be related to the large number of nationals from these countries working in Europe and the oil rich countries in the Middle East. Similarity of religion and language makes it easier for citizens of North African countries to migrate to the oil rich countries compared with citizens of the other parts of the continent. In addition, the geographical location of these countries to Europe explains why they have a larger share of remittances compared with other countries in Africa. Available evidence shows that EU destination countries of migration from North Africa are: France, Germany, Belgium, the Netherlands, Spain and Italy. Among the Sub-Saharan African countries Nigeria was the leading recipient with US dollars 9.6 billion in 2009. Other leading destinations in the region were Sudan (US dollars 3.0 billion) and Kenya (US dollars 1.7 billion). Fig. 4 shows that over the period under review, Nigeria emerged as the highest recipient of remittances in Sub-Saharan Africa. On average, Nigeria received US$3.23 billion annually. This trend is a reflection of high levels of emigration in Nigeria. Nigeria has, over the years, experienced considerable emigration mainly due to the large size of its population and also due to political upheavals experienced in the country. It is important to note that Nigeria is the most populated country in Africa and therefore faces higher emigration as evidenced by its being rated as having the biggest African Diaspora. The history of politically influenced emigration in Nigeria dates back to 1965 when deteriorating political environment and electoral crisis caused many people to leave Nigeria. Between 1967 and 1970, the Biafra war also forced more Nigerians to emigrate. Following the Biafra war, Nigeria was ruled by a military dictatorship, thereby causing even more people to seek refuge outside Nigeria. Such an enormous number of emigrants, accounts for large amounts of remittances to Nigeria (Montclos 2005). As shown in Table 1, the financial crisis elicited subdued remittances flows to Africa region. As shown earlier the North African countries are the major remittance receiving countries and the financial crisis did not spare this region. All the countries in this region registered declining remittances inflows in 2009. Egypt recorded the highest declining during this period at 17.8 percent followed by Morocco at 9.1 percent and Algeria at 6.5 percent. Among the Sub-Saharan African region the worst affected countries during this period are: Zambia (39.5 percent), Sao Tome and Principe (33.3 percent), Ethiopia (32.3 percent) and Guinea (22.5 percent). However, it is worth noting that the major remittances receiving countries in Sub-Saharan Africa suffered modest declines in remittances: Nigeria (4.0 percent), Sudan (3.5 percent), Kenya (0.3 percent) and Senegal (7.5 percent). However, some countries in Africa show good growth performance of remittances in 2009: Seychelles (60.6 percent), Sierra Leone (69.9 percent), Rwanda (36.6 percent), Tanzania (24.9 percent) and Gambia (22.8 percent). However, these countries are not among the leading remittances destinations in Africa. The growth of remittances to these countries during this period may be related to the fact that most of these countries do not largely depend on remittances from Europe and North America as it is the cases with other countries. The migrants would be working in neighboring countries where the full effects of the financial crisis were not witnessed.

2. Impact of remittances: what literature says? A number of studies have investigated the effect of remittances on a range of socio-economic outcomes on the receiving countries. However, this seems to be a debate that is far from concluded. The following are the key issues captured in the literature.

248

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

• As shown in the literature remittances (see Adams, 2004; Adams & Page, 2003, 2005; Gupta, Pattillo, & Wagh, 2007) are geared to supporting the welfare of those relatives left behind and therefore contributes to the poverty eradication of the recipient. However, as shown by Carling (2004), poorer and lower-skilled households may get marginal benefits from remittances as a result of prohibitive costs of migration and the stringent immigration policies in advanced economies which tend to favor skilled workers. As a result, it is argued remittances may improve the per capita income of the recipient country but may worsen income inequality in the receiving country. This may result in urban–rural inequality, as remittances are mainly used to finance investments in urban areas. In addition, some authors (see Chami et al., 2003), argue that remittances may reduce recipients’ motivation to work, creating permanent financial dependency, and slowing down economic growth. • Remittances represent a source of savings and capital for investment in education (see Yang, 2008) health (Anton, 2010) and entrepreneurship (see Yang, 2004; Woodruff & Zenteno, 2007; Woodruff, 2007; Massey & Parrado, 1998), all of which have an effect on productivity, employment and ultimately on economic growth and development. • As shown by Aggarwal et al. (2010), remittances may influence the growth of financial development in recipient countries. This is premised on the notion that some remittance recipients may be persuaded to convert their remittances into deposits with financial institutions when these remittances are channeled through the banking system. This may result in more funds becoming available for lending by commercial banks to the private sector. If this argument is valid, financial development will in turn enhance growth as shown in the literature (see Misati & Nyamongo, 2011). • Remittances may improve a country’s credit worthiness for external borrowing. This is because remittances are included in the exports of goods and services. Higher remittances will therefore improve the country’s debt to exports ratio, an indicator that is critical in the assessment of a country’s credit worthiness. • As shown by Chami, Fullenkamp, and Gapen, M. (2009) and Ratha (2003), remittances tend to rise in times of economic downturns thus smoothing household consumption. This phenomenon has been observed in a number of countries during the just ended financial crisis period. • Some countries, as shown by Ketkar and Ratha (2001), take advantage of the stable time profile of remittances and have come to accept remittances as collateral against which both public and private sector entities may borrow in international capital markets longer dated debts at relatively lower rates. Available evidence shows that the first major securitization deal involving workersˇı remittances occurred in 1994 in Mexico. Further evidence shows that since then, other countries have followed suit. The largest issuers of remittance backed securities nowadays are Turkey which accounts for 35 percent of total remittance backed securities, Brazil at 31 percent and Mexico at 24 percent. • Remittances enhance foreign exchange inflows in a country. Large inflows of foreign exchange into small open economies, such as those in Africa, may lead to exchange rate appreciation and lower export competitiveness. However, as shown by IMF (2005), since remittances tend have a stable time profile, the “Dutch disease” effects of remittances concerns may not be credible. 3. Methodology 3.1. Estimation strategy Here we explain the estimation strategy used in this paper. As a starting point we formulate the standard growth model in a manner consistent with Barro (1989, 1991) as: yit = (ˇ1 − 1)Yt−1 + ˇ2 Xi,t + ˇt + i + εi,t

(1a)

where y is the growth rate of real GDP per capita, Y is the real GDP per capita; X is a vector of variables found in standard growth models including: the ratio of gross investment to GDP; inflation rate, human capital formation; ratio of government consumption to GDP and openness variable (sum of exports and imports to GDP), εij is the idiosyncratic error term. While ˇt and i are time and country specific effects, respectively. In the above formulation ␤1 − 1 is the convergence coefficient. As shown in the literature, many studies augment Eq. (1a) with a host of other factors that may influence economic performance.

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

249

In this study we hypothesise that remittances and financial development will play important roles in enhancing economic growth performance. Therefore we extend Eq. (1a) to take the form: yit = (ˇ1 − 1)Yt−1 + ˇ2 Xi,t + ˇ3 REMY + ˇ4 REMVit + ˇ5 FDit + ˇ6 (REMY.FD)it + ˇt + i + εi,t (1b) where REMY is amount of remittances received from abroad (normalized by GDP); REMV a measure of volatility of remittances; FD is a set of financial development indicators, and REMY.FD is an interaction variable. As shown in Equation 1b remittances are critical to economic growth performance. Here we hypothesize that higher level of remittances will impact on economic growth. This follows the work of Giuliano and Ruiz-Arranz (2009) and Rao and Hassan (2011). Further motivation is found in Aggarwal et al. (2010) where it is shown that higher financial development tend to be associated with higher remittances. Following this line of thinking therefore, it may be inferred that higher remittances will occasion higher financial development which will in turn positively impact the level of economic growth. In this study we include a variable to capture volatility of remittances in the augmented growth model. Earlier we have shown the link between remittances and economic growth using the financial sector link. However, what has not been shown is what happens if there is volatility of these flows to the recipient countries, as it did during the financial crisis of 2007–2009. We hypothesise that volatility in remittances will dampen the growth performance of the recipient country. The argument here is as follows: assuming all the remittances are channeled to the recipient country through the banking sector. This therefore becomes a dependable source of funds that will boost financial development and therefore economic growth. Suppose there is a negative shock in the flow of remittances in a particular year, what will happen? We argue that this will lead to a substantial decline in the growth of credit to both the private and the public sector. Consequently, investment will be constrained and therefore economic growth will be injured. Therefore we hypothesize here that high volatility will negatively impact economic growth. In this study two indicators that measure financial development are used, namely ratio of credit to the private sector to GDP (DC) as a measure of financial depth and the ratio of broad money supply to GDP (M2) which is a monetization variable. The ratio of broad money to GDP (M2) is the most commonly used measure of financial development (see Calderon & Liu, 2003; King & Levine, 1993). A higher ratio of M2 to GDP indicates a larger financial sector and bigger financial intermediation. This ratio shows the real size of the financial sector of the country. If the financial sector grows faster than the real sector, this ratio increases over time. The other indicator of financial development is the ratio of credit extended to the private sector to GDP which represents the actual resources that are channeled to the private sector. In Eq. (1b) the interaction term (REMY.FD) is incorporated. This variable serves to show the role of remittances on economic growth using the financial sector transmission mechanism. The inclusion of the interaction term in this equation is based on the debate in the literature on whether these two variables are complements or substitutes. The proponents of the substitutability hypothesis argue that remittances relaxes the lack of financial development condition in emigration countries, by allowing poor people to invest in high return projects despite their difficulties to obtain credit (see Calderón, Fajnzylber, & López, 2007; Giuliano & Ruiz-Arranz, 2009). On the other hand, the complementarity hypothesis is built on the notion that remittances and financial development support one another (see Aggarwal et al., 2010 and Martínez, Soledad, Mascaró, & Moizeszowicz, 2007). Here it is shown that a higher degree of financial development allows migrants to send money home cheaply, faster and safely. If remittances are transmitted in large amounts they may stimulate the interest of financial institutions and public authorities, bringing about higher levels of competition between financial institutions, as well as institutional reforms with a view to channeling remittances towards productive investment. Human capital formation is widely acknowledged in the literature as critical to the growth process with countries registering higher human capital formation expected to post higher growth rates. In the literature human capital formation is proxied by school enrollment rates. Some studies use secondary school gross enrolment rate while others use the primary school gross enrolment rate (see Ajala & Kerebih, 2008 and Qaisar, 2001). In both cases, such proxies have yielded positive coefficients in

250

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

support of theory. In this study we use the primary school gross enrolment rate as data on this variable is relatively available compared to the secondary school gross enrolment rate. We therefore expect gross primary school enrollment to positively influence growth through its effect on productivity (see Barro, 1989). In the literature there is consensus about the positive role of capital accumulation in growth. There is also consensus that inclusion of this variable in the growth model will create endogeneity problem which may be addressed via instrumental variable estimation techniques. In this study therefore, we use investment to capture the role of capital accumulation. Inflation is included to capture the effect of macroeconomic stability. Inflation which measures the degree of uncertainty about the future market environment is expected to have a negative relationship with economic growth. In an uncertain market environment firms become reluctant to make longrun commitments in the presence of higher price variability (Caporale, Rault, Sova, & Sova, 2009). This is especially true if the high inflation is also associated with increased price variability. The effect is a more conservative investment strategies than would otherwise be the case, ultimately leading to lower levels of investment and economic growth. Total government expenditures as a share of GDP is expected to correlate ambiguously with growth depending on the type of government expenditure and whether it crowds in or crowds out private investment. Government expenditure engaged in the provision of non-rival and non-excludable public services to the economy is complementary to private investment and thus is expected to positively impact economic growth. It is also possible that high resource consumption by the public sector would undermine the efficiency of resource allocation or crowd out resource availability to the more efficient sectors of the economy. As shown in the vast literature, government consumption may have positive or negative impact on economic growth. Negative effect stems from the notion that large governments tends to crowd out the private sector while the positive effect is founded on the reasoning that higher government expenditure may create an enabling environment that may spur economic growth However, in Africa where government consumption is mainly wasteful we expect higher government consumption to impact economic growth negatively.

3.2. Data sources and type Here we describe the data used in this study. The full sample consists of 36 countries from the African continent, the basic data covering 1980–2009 used in this study is obtained from various sources. The basic data is manipulated to be consistent with that used in other studies. As shown in Giuliano and Ruiz-Arranz (2009), studies do not use annual observations directly. Instead, averaging of the observations is done to account for business cycle fluctuations. In this regard, we split the sample period 1980–2009 into 10 non-overlapping 3-year periods. While most studies rely on 5 year averages, this study settled on 3-year averages in order to avoid losing substantial number of countries used in the study. In this regard, all countries with data points being less than 3 for any key variable, after averaging, were dropped from the study. The full description of the data is as follows: The dependent variable is the real GDP per capita growth. This variable is obtained directly from the World Bank: African Development indicators. In addition the real GDP per capita used here is in US dollars (2000 prices). Data on remittances as a ratio of GDP is obtained from the World Bank: African development indicators. In terms of definition, remittances data used here comprise of the following distinct categories of transfers. (1) Workers’ remittances category which records current transfers to nonresidents by migrants who are employed in, and considered a resident of, the countries that host them. (2) Employee compensation category is composed of wages, salaries, and other benefits earned by individuals in countries other than those in which they are residents for work performed for and paid for by residents of those countries. (3) Migrants’ transfers category comprises of contra-entries to the flow of goods and changes in financial items that arise from individuals’ change of residence from one country to another, such as movement of accumulated savings when a migrant returns permanently to the home country. The ratio of remittances to GDP ratio is then averaged over a period of three years. The volatility of remittances is computed from the ratio of remittances to GDP. The proxy for volatility is the standard deviation of this ratio over a period of three years.

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

251

In this study we use 2 variables to proxy financial development. All these measures are obtained from the World Bank: African development indicators. These measures include the broad money supply as a ratio of GDP. This is equal to currency plus demand and interest bearing liabilities of banks and non-financial intermediaries divided by GDP. By using this measure we are able to account for the actions of three financial institutions namely, the central bank, deposit money banks and other financial institutions. Also, credit provided by the banking sector to GDP, which reveals the extent of financial intermediation by the banking system, including credit to the public and private sectors. In addition, inflation, measured as the annual percentage change in the consumption price index. Openness to international trade, defined as the ratio of the sum of exports plus imports of goods to total output. Human capital formation measured as the gross primary enrolment rate, government consumption to GDP ratio are also obtained from the World bank. However, we average these ratios over a period of three non-overlapping years. 4. Empirical results The descriptive statistics of the variables used in this study is reported in Table 2. The descriptive statistics of the variables used in this study is reported in Table 2 while estimation results are reported in Tables 3–5. In these estimations we employ the following strategy: (1) we estimate the pooled, fixed effects and the random effects models using both the ordinary least squares and the two stage least squares. This is based on the evidence in the literature (see Chami et al., 2003; Catrinescu, Leon-Ledesma, Piracha, & Quillin, 2008); and Faini (2007) where it is shown that the possibility of causality between remittances and growth may affect the reliability of the estimated coefficients. The possibility of reverse causality suspected here emanates from the fact that remittances could affect growth of recipient countries and therefore impact on the amount of future remittances received. Therefore estimation of the model implemented in this study using ordinary least squares would lead to endogeneity bias. To address this problem, we use the Two-Stage Least Squares (TSLS) instrumental variables method and try to find variables that are highly correlated with the endogenous variable, but not related to the error term. This raises further issues, how do we find the appropriate instruments? The literature has emphasized the difficulty of finding appropriate instruments for remittances that is not subject to reverse causality and the weakness of instruments used by some authors. Some instruments have appeared in the literature but have faced criticism, for example, a country’s legal systems and creditor rights (La Porta et al., 1997). Rajan and Subramanian (2005) use the distance from the country of origin as an instrument for remittances. These variables suffer from the drawback that they do not vary over time, Chami et al. (2003) use differences in income and interest rates between migrants’ home and host countries where host countries are represented by the proxy USA; while Rajan and Subramanian (2005) and Faini (2007) uses the distance between migrants’ home and host countries. However, these instruments suffer from the drawback that they do not fully reflect the flows of remittances. For instance, distance between home and host countries do not vary over time, so it is not possible to use it in a panel framework. For this reasons, lagged values of the exogenous variables often act as instruments. In addition, we test for the validity of the estimated models. The F-test is used to test for the validity of the fixed effects model relative to the restricted/pooled model. In addition the validity of the random effects model is conducted using the Hausman-test. Both the F-statistic and the Hausman tests, where appropriate, are reported in Tables 3–5. F-test allows us to reject the null hypothesis of homogeneity at 1% level and conclude to the presence of individual specific factors that drive growth. This therefore suggests that the best models applicable here are either fixed-effect or random-effect. The slope coefficients are reported in the tables provided, however, the intercept coefficients/fixed effects are not reported. However, it is important to explain the important observations from the fixed effects models, where the estimated country specific affects point out whether there are country specific factor that enhance or inhibit economic growth during the period under study. Among the countries that have conditions that inhibit growth include, Burundi, Rwanda, Cote D’Ivoire, Kenya, Malawi, Zambia and Togo. In the case of Burundi and Rwanda, during the period under study these countries endured political and social problems leading to the Genocide that occurred in Rwanda in

252

YPCG Mean Median Maximum Minimum Std. Dev. Skewness Kurtosis Jarque-Bera Probability Observations Cross sections

1.37 1.41 16.99 −11.23 3.14 −0.02 5.74 87.04 0.00 279 36

YPCR 1016.41 411.94 8413.74 108.50 1325.90 2.58 11.00 1053.32 0.00 279 36

REMY

REMV

4.77 1.72 85.35 0.01 10.74 5.36 35.30 13459.44 0.00 279 36

0.63 0.24 13.40 0.00 1.36 5.86 45.21 22305.96 0.00 279 36

Gl 21.99 20.85 62.93 3.92 8.59 1.42 7.14 293.27 0.00 279 36

INF

PRI

GOV

TR

M2

DC

10.71 7.85 111.47 −8.24 13.07 3.78 22.79 5219.69 0.00 279 36

85.34 91.84 152.90 16.51 28.64 −0.30 2.57 6.36 0.04 279 36

15.32 14.98 36.60 4.65 5.39 0.90 4.27 56.87 0.00 279 36

73.39 63.40 289.29 11.34 39.06 1.33 6.03 189.31 0.00 279 36

32.78 25.51 103.38 7.31 20.46 1.49 4.69 136.61 0.00 279 36

24.41 16.58 143.57 1.94 22.98 2.30 9.40 721.32 0.00 279 36

Where YPCG is the growth rate of per capita GDP, YPCR is the real per capita GDP, REMY is the ratio of remittances to GDP, REMV is the standard deviation of remittances to GDP; GI is the ratio of gross investment to GDP, INF is the inflation rate, PRI is the gross primary enrolment rate, GOV is the ratio of government consumption to GDP, TR is the ratio of total trade to GDP, M2 is the ratio of m2 to GDP and DC is the ratio of domestic credit to GDP.

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

Table 2 Descriptive statistics.

OLS

YPC(−1) GI INF PRI GOV TR F-statistics Hausman test (Chi-sq statistic) R2 Number of observations

2SLS

PM (1)

FEM (2)

REM (3)

PM (4)

FEM (5)

REM (6)

−0.183* (−1.832) 1.436*** (2.726) −0.009 (−0.893) 1.573*** (2.698) −1.217** (−1.216) −0.089 (−0.194) – – 0.08 298

−3.872*** (−3.656) 0.060 (0.099) −0.019* (−1.881) 4.036*** (4.896) −2.773*** (−3.874) 2.300*** (2.824) 4.167 – 0.41 298

−0.450* (−1.763) 0.688 (1.299) −0.017* (−1.798) 2.539*** (3.961) −1.947*** (−3.227) 0.693 (1.276) – 40.692 0.11 298

0.213* (−1.887) 1.110*** (2.879) −0.019*** (−2.684) 1.674*** (3.778) −1.524*** (−3.478) 0.291 (0.877) – – 0.16 298

−3.283*** (−4.507) 0.986*** (2.279) −0.034*** (−5.298) 3.765*** (6.373) −2.065*** (4.133) 2.913*** (5.722) 8.055 [35,256] – 0.60 298

−1.883* (−1.891) 1.436*** (2.918) −0.009 (−0.956) 1.573*** (2.888) −1.217** (−2.232) −0.089 (−0.208) – 49.415 0.08 298

t-Values in parentheses. PM is the pooled model, FEM is the fixed effects model, REM is the random effects model, YPCG is the growth rate of per capita GDP, YPCR is the real per capita GDP, REMY is the ratio of remittances to GDP, REMV is the standard deviation of remittances to GDP; GI is the ratio of gross investment to GDP, INF is the inflation rate, PRI is the gross primary enrolment rate, GOV is the ratio of government consumption to GDP, TR is the ratio of total trade to GDP. * Significant at 10 percent. ** ***

Significant at 5 percent. Significant at 1 percent.

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

Table 3 Baseline model: dependent variable: growth of GPG per capita.

253

254

Table 4 Dependent variable: Growth of GPG per capita. OLS PM (1)

Number of observations

−0.163* (−1.764) 2.410*** (4.214) 0.009 (0.661) 1.613*** (2.789) −0.744 (−1.265) 0.683 (1.507) 0.291** (2.549) – – – 0.16 271

PM (2) −0.289* (−1.829) 2.757*** (4.664) 0.018 (1.265) 1.640*** (2.799) −0.488 (−0.822) −0.529 (−1.168) 0.320** (2.442) −0.336* (−1.778) – – 0.18

FEM (3) −3.569*** (−2.775) 1.706** (2.401) 0.001 (0.086) 4.051*** (4.353) −1.998*** (−2.366) 1.642* (1.615) 0.181 (0.575) – 2.639 – 0.40

FEM (4) −3.588*** (−2.739) 1.480** (2.002) −0.009 (−0.479) 4.432*** (4.573) −1.768** (2.061) 1.037 (1.114) 0.561* (1.653) −0.265 (−1.325) 2.555 – 0.41

271

271

271

PM (8)

FEM (9)

FEM (10)

REM (5) −0.398* (−1.609) 2.248*** (3.794) 0.003 (0.231) 2.259*** (3.479) −1.186* 0.280 (0.514) 0.334** (2.459) – – 24.518 0.14 271

REM (6) −0.427* (−1.717) 2.439*** (4.069) 0.009 (0.591) 2.240*** (3.539) −0.807 (−1.314) −0.348 (−0.678) 0.369** (2.589) −0.282 (−1.568) – 32.604 0.16 271

2SLS PM (7) YPC(−1) GI INF PRI GOV TR REMY REMV F-statistics Hausman test (Chi-sq statistic) R2 Number of observations

−0.201* (−1.961) 2.507*** (5.543) 0.001 (0.128) 1.525*** (3.660) −0.880** (−2.120) 0.391 (2.209) 0.158* (1.843) – – – 0.26 271

−0.327** (−2.094) 2.627*** (5.815) 0.008 (0.698) 1.675*** (4.192) −0.648* (−1.633) −0.318 (−1.036) 0.226** (2.571) −0.31*** (−2.723) – – 0.28 271

−3.416*** (−3.822) 0.543 (0.983) −0.018 (−1.315) 3.433*** (6.345) −0.923** (−2.067) 2.239*** (3.293) −0.056 (−0.259) – 5.429 – 0.66 271

−3.997*** (−4.221) 0.379 (0.673) −0.025* (−1.744) 3.885*** (5.711) −0.849** (−2.419) 2.028*** (2.957) 0.155 (0.701) −0.181 (−1.422) 5.101 – 0.77 271

REM (11) −0.162* (−1.837) 2.410*** (4.623) 0.009 (0.725) 1.612*** (3.060) −0.744 (−1.388) 0.683* (1.653) 0.291*** (2.797) – – 61.127 0.16 271

REM (12) −0.289* (−1.845) 2.757*** (5.068) 0.018 (1.375) 1.639*** (3.041) −0.488 (−0.893) −0.529 (−1.269) 0.320*** (2.654) −0.336* (−1.931) – 53.804 0.18 271

t-Values in parentheses. PM is the pooled model, FEM is the fixed effects model, REM is the random effects model, YPCG is the growth rate of per capita GDP, YPCR is the real per capita GDP, REMY is the ratio of remittances to GDP, REMV is the standard deviation of remittances to GDP; GI is the ratio of gross investment to GDP, INF is the inflation rate, PRI is the gross primary enrolment rate, GOV is the ratio of government consumption to GDP, TR is the ratio of total trade to GDP, M2 is the ratio of m2 to GDP and DC is the ratio of domestic credit to GDP. * Significant at 10 percent. **

Significant at 5 percent.

***

Significant at 1 percent.

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

YPC(−1) GI INF PRI GOV TR REMY REMV F-statistics Hausman test (Chi-sq statistic) R2

Table 5 Dependent variable: growth of GPG per capita. OLS – fixed effects (1)

Number of observations

(2)

−0.233* (−1.929) 3.059*** (6.011) −0.006 (−0.668) 1.156** (2.060) −1.416** (−2.494) −0.403 (−0.919) – – 0.122 (0.471) – – – 0.16 271

(3)

−0.298* (−1.696) 2.632*** (5.833) 0.007 (0.642) 1.686*** (4.205) −0.606 (−1.459) −0.347 (−1.107) 0.231** (2.574) −0.310*** (−2.709) −0.060 (−0.310) – – 0.28 271

(4)

−0.386** (−2.121) 2.514*** (6.251) 0.004 (0.401) 1.401*** (3.431) −0.620 (−1.502) −0.332 (−1.086) 0.333* (−1.727) −0.285** (−2.583) 0.065 (0.334) 0.223*** (2.802) – – 0.30 271

(5)

−0.221* (−1.986) 2.998*** (5.788) −0.008 (−0.861) 1.083* (1.939) −1.419** (−2.567) −0.365 (−0.838) – – – – 0.229 (0.601) – 0.16 271

−0.123* (−1.901) 2.876*** (4.860) 0.012 (0.806) 1.684*** (2.847) −0.453 (−0.760) −0.615 (−1.334) 0.366*** (2.652) −0.326* (−1.709) – – −0.557 (−1.261) – 0.18 271

2SLS – fixed effects (6) PM YPC(−1) GI INF PRI GOV TR REMY REMV DC DC* REMW M2 F-statistics R2 Number of observations

(7)

−3.035 (−3.697) 1.000* (1.895) −0.027*** (−4.476) 3.322*** (5.522) −2.110*** (−3.833) 1.677*** (2.932) – – 0.084 (0.261) – – 5.674 0.58 ***

271

(8)

−3.30 (−3.482) 0.415 (0.728) −0.027* (−1.794) 3.906*** (5.732) −0.954** (−2.058) 2.069*** (2.987) 0.133 (0.581) −0.176 (−1.363) 0.092 (0.343) – – ***

0.66 271

(9)

−3.670 (−3.601) 0.572 (1.034) −0.031** (−2.233) 2.886*** (5.618) −0.949** (−2.469) 2.131*** (3.363) 1.078*** (2.682) −0.243* (−1.872) 0.134 (0.547) 0.498*** 3.457) – 5.076 0.74 ***

271

(10)

−2.706 (−3.414) 0.923*** (1.786) −0.029*** (−4.750) 3.321*** (5.444) −2.158*** (−4.099) 1.939*** (3.384) – – – – 0.489 (0.793) 5.696 0.59 ***

271

−3.527*** (−3.345) 0.485 (0.853) −0.032** (2.202) 3.786*** (5.484) −0.947** (−2.449) 2.171*** (3.162) 0.135 (0.612) −0.150 (−1.165) – – 0.493 (0.820) 5.127 0.73

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

YPC(−1) GI INF PRI GOV TR REMY REMV DC DC* REMV M2 F-statistics R2

271

255

t-Values in parentheses. PM is the pooled model, FEM is the fixed effects model, REM is the random effects model, YPCG is the growth rate of per capita GDP, YPCR is the real per capita GDP, REMY is the ratio of remittances to GDP, REMV is the standard deviation of remittances to GDP; GI is the ratio of gross investment to GDP, INF is the inflation rate, PRI is the gross primary enrolment rate, GOV is the ratio of government consumption to GDP, TR is the ratio of total trade to GDP, M2 is the ratio of m2 to GDP and DC is the ratio of domestic credit to GDP. * Significant at 10 percent. ** Significant at 5 percent. *** Significant at 1 percent.

256

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

1992 in which approximately one million people were killed. Although stability has been restored the main challenge that may have faced these countries after the genocide was the resources constraints to rebuild these countries. Cote D’Ivoire and Kenya also, reported negative country specific effects across most of the estimations, Cote D’Ivoire, during the period under study endured two major events which may have occasioned poor performance during the period. In the 1980s–1990s the commodity price of its foreign exchange earning crop – Cocoa declined in the international market. This development may have impacted the growth potential of Cote D’Ivoire during the time. Just when the prices were recovering in the early 2000s, the country witnessed political developments resulting in the emergence of rebels fighting the government and took control of the northern half of the country. This development may have impacted on government programmes in the affected areas and also the investors may have fled the country. The notable case is the relocation of the African Development Bank (ADB) from Cote D’Ivoire to Tunisia during this period as a result of the political tensions. Kenya, also witnessed negative country specific effect during the period under study. This may be as a result of the following developments: (1) During the 1980s the commodity prices for the Kenya’s main export crop – coffee had declined in the international markets and also the poor economic management of the economy during this time resulted in poor growth performance (2) in the 1990s the country embraced economic liberalisation – price and exchange controls were removed with poor safety nets to cushion the poor from the negative effects of such development. In addition, due to poor policies of the government Kenya faced foreign aid freeze leading to near collapse of many sectors of the Kenyan economy. While the multi-party politics had been introduced in the country in the 1990s, the period prior to elections: 1992, 1997, 2002 and 2007 witnessed heightened political tensions that was not pleasing to both the domestic and foreign investors. The notable event was the post election violence which erupted in early 2008 which led to disruption of economic activities. This also coincided with the international financial crisis of the 1997–2009. In the sample, however, there are countries which show positive coefficients in most of the estimations. These countries include: Namibia, Botswana, South Africa and Sudan. In the case of Botswana, this is a country that is endowed with natural resources and is the best performer in a range of governance matters. For example, according the Corruption perceptions index (CPI) published by the transparency international, Botswana is the least corrupt country in Africa. In terms of all the World Bank’s Governance indicators (Voice and Accountability, Political Stability and Absence of Violence, Government Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption) Botswana is the best performer in the continent. Namibia is mineral rich and ranks second to Botswana in terms of the quality of institutions. These conditions may have made these countries to post positive country specific effects. South Africa is mineral rich; it has large deposits of platinum and gold. The exports of these minerals may have enhanced its growth performance during the period under study. In addition, since 1994 when the apartheid regime was replaced with a democratically elected government the country took advantage of the international goodwill to advance its growth agenda by focusing on the international markets. Sudan as well posts positive fixed effects in most of the specifications. Although Sudan was embroiled in the civil war involving the Southern and northern parts of the country during the period under study, large deposits of petroleum was discovered in the country during this same period and appear to have boosted the growth performance of this country. To investigate the role of remittances, financial development and economic growth we present a range of results. We follow the approach of first estimating the growth model following the standard variables as shown in Table 3 then financial development indicators and remittances are included in Tables 4 and 5. The role of remittances is shown in the various estimations in Table 4. Adding remittances to GDP in the pooled model (columns 1 and 7) shows that the estimated coefficients are positive and significant at the conventional levels of testing. However, the fixed effects models (columns 3 and 9) yield insignificant results – in the OLS estimation the estimated coefficient is found to be positive but insignificant at the conventional levels of testing while the 2SLS method yields a negative and insignificant coefficient. Further estimation using the random effects model (columns 5 and 11) yield coefficients that are positive and significant at the conventional levels of testing. Including the volatility

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

257

variable in the specifications (columns 2, 4 and 6, 8, 10 and 12) leaves the estimated coefficients of the remittances to GDP largely positive and significant at the conventional levels of testing. Further evidence of the importance of remittances to economic growth are shown in Table 5, where the financial development indicators are introduced into the model, and it is found that the estimated coefficients are largely positive and significant at the conventional levels of testing. This evidence therefore suggest that remittances are important in explain economic growth in Africa. The role of volatility of remittances is shown in Table 4 as well. As shown in the pooled model (columns 2 and 8) the estimated coefficient is negative and significant at the conventional levels of testing. The fixed effect model (columns 4 and 10) also produce negative estimated coefficients but marginally significant in the OLS estimation. The random effects model (columns 6 and 12) also yield results that are similar to those obtained in the fixed effects model. Further evidence of the role of volatility of remittances is shown in Table 5 where it is found that the estimated coefficients are consistently negative but not unanimously significant at the conventional levels of significance. In this study we had hypothesised that volatility in remittances will dampen the growth performance of the recipient country. The evidence here tends to support the view that there is a negative relationship between volatility of remittances and economic growth. However, this evidence is not strong as some of the estimated coefficients are not found to be significant at the conventional levels of testing. The role of financial development is shown in Table 5 where the fixed effects model results are shown. Here results shown in columns 1 and 6 show domestic credit as ratio of GDP included in the estimation. In both cases it is found that the estimated coefficients of domestic credit to private sector as a ratio of GDP are positive but not significant at the conventional levels of testing. Using a different measure of financial development, m2 as a ratio of GDP also yield positive but insignificant results. This point to the weakening role of financial development in economic growth among the countries in our sample. Further analysis is done by incorporating remittances as a ratio of GDP and volatility of remittances in the estimated models (columns 2 and 7). In this case the estimated coefficient of domestic credit as a ratio of GDP is not found significant at the conventional levels of testing. Using m2 as a ratio of GDP (column 9) does not seem to change the significance of the estimated coefficients. All these therefore point to the fact that the earlier studies that find financial development impacting economic growth in a significant way are not supported here. Evidence elsewhere (see Giuliano & Ruiz-Arranz, 2009) tends to support the notion that countries with higher level of financial development tend to have a lower marginal effect of remittances on growth. Table 5 presents the results where the interaction variable is added to the regressions. As shown in columns 3 and 8, the estimated coefficients of the interaction variable are found to be positive and significant at the conventional levels. This finding supports the complementarity hypothesis and corroborates the findings by Mundaca (2005) and Bettin and Zazzaro (2008). This finding may be motivate as follows: countries with high levels of financial development help migrants to send more money home and, in turn, a significant inflow of remittances contributes to promoting financial democracy – a better access of the population to services offered by financial institutions. Such process may occasion a virtuous circle, where an increase in remittances brings about a higher level of financial development that allows migrants to send even more money with efficient financial institutions helping to channel these remittances towards productive investment projects (Terry & Wilson, 2005). The increase in remittance flows to countries with relatively developed financial markets tend to occasion a gradual process of institutionalization and bankarization – the implementation of a structured network of professional financial intermediaries targeting the unbanked population. This increase in such networks may increase competition leading to a decline in intermediation costs, hence benefitting remittance recipients. The convergence theory is tested in this study using the lagged GDP per capita. In all the estimations it is found that the estimated coefficients are negative and significant at the conventional levels of testing. However, it is shown that the magnitude of the estimated coefficients appears to display wide variations. However, these results are consistent with the neoclassical model which postulates that the economy tends to approach its long run position if the starting per capita income is low. The result therefore supports the conditional convergence hypothesis in which case poor countries grow faster than richer countries. The result corroborates the work of Barro and Xavier (1997), Easterly and Levine (1997); Sachs and Warner (1997).

258

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

The estimated coefficients of private investment are found to be positive as expected. In the pooled models in Table 3 the ratio of private investment to GDP ratio is positive and significant at the conventional level of testing. Including the remittances to GDP ratio leaves the estimated coefficient largely unaffected. Including the financial development indicators leaves the results largely unaffected. The same observation is made when estimation is done using the fixed and random effects models as shown in Tables 3–5. This finding therefore gives support to the notion that higher level of private investment leads to higher economic growth. The role of human capital on growth proxied by the primary school enrolment rate is also investigated in this study. In the baseline model in Table 3, it is found that the estimated coefficients of gross primary school enrolment rate are positive and significant at the conventional levels of testing in both the OLS and 2sls estimations. Introducing the remittances as a ratio of GDP and the volatility of remittances in Table 4 leaves the results unchanged. In addition, introducing financial development indicators into the estimation as shown in Table 5 leaves the results unambiguously positive and significant at the conventional levels of testing. This therefore shows that the estimated coefficients are not sensitive to the model formulation which may suggest that the coefficients obtained here are robust. This finding therefore gives support to the view that higher level of human capital formation has growth enhancing effect. The role of inflation in dampening economic growth is also investigated here. In the baseline model in Table 3, inflation is found to be consistently negative in all the specifications both under the OLS and the 2sls estimation approaches. However, it is noted that the estimated coefficients are not unanimously significant at the conventional levels of testing. Further analysis on the role of inflation is shown in Tables 4 and 5 where it is shown that the estimated coefficients are sensitive to the specification. In some case the estimated coefficients are positive and not significant. However, in most of the case, particularly where the estimated coefficients are found to be positive they are also found to be significant at the conventional levels of testing. The negative coefficient supports the traditional view that higher economic growth may only be achieved in an environment of low and stable inflation rate. Low inflation is advocated for because it creates an environment that is easy to predict into the future. This is because investors are more worried about the future and will tend to attach their long term investment decisions based on level of certainty they see in a country. The role of government in the economic growth of the countries in the sample is also tested. In the models shown in Tables 3–5, it is found that the estimated coefficients of government consumption to GDP ratio is unanimously negative and largely significant in all the specifications. In the standard growth model shown Table 3, the estimated coefficients are found negative and significant at the conventional levels of testing and the same result hold in other specifications in Tables 4 and 5. This finding seems to give credence to the notion that higher involvement of the government in economy will have significant negative consequences on the growth performance. 5. Conclusions This study investigates the role of remittances and financial development on economic growth in a panel of 36 countries in Saharan Africa over the period 1980–2009. It is motivated by the realisation that up to very recently remittances was not considered useful and as a result was ignored by both policy makers and academics. However, in recent times there seems to be shift of focus to the role of remittances on economic growth. Available evidence shows that recorded remittances in 2009 were nearly three times the amount of official aid and almost as large as foreign direct investment (FDI) flows to developing countries and that in 2010 stood, worldwide remittance flows, exceeded US $440 billion of which US $325 billion were transmitted to developing countries. The main findings of the study are as follows: (1) the importance of financial development in boosting economic growth appears weakened, at least among the countries under study; (2) remittances appear to be an important source of growth for these countries in Africa during the period under study; (3) volatility of remittances appears to have a negative effect on the growth of countries in Africa; (4) remittances appear to be working as a complement to financial development. This study has important implications for policy: (1) the declining importance of financial development in boosting economic growth implies despite efforts to encourage financial development in

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

259

Africa the role of financial development should be viewed as facilitative; (2) following the finding that remittances are important in boosting economic growth in Africa and that its volatility has an important consequence of economic growth suggests that governments in Africa should put in place measures to ensure that the flow of remittances is not hampered; (3) government in Africa should strive to promote formal or official channels of transmission of the remittances. This will ensure that the complementary nature of remittances to financial development identified in this study will be enhanced.

References Adams, R. (2004). Remittances and poverty in Guatemala. World Bank policy research paper, No. 3418. Adams, R.H., & Page, J. (2003). International migration, remittances, and poverty in developing countries. Policy Research Working Paper Series 3179, The World Bank. Adams, R. H., & Page, J. (2005). Do international migration and remittances reduce poverty in developing countries? World Development, 33, 1645–1669. Aggarwal, R., Demirgüc¸-Kunt, A., & Martínez Pería, M. S. (2010). Do remittances promote financial development? Journal of Development Economics, doi:10.1016/j.jdeveco.2010.10.005 Ajala, O., & Kerebih, A. (2008). Accessibility in equality to basic education in Amhara, Ethiopia. Ethiopian Journal of Education and Science, 3(2), 11–26. Allen, F., & Giovanneti, G. (2010). The effects of the financial crisis on Sub-Saharan Africa. Review of Development Finance, 1(1), 1–27. Amuedo-Dorantes, C., & Pozo, S. (2004). ‘Workers’ remittances and the real exchange rate: a paradox of gifts’. World Development, 32(8), 1407–1417. Anton, K. I. (2010). The impact of nutritional status of children in Ecuador. International migration review, 44, 269–299. Bagehot, W. (1873). Lombard street: A description of the money market. London: H.S. King. Barajas, A., Chami, R., Fullenkamp, C., Gapen, M., & Montiel, P. (2009). Do workers’ remittances promote economic growth. IMF working paper WP/09/153. Barajas, A., Chami, R., Fullenkamp, C., & Garg, A. (2010). The global financial crisis and workers’ remittances to Africa: What’s the damage? IMF working paper 10/24. Barro, R. (1989). Economic growth in a cross section of countries. NBER working paper No. 3120. Barro, R. J. (1991). Economic growth in a cross section of countries. Quarterly Journal of Economics, 106(2), 407–433. Barro, R., & Xavier, S. (1997). Technological diffusion, convergence, and growth. Journal of Economic Growth, 2(1), 1–26. Bettin, Giulia, & Zazzaro, Alberto. (2008). Remittances and financial development: Substitutes or complements in economic growth? Mimeo: Università Politecnica delle Marche. Brown, P. C. (1994). Migrants’ remittances, savings, and investment in the South Pacific. International Labor Review, 133(3), 347–367. Calderon, C., & Liu, L. (2003). The direction of causality between financial development and economic growth. Journal of Development Economics, 72, 321–334. Calderón, Cesar, Fajnzylber, Pablo, & López, Humberto. (2007). Remittances and growth: The role of complementary policies. In Pablo Fajnzylber, & Humberto López (Eds.), Remittances and development: Lessons from Latin America (pp. 33–68). Washington, DC: The World Bank. Caporale, G. M., Rault, C., Sova, R., & Sova, A. (2009). Financial depth and economic growth in new E.U. Members. http://www.wbiconpro.com/234-Rault.pdf. Accessed January 2011. Carling, O. J. (2004). Emigration, return, and development in cape verde: The impact of closing borders. Population, Society, and Place, 10(2), 113–132. Catrinescu, N., Leon-Ledesma, N., Piracha, M., & Quillin, B. (2008). Remittances, institutions and economic growth. World Development, doi:10.1016/j.worlddev.2008.02.004 Chami, R., Fullenkamp, C. & Gapen, M. (2009). Measuring Workers’ Remittances: What should be kept in and what should be left out. Mimeo, International Monetary Fund. Chami, R., Fullenkamp, C., & Jahjah, S. (2003). Are migrant remittance flows a source of capital for development? IMF working paper, International Monetary Fund, Washington, DC, August. Deodat, A. (2011). Financial development, international migrant remittances and endogeneous growth in Ghana. Studies in Economics and Finance, 28(1), 68–89. Easterly, W., & Levine, R. (1997). Africa’s growth tragedy: policies and ethnic divisions. Quarterly Journal of Economics, 112(4), 1203–1250. Faini, R. (2007). Remittances and the brain drain: Do more skilled migrants remit more? World Bank Economic Review, 21(2), 177–191. Giulia, B., & Zazzaro, A. (2011). Remittances and financial development: Substitutes or compliments to economic growth. Bulletin of Economic Research, doi:10.1111/j. 1467-8586.2011.00398.x Giuliano, P., & Ruiz-Arranz, M. (2009). Remittances, financial development, and growth. Journal of Development Economics, 90, 144–152. Gupta, S., Pattillo, C., & Wagh, S. (2007). Impact of remittances on poverty and financial development in Sub-Saharan Africa. IMF working paper, WP/07/38. Hicks, J. (1969). A theory of economic history. Oxford: Clarendon Press. IMF. (2005). Two current issues facing developing countries: Workersˇı remittances and economic development. World Economic Outlook, April.

260

E.M. Nyamongo et al. / Journal of Economics and Business 64 (2012) 240–260

Ketkar, S., & Ratha, D. (2001). Securitization of future flow receivables: A useful tool for developing countries. Finance and Development, IMF, 38(1). King, R. G., & Levine, R. (1993). Finance, entrepreneurship and growth. Journal of Monetary Economics, 32, 30–71. La Porta, R., De Silanes, F. L., Shleifer, A., & Vishny, R. (1997). Legal determinants of external finance. Journal of Finance, 52, 1131–1150. Martínez, Pería, Soledad, María, Mascaró, Yira, & Moizeszowicz, Florencia. (2007). Do remittances affect recipient countries’ financial development? In Pablo Fajnzylber, & Humberto López (Eds.), Remittances and development: Lessons from Latin America (pp. 171–215). Washington, DC: The World Bank. Massey, D., & Parrado, E. (1998). International migration and business formation in Mexico. Social Science Quarterly, 79, 1–20. Misati, R. N. (2007). Liberalization, stock market development and investment efficiency in Africa. International Review of Business Research Papers Journal, 3(4), 183–191. Misati, R. N., & Nyamongo, E. M. (2010). Financial liberalization, financial fragility and economic growth in Sub-Saharan Africa. Journal of Financial Stability, doi:10.1016/j.jfs.2011.02.001 Misati, R. N., & Nyamongo, E. M. (2011). Financial development and private investment in Sub-Saharan Africa. Journal of Economics and Business, 63(2), 139–151. Mundaca, Gabriela. (2005). Can remittances enhance economic growth? The role of financial markets development. Mimeo: University of Oslo. Mundaca, D. (2009). Remittances, financial markets development and economic growth: The case of Latin America and the Caribbean. Review of Development Economics, 13(2), 288–303. Nyamongo, E. M., & Misati, R. N. (2010). Modelling the time varying volatility of stock market returns in Kenya. African Journal of Economic and management studies, 1(2), 183–196. Nyamongo, E.M., & Misati, R.N. (2011). Remittances and Banking Sector Development in Sub Saharan Africa. Paper presented at the Global Development Forum, Dubai, United Arab Emirates, November. Qaisar, A. (2001). Endogeneous growth and human capital: A comparative study of Pakistan and Sri Lanka. The Pakistan Development Review, 40(4), 987–1007. Rajan, R. & Subramanian, A. (2005). What undermines aid’s impact on growth? IMF working paper no. 05/126. Rao, B. B., & Hassan, G. M. (2011). A panel data analysis of the growth effects of remittances. Economic modelling, 28, 701–709. Ratha, D. (2003). Worker Remittances: An Important and Stable Source of External Development Finance, Global Development Finance. World Bank. Sachs, J. D., & Warner, A. M. (1997). Fundamental sources of long run growth. American Economic Review, 87, 184–188. Schumpeter, J. A. (1911). The theory of economic development: An inquiry into profits, capital, credit, interest and the business cycle. New York: Oxford University Press. Translated and Reprinted 1961. Sufian, M., & Siridopoulos, M. (2010). Does worker’s remittances affect growth: evidence from seven MENA labor exporting countries. International Research Journal of Finance and Economics, (46), 181–194. Terry, Donald, & Wilson, Steven. (2005). Beyond Small Change: Making Migrant Remittances Count. Washington, DC: InterAmerican Development Bank. Woodruff, C. (2007). Mexican microenterprise investment and employment: The role of remittances. INTAL Working Papers 1452, Inter-American Development Bank, INTAL. Woodruff, C., & Zenteno, R. (2007). Migration networks and microenterprises in Mexico. Journal of Development Economics, 82(2), 509–528. World Bank. (2003). Workersˇı remittances – an important and stable source of external development finance. Global development finance 2003. Washington, DC: World Bank. World Bank. (2004). Harnessing cyclical gains for development. Global development finance 2004. Washington, DC: World Bank. World Bank. (2011). Migration and remittances factbook 2010. Washington: World Bank. Yang, D. (2004). International Migration, Human Capital, and Entrepreneurship: Evidence from Philippine Migrants’ Exchange Rate Shocks. Ford School of Public Policy Working Paper No. 02-011, University of Michigan, Ann Arbor. Yang, D. (2008). International migration, remittances and household investment: Evidence from Philippines migrants’ exchange rate shocks. The Economic Journal, 118, 591–630.