Trade Openness and Diversification of External Financial Flows for Development: An Empirical Analysis

Abstract

The implementation of sustainable development goals (SDGs) adopted in 2015 by the international community in the Agenda 2030 requires a substantial mobilization of financial resources. In the meantime, Goal 17 of this Agenda recognizes trade as an important means of the implementation of the SDGs. The current article investigates empirically the impact of openness to international trade on the diversification of external financial flows for development, which could help developing countries achieve the SDGs by 2030. To that end, three major external flows for development have been considered: development aid inflows, migrants’ remittances inflows and foreign direct investment (FDI) inflows. The analysis relies on a panel data set comprising 116 countries, over the period 1970–2017. The empirical analysis relies primarily on the two-step system generalized method of moments (GMM) approach and shows that greater trade openness exerts a positive and significant impact on the diversification of external financial flows for development, in particular, in the least developed countries (LDCs). As a result, greater openness to international trade could be an important tool for external capital flows diversification in developing countries.

JEL Classification: F13, F14, F21, F24, F35, O20

Keywords

Diversification of external financial flows for development development aid remittances foreign direct investment trade openness

Introduction

The Agenda 2030¹ for the sustainable development goals (SDGs) and the Addis Ababa Action Agenda² adopted by the international community, respectively, at the United Nations Summit in September 2015 and at the Third International Conference on Financing for Development in July 2015 called for a substantial mobilization of financial resources in order to meet these goals by the agreed deadline of 2030. According to the United Nations Conference on Trade and Development (UNCTAD) (2014), developing countries would need to invest around US$4 trillion per year to achieve the SDGs by 2030. At the same time, trade has been recognized in Goal 17³ of Agenda 2030 as an important means of the implementation of this Agenda.

Recent years’ statistics on resource flows (see, e.g., World Bank, 2016, p. 34) have shown that remittances inflows to developing countries were far higher than official development aid (ODA) (e.g., remittances represented more than three times development aid in 2013 and 2014). Moreover, they suggest that remittances are even greater than foreign direct investment (FDI) inflows once China is excluded.

Among these external capital inflows to developing countries, official development aid (ODA), remittances and FDI inflows are known to be relatively more stable, that is, less volatile than portfolio inflows, which are of short-term nature and are highly reversible, as compared to the three other flows. Therefore, from now onwards, we focus in our analysis on the three major external financial flows for development, namely ODA, remittances and FDI inflows.

The bulk of studies, which in the development literature have examined the determinants of these three external financial flows for development, have been conducted by separating the analysis for each type of external financial flows for development. Among these studies, those that have examined the role of trade in mobilizing the external financial flows for development include, for example, Alesina and Dollar (2000) and Gnangnon (2017) for ODA inflows, and Markusen (1984), Helpman and Krugman (1985), Markusen and Venables (1995), Edwards (1992), Gastanaga, Nugent, and Pashamova (1998), Asiedu (2002), and Brun and Gnangnon (2017) for FDI inflows. Concerning remittances inflows, there is—to the best of our knowledge—no such study that has examined the impact of trade openness on remittances flows, although some studies (e.g., Head and Ries, 1998; Markusen, 1983; Mundell, 1957; Rauch, 2001; Richards, 1994) have looked at the relationship between trade and migration, which may be considered as an indirect way to assess or examine the impact of trade openness on remittances. It is important to note in this regard that an increase in the migrant stocks may not necessarily lead to higher remittances inflows to the home countries of migrants.

That being said, it is worth highlighting that one of the studies that has examined in a framework of simultaneous equations the impact of trade openness on financial flows for development (including government revenue, ODA and FDI inflows) is that of Brun and Gnangnon (2017). These authors show evidence that trade openness is an important driver of these financial flows for development.

The current study aims to contribute to the existing literature on the impact of trade openness on external financial flows for development, not by examining once again this relationship for each capital flows, but rather by investigating the impact of trade openness on the diversification of external financial flows for development. The empirical analysis provides evidence that trade openness is strongly conducive to diversification of financial flows for development in recipient countries, in particular, in low-income countries. This clearly indicates that trade openness could be an important means of financing the implementation of the SDGs.

The rest of the article is structured as follows. The second section discusses the effect of trade openness on the diversification of external financial flows for development. The third section discusses the other possible determinants of diversification of external financial flows for development. The fourth section presents our measure of the diversification of external financial flows for development. The fifth section presents the empirical model for the analysis. The sixth section discusses the econometric strategy for the estimation of this model, and the seventh section interprets the empirical results. The eighth section concludes.

Discussion on the Effect of Trade Openness on the Diversification of External Financial Flows

In this part of the analysis, we discuss the impact of trade openness on the diversification of financial flows for development. In other words, this section aims to examine the theoretical impact of a country’s degree of trade openness on its ability to attract these three types of financial flows. It is important to clarify that the objective of the study is not to investigate the complementarity and substitutability between these inflows, but rather if trade openness contributes to attracting all of these inflows considered jointly. While the theoretical impact of trade openness on development aid flows and FDI has been largely discussed in the literature,⁴ the direct impact of trade openness on remittances has received less attention, although the interplay between trade flows and migration has been investigated in the economic literature.

It is also important to stress here that we are examining the impact of trade openness, and not trade policy, on the diversification of financial flows; a country’s overall degree of trade openness reflects here both its trade policy and also other domestic policies (investment in human capital, institutional quality, exchange rate policy, etc.), which together determine the country’s level of openness to international trade. Therefore, our analysis involves the assessment of de facto trade openness on the diversification of external financial flows for development.

Discussion on the Possible Channels Through Which Trade Openness Could Influence Remittances

Migrants’ remittances are sent to the home country for both altruistic and investment motives (e.g., Chami, Fullenkamp, & Jahjah, 2003; Johnson & Whitelaw, 1974; Lucas & Stark, 1985).

The use by recipient countries of remittances to finance consumption or imports has been discussed in the empirical-related literature. According to Paine (1974), once families that receive remittances have tasted foreign goods and living standards, the propensity to import out of remittances becomes high. Durand and Massey (1992) obtain that most of migrant households’ income derived from remittances is used for consumption rather than for productive investments. At the same time, Zaman and lmrani (2005) find that higher remittances do not increase the demand for imported consumer goods but are rather used to finance the import of capital goods and raw materials. In the same vein, Adams (2006) finds that households that receive remittances use a large share of it for housing, education and health care financing and a small share for food and other non-durable goods consumption. All these show that migrants’ remittances are used for various purposes.

When guided by altruistic motives, remittances end up primarily financing domestic consumption. In this case, lower tariffs and non-tariff barriers on imports, thanks to greater trade openness, could contribute to financing consumption, in particular, if this consumption translates into higher imports. Hence, greater trade openness could induce higher altruism-motivated remittances in order to fuel higher consumption. Likewise, higher trade openness could lower the costs of locally produced goods and services by reducing the costs of intermediate inputs needed to produce such goods and services. As a result, domestic consumption could increase and remitters may be more willing to send remittances for consumption purposes.

When motivated by investment purposes, remittances aim primarily to take advantage of high returns or other investment opportunities in the home country. Greater trade openness, including through lower trade barriers, could encourage a rise in the remittances that finance imports of goods and services that are needed to support projects of migrants’ associations in the home countries.

The relationship between trade openness and remittances can also be viewed through the lens of the impact of trade flows on migration. It is worth noting in this regard that higher migration may not necessarily translate into higher remittances to home countries. For example, Foad (2010) provides evidence that there is a threshold of the stock of migration in the receiving country above which migration has a positive impact on trade, as below this threshold, trade might not be profitable. Elbadawi and Rocha (1992) have focused on North African and two European countries and found evidence that the stock of migrants in the host country exerts a positive effect on migrant remittances. Similarly, trade openness may not necessarily induce higher trade flows, including higher exports, as its ultimate effect on trade flows could depend on other factors such as human capital investment, business environment and the infrastructural quality. Trade has been argued to be a substitute for migration as it creates rising prosperity that can reduce the need to migrate. For example, according to the conventional factor–price equalization theorem (the Heckscher–Ohlin–Samuelson model), there is a substitution type of relationship between trade and migration (Mundell, 1957). In the meantime, other studies like Markusen (1983) have demonstrated that trade could be complementary with migration if one of the following assumptions of the analysis is relaxed: (a) constant returns to scale; (b) identical technologies; (c) perfect competition; or (d) no domestic distortions. Relying on the experience of trade liberalization in Asia and Latin America, Richards (1994) shows evidence of a complementarity between trade and migration. In the same vein, Head and Ries (1998) and Rauch (2001) also provide evidence of the complementarity between trade and migration.

Overall, we can expect that greater trade openness would induce a rise in remittances, though we do not rule out situations where trade openness may induce lower remittances.

Discussion on the Effect of Trade Openness on Development Aid Flows

The strand of the development aid literature that has examined the determinants of development aid has shown the importance of trade openness in donors’ aid allocation. Alesina and Dollar (2000) provide evidence that donors allocate more aid to reward developing countries for the good quality of their economic policies, in particular, their trade liberalization policies. Hence, by enhancing competitiveness and providing signals to the international community of their commitment to sound macroeconomic policies, including trade and trade-related policies, recipient countries of development aid could induce donors to provide them with higher development aid flows. For example, Brun and Gnangnon (2017) have provided empirical evidence that trade openness is conducive to higher aid flows to recipient countries.

Discussion on the Effect of Trade Openness on Development Foreign Direct Investment Flows

This section is essentially drawn from Brun and Gnangnon (2017). The international business literature on the determinants of FDI inflows has shown that the impact of trade openness on FDI inflows depends on the type of FDI inflows undertaken in the recipient country. Indeed, the literature on international business theory has identified broadly four models of FDI undertaken by multinational enterprises (MNEs): horizontal FDI models; vertical FDI models; export-platform FDI models; and complex-vertical FDI models. The theory underlying horizontal FDI models posits that MNEs are market seeking and want to expand overseas to avoid trade costs (Markusen, 1984; Markusen & Venables, 1995). Hence, high trade barriers in any destination country allow MNEs engaged in horizontal-type FDI inflows to serve the local market and to benefit from the protection of their output from imports of foreign competitors (tariff-jumping hypothesis). These types of FDI inflows lead to a substitutionary relationship with trade. The theory associated with vertical FDI models (e.g., Helpman, 1984; Helpman & Krugman, 1985) offers that FDI and trade could be complementary when vertical FDIs are involved due to the fragmentation of the production process geographically. Multinationals undertaking vertical FDIs are primarily motivated by low production costs in host countries and aim to serve both the domestic and foreign markets. As a result, the lower the parent country’s tariffs, the stronger the complementarity between FDI and trade. Export-platform FDI models have recently emerged to explain the paradox—stemming from the theory underlying horizontal FDI models—associated with the rise in world trade thanks to multilateral trade liberalization and the concomitant surge in FDI inflows (see, e.g., Collie, 2011; Neary, 2009). These types of FDIs are undertaken by MNEs in a host country with a view to serve both the local market and the surrounding countries. Market access conditions experienced by the host country exporters in neighbouring countries appear to be relevant for these types of FDIs (see Fugazza & Trentini, 2014). Moreover, high initial host country’s tariffs discourage these types of FDIs. Finally, complex-vertical FDI models are the most advanced investment strategies and are particularly motivated by the minimization of production costs. These FDIs entail the creation by MNEs of several production locations specialized in different phases of production. Complex-vertical FDIs require third countries’ access to the host country and the host country’s openness to the rest of the world. Based on this theoretical discussion, it is a priori difficult to anticipate the effect of trade openness on FDI inflows. Therefore, we conclude that the effect trade openness on FDI depends on the type of the FDI undertaken by MNEs (e.g., Asiedu, 2002).

From the empirical perspective, Edwards (1992), Gastanaga et al. (1998) and Hausman and Fernández-Arias (2000) show evidence that reforming trade regimes by making countries more open to international trade contributes to attracting more FDI. More recently, Brun and Gnangnon (2017) provide evidence that trade openness leads to lower FDI inflows to least developed countries (LDCs) (multinationals engage in FDI inflows in these countries when the latter levy higher trade barriers), but exerts a positive impact in FDI inflows in countries that are not LDCs.

Discussion on the Other Possible Determinants of Diversification of External Financial Flow

In this part of the analysis, we examine those factors (other than a country’s level of trade openness) that are common to these three strands of the literature (i.e., the literature on the determinants of remittances, FDI inflows and development aid flows) that influence the impact of trade openness on the diversification of financial flows for development. The lack of a comprehensive unified theoretical framework on the determinants of diversification of external financial flows for development leads us to draw on the existing literature on the determinants of each of the above-mentioned financial flows and consider the following determinants of diversification of financial flows for development: the level of economic development, measured by countries’ real per capita income; the depth of domestic financial markets; the accumulation of human capital; the size of the population; the institutional quality; and the world economic stance (proxied by the level of income in other countries, rather than the considered recipient country) and the inflation rate, as a proxy for macroeconomic stability in recipient countries of financial flows for development.

Impact of the Recipient Country’s Level of Economic Development on the Diversification of Financial Flows for Development

For development aid allocation, the level of economic development (proxied by real per capita income) acts as a proxy for recipients’ needs (e.g., Berthélemy, 2006), as recipients with a high level of development are expected to receive less aid. In the meantime, countries with a high level of development may benefit from higher donors’ aid flows supply if bilateral aid is guided by self-interest. At the same time, there is no clear-cut conclusion on the impact of the level of economic activity in the home countries on migrants’ remittances received by these countries. Indeed, many studies (e.g., Bouhga-Hagbe, 2006; Singh, Haacker, Lee, & Le Goff, 2011; Yang & Choi, 2007) have found a negative relationship between the level of economic activity in the home country and remittances received; other studies such as Higgins, Hysenbegasi, and Pozo (2004) and Aydas, Metin-Ozcan, and Neyapti (2005) have concluded that remittances tend to rise with improvements in recipient countries’ per capita income. As for the FDI inflows, the empirical literature has shown that the size of the local market, proxied by the level of economic activity in the recipient country, matters significantly to encourage FDI inflows. For example, according to Asiedu (2002), the size of the local market could be viewed by market-seeking FDIs as an opportunity to enter the host market (e.g., Okafor et al., 2015; Prasad, Rogoff, Wei, & Kose, 2003; Vo & Daly, 2007). However, multinationals oriented towards FDI may also be pursuing the objective of producing in the host country and exporting to the rest of the world. In such a case, an increase in the domestic market size may not necessarily be associated with higher FDI inflows in the host country. All in all, it is difficult to draw a conclusion on the ultimate impact of the recipient country’s exchange rate policy on the diversification of the flows it would receive.

Impact of the Recipient Country’s Financial Openness on the Diversification of Financial Flows for Development

The effect of financial openness (or capital account openness) on the diversification of external financial flows for development depends on the extent to which it affects each of the three types of external financial flows variables. While to the best of our knowledge there is no study on the effect of financial flows on overall development aid, a recent paper (Gnangnon, 2019) has examined the effect of financial openness on Aid for Trade (AfT), which is an important part of the overall development aid. Gnangnon (2019) has obtained empirical evidence that the effect of financial openness on AfT flows depends on the degree of financial openness, with countries with relatively low level of financial openness receiving higher amounts of AfT flows. The effect of financial openness on remittances has received scant attention in the literature. We postulate here that by easing restrictions on capital flows across a country’s borders, financial openness would affect the country’s real exchange rate and hence the amount of remittances that this country could receive. For example, as capital account restrictions are associated with a higher probability of an exchange rate crisis (Glick & Hutchinson, 2005; Rodriguez & Wu, 2013), it could adversely affect the amount of remittances sent. In this context, we expect greater financial openness to be positively associated with higher remittances inflows. Concerning FDI inflows, capital controls could result in higher FDI inflows (Butkiewicz & Yanikkaya, 2008). Similarly, Brafu-Insaidoo and Biekpe (2014) have reported that capital account liberalization, including through the liberalization of inward FDI, directly increases FDI inflows. Overall, as there is no clear direction as to which financial openness could influence the three types of external capital inflows, we argue that the effect of financial openness on the diversification of external financial flows is ultimately an empirical matter.

Impact of the Recipient Country’s Depth of Financial Development on the Diversification of Financial Flows for Development

To the best of our knowledge, there has not been any study in the development aid literature that has been specifically devoted to the effect of the depth of domestic financial markets in a recipient country on donors’ aid allocation to these recipient countries. Nevertheless, we argue here that countries that are financially less developed are likely to benefit from higher development aid flows as compared with countries that experience higher depth of domestic financial development. The literature on the determinants of FDI inflows has also identified the depth of financial market development as an important factor to explain foreign investment inflows. This factor acts as a proxy for the imperfection of financial markets: the deeper a country’s financial markets, the more capital flows it will attract. Alfaro, Chanda, Kalemli-Ozcan, and Volosovych (2008) argue that deeper financial markets may allow FDI inflows to finance short- and long-term transactions more easily and meet capital needs in the local market. Regarding remittances, it has been argued that well-functioning financial markets can help lower the costs of conducting financial transactions and consequently direct remittances to projects that yield the highest returns. At the same time, in the context of inefficient or non-existent credit markets, remittances may increase and act as a substitute to this credit constraint by helping local entrepreneurs bypass the lack of collateral or high lending costs and start productive activities (e.g., Giuliano & Ruiz-Arranz, 2009). Giuliano and Ruiz-Arranz (2009) have provided evidence that remittances boost growth in countries with less developed financial systems by providing an alternative way to finance investment and help overcome liquidity constraints. Hence, we could also expect countries with less developed financial systems to receive significant amounts of remittances.

Overall, this discussion does not allow us to definitely conclude on the direction of the impact of the level of financial development in the recipient country on the diversification of the flows it would receive.

Impact of the Recipient Country’s Level of Human Capital on the Diversification of Financial Flows for Development

As noted by Kpodar and Le Goff (2011), if external development aid reacts to human capital needs, then we should expect the level of human capital accumulated in a recipient country to be negatively associated with the amount of development aid that this country receives from donors. In other words, countries with better human capital receive less aid, everything being equal. As remittances could be used in the recipient economies to enhance human capital development through investment in higher education and health (e.g., Acosta, Fajnzylber, & Lopez, 2007; Ratha, 2013), we expect that countries with lower level of human capital development would receive higher amounts of remittances. The positive impact of human capital development, including through education and skill acquisition, on FDI inflows has also been emphasized in the related economic literature (e.g., Asiedu, 2006; Okafor et al. 2015). However, while authors like Cleeve et al. (2015) have obtained evidence that human capital (proxied by educational attainment) exerts a significant positive impact on FDI inflows to sub-Saharan African (SSA) countries, they have not observed evidence of the increasing importance of human capital on FDI flows to SSA over time. This probably reflects the fact that the effect of human capital on FDI inflows depends on the type of FDI flowing into this set of countries. Kar and Beladi (2004) have provided evidence that larger trade volumes are complementary with the rate of migration of skilled labour. This might suggest that greater trade openness in migrants’ home countries could be associated with higher remittances flows to their countries of origin, and hence contribute to the diversification of financial flows for development. On the basis of this discussion, we cannot conclude a priori on the direction of the impact of the recipient country’s level of human capital on the diversification of the flows it would receive.

Impact of the Size of the Recipient Country’s Population on the Diversification of Financial Flows for Development

The size of the recipient country’s population has been argued to be an important determinant of bilateral aid allocated by donors (e.g., Alesina & Dollar, 2000; Trumbull & Wall, 1994). Younas (2008) summarizes the importance of the population for aid allocation as follows: (a) the marginal impact of aid decreases as the population increases; (b) high-population countries lack the administrative expertise to absorb large amounts of aid; and (c) it is relatively easier for donors to wield political influence over a smaller country than a large country. Similarly, the size of the population could positively influence FDI inflows if the main objective of the multinationals undertaking the FDI in a host country is to serve the domestic market. However, if the chief motivation of these multinationals is to export either to neighbouring countries of the host countries or to the countries pertaining to the same regional integration area with the host country, or even to the rest of the world in general, then the population would matter less for attracting these FDIs (e.g., Gnangnon & Iyer, 2017). As for remittances, we argue here that the small countries, that is, those with a small population, likely face significant constraints on trade and economic growth opportunities that remittances could alleviate. Therefore, we expect recipient-countries with higher size of population to receive higher inflows of remittances.

Based on this discussion, we cannot conclude a priori on the direction of the impact of the size of the recipient country’s population on the diversification of the flows it would receive.

Impact of the Recipient Country’s Institutional Quality on the Diversification of Financial Flows for Development

The economic literature has shown that the quality of institutions matters for the three types of financial flows considered in this study. The theoretical and empirical impact of the institutional quality on donors’ aid allocation is ambiguous (for more details, see, e.g., In’airat, 2014). Some studies have obtained a positive effect of institutional quality on aid allocation, while others have found a negative or a non-statistically significant effect. For example, Alesina and Weder (2002) show that among donors, corruption is a decisive factor for aid allocation only for Australian and Scandinavian donor countries. Svensson (2000) and Neumayer (2003) find a weak role of corruption in the selection of recipient countries that will benefit from foreign aid. Alesina and Dollar (2000) found that apart from France, Italy, Belgium and Austria, most donors provide more aid to recipients with better political and civil rights. Trumbull and Wall (1994) observe that an altruistic donor may not necessarily provide more aid to poorer nations with the view to punish the recipient government for political oppression. Bandyopadhyay and Wall (2007) provide evidence that an improvement in civil/political rights in recipient countries is associated with higher aid receipts. Buchanan et al. (2012) provide evidence that institutional quality measured by the principal component of six governance indicators (voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rules of law and control of corruption—see Kaufman et al., 2010) exerts a positive and significant impact on FDI inflows and reduces the volatility of the latter. Good institutional quality has also been found to exert a positive effect on remittances inflows (see, e.g., Ajide & Raheem, 2016; Lartey & Mengova, 2016). Overall, we expect good institutional quality (good operation of institutions) to induce a diversification of financial flows of development in recipient countries.

Impact of the World Economy Stance on Recipient Country’s Diversification of Financial Flows for Development

As far as remittances are concerned, we expect that the level of economic activity in the resident country of migrants would influence the remittances that migrants would send back to their countries of origin. In particular, improved economic conditions in the migrants’ resident countries provide further employment opportunities for migrants, increase their earning prospects and put them in a better position to remit more money to their home countries (e.g., Chami et al., 2008; IMF, 2005; Schiopu & Siegfried, 2006). For a given recipient country, we measure the level of economic activity in the resident country of migrants by the stance of the world economy, the latter being measured by the difference between the world’s real GDP and the real GDP of the concerned recipient country. Similarly, we expect the improvement in the world economic conditions to be associated with higher aid flows (as donors would likely be wealthier) and higher FDI inflows. Overall, we expect the improvement in the world economic conditions to be associated with a diversification of financial flows to recipient countries.

Finally, the inflation rate could be an important determinant of diversification of financial flows for development, including through its important contribution to achieving macroeconomic stability in the recipient countries of the financial flows. For example, greater macroeconomic stability (through lower inflation rate) would likely attract higher FDI inflows as well as higher remittances inflows. Similarly, donor countries of development aid could also be willing to supply higher development aid flows to recipient countries that endeavour to achieve macroeconomic stability. As a consequence of all these, we could expect a lower inflation rate to be positively associated with diversification of financial flows for development.

Description of the Measure of the Concentration (Or Diversification) of Financial Flows for Development

Let us denote ‘AID’ as our measure of development aid inflows variable (in current USD), which is measured by the net ODA disbursement (ODANET). It represents gross amount of aid disbursements (i.e., the total aid disbursements over a given accounting period) minus the repayments of loan principal or recoveries on grants received during the same period.

The index of concentration of external financial flows for development denoted as ‘CONCFLOWS’ has been calculated using one of the ‘AID’ variables described above, along with the two types of external capital inflows considered in the analysis, namely remittances inflows (in current US$), denoted as ‘REMIT’, and FDI inflows (in current US$), denoted as ‘FDI’.

Step 1: The first step is to transform each of these variables into an index (denoted as ‘VARIndex’), which ranges from 0 to 100. The formula used is as follows:

V A R I n d e x = \frac{V A R_{i t} - \min (V A R_{i t})}{\max (V A R_{i t}) - \min (V A R_{i t})} * 100

(1)

where VARValue is one of the capital flows variables described above. ‘Min(VAR)’ and ‘Max(VAR)’ represent respectively the minimum and maximum values of ‘VAR’ in the sample.

Step 2: We rely on the Herfindhal index to compute our index of concentration of financial flows for development ‘CONCFLOWS’ index using the following formula:

C O N C F L O W S_{i t} = \frac{{(A I D I n d e x)}_{i t}^{2} + {(R E M I T I n d e x)}_{i t}^{2} + {(F D I I n d e x)}_{i t}^{2}}{{(A I D I n d e x_{i t} + R E M I T I n d e x_{i t} + F D I I n d e x_{i t})}^{2}}

(2)

where ‘AIDIndex’ denotes the index variable calculated in Step 1 for the ‘ODANET’ variable. ‘REMITIndex’ and ‘FDIIndex’ stand respectively for the indices calculated in Step 1 of the variables ‘REMIT’ and ‘FDI’. Higher values of the index ‘CONCFLOWS’ reflect a higher concentration of financial flows for development, while lower values of this index indicate a higher degree of diversification of financial flows for development.

Model Specification

We draw on the discussion provided in the third section and, in the absence of a theoretical model on the macroeconomic determinants of diversification of financial flows for development, we adopt a pragmatic approach. Hence, we estimate the impact of trade openness on the diversification of external financial flows for development by postulating the following model:

\begin{matrix} C O N C F L O W S_{i t} = α_{0} + α_{1} C O N C F L O W S_{i t - 1} + α_{2} O P E N_{i t} + \\ α_{3} F I N K O F_{i t} + α_{4} \log {(G D P C)}_{i t} + \\ α_{5} F I N D E V_{i t} + α_{6} E D U_{i t} + α_{7} \log {(W G D P)}_{i t} + \\ α_{8} \log {(P O P)}_{i t} + α_{9} I N F L_{i t} + α_{10} P O L I T Y 2_{i t} + \\ μ_{i} + λ_{t} + ω_{i t} \end{matrix}

(3)

where i represents the country’s index and t denotes the time period. The unbalanced panel data set comprises 116 countries spanning the period 1970–2017. In particular, we use non-overlapping sub-periods of 6-year average data, which include 1970–1975; 1976–1981; 1982–1987; 1988–1993; 1994–1999; 2000–2005; 2006–2011; and 2012–2017. α₀ to α₁₀ are parameters to be estimated. μ_i represents countries’ fixed effects and ε_it is a well-behaving error term. λ_t are time dummies that represent global shocks affecting external capital inflows for all the countries together.

The dependent variable ‘CONCFLOWS’ represents the measure of concentration of financial flows for development described above (see section 4). The one-period lag of this variable has been introduced in model (3) to capture the eventual persistence (dynamic) in the index of financial flows for development. The variable ‘OPEN’ is the key regressor of interest in the analysis. It is primarily measured by the de facto trade globalization index (denoted as ‘OPENKOF’) (see Dreher, 2006; Gygli, Haelg, Potrafke, & Sturm, 2019). This is an important component of overall globalization index. It is calculated as the weighted indicators of trade in goods, trade in services and the trade partner diversity.

In addition, we use for robustness check the traditional trade openness measure traditionally used in the empirical literature, which is the sum of exports and imports of goods and services measured as a share of GDP (denoted as ‘OPENNEW’). The definition and source of all variables used in model (3) are described in Appendix 1. Appendix 2 reports respectively the standard descriptive statistics, and Appendix 3 displays the list of countries used in the analysis.

Before moving on to the next section, we provide a first insight into the relationship between the index of concentration of external financial flows for development (‘CONCFLOWS’) and the trade openness variable (‘OPENKOF’) over the entire sample as well as over the two sub-samples, including LDCs⁵ and other countries in the sample, which we qualify as NonLDCs. Thus, in Figure 1, we present the evolution of the index of concentration of financial flows for development and the trade openness indicator over the full sample. Figure 2 shows the evolution of the same two indicators over the sub-samples of LDCs and NonLDCs. Figure 3 displays the correlation pattern between CONCFLOWS and OPENKOF. Over the full sample, the index of concentration of financial flows for development and the indicator of trade openness move in opposite directions. This pattern is confirmed for the sub-samples of LDCs and NonLDCs in Figure 2. Interestingly, while over the entire period trade openness in NonLDCs is higher than in LDCs, the index of concentration of financial flows has been higher in NonLDCs than in LDCs from 1970–1975 to 1988–1993. For the rest of the period, this index has become higher in LDCs than in NonLDCs. This, therefore, indicates that from 1970–1975 to 1988–1993, LDCs have experienced a higher level of diversification of financial flows for development than NonLDCs, whereas, for the remaining period, NonLDCs have enjoyed a greater degree of diversification of financial flows for development compared to LDCs.

Figure 1.

Source: The author.

Figure 2.

Source: The author.

Figure 3.

Source: The author.

Figure 3 shows a negative correlation pattern between the index of concentration of financial flows for development and the indicator of trade openness. This negative correlation pattern is observed over the full sample, as well as over the sub-samples of LDCs and NonLDCs. This suggests that greater trade openness is positively correlated with the diversification of financial flows for development.

Estimation Strategy

At the outset, we use two traditional estimators to estimate model (3) without the one-period lag of the dependent variable as regressor. These estimators include the within fixed effects (denoted as ‘FEDK’) where standard errors are corrected using the Driscoll and Kraay (1998) technique. This technique helps account for heteroscedasticity, serial correlation and contemporaneous cross-sectional dependence in the error term. The second estimator is the feasible generalized least squares (denoted as ‘FGLS’), with panel-specific first autoregressive (AR[1]) autocorrelation structure. The results of the outcomes of the estimations based on these estimators are reported in Table 1. The main problem with the use of these estimators is that they do not help address the endogeneity problems that could plague model (3), including the simultaneity bias associated with the reverse causality between the dependent variable and some regressors. The latter include the variables ‘OPENKOF’, ‘FINKOF’, ‘FINDEV’, ‘GDPC’, ‘POLITY’ and ‘INFL’. In addition, the specification of model (3) that has been estimated using FEDK and FGLS estimators does not include the one-period lag of the dependent variable. The absence of this variable also introduces an omitted variable bias in model (3). To account for these endogeneity concerns, we estimate the dynamic model (3) using the two-step system generalized method of moments (GMM) estimator developed by Blundell and Bond (1998). This estimator helps address not only the simultaneity bias problems but also the endogeneity problem associated with the correlation between fixed effects and the one-period lag of the dependent variable. The two-step system GMM estimator is also known to perform better than the one-step GMM estimator and the difference GMM estimator suggested by Arellano and Bond (1991) (e.g., Blundell & Bond, 1998; Bond, 2002; Davidson & Mackinnon, 2004; Roodman, 2009). We ascertain the validity of the two-step GMM system estimator through the following diagnostic tests: the standard Sargan test of overidentifying restrictions, which determines the validity of the instruments used in the estimations; the Arellano–Bond (AB) test of first-order serial correlation in the error term and no second-order autocorrelation in the residuals; and the results of the AB test of third-order autocorrelation to show the lack of autocorrelation at the third order in our estimates. In addition, we report the number of instruments used in the estimations, given that according to Roodman (2009), the tests described above may lose power if the ratio of the number of countries to the number of instruments used is less than 1. In other words, the number of instruments should be lower than the number of countries to ensure the validity of the two-step GMM system estimator.

Table 1.

Impact of Trade Openness on the Diversification of Financial Flows

Variables	FEDK	FGLS with Panel-specific AR(1) Autocorrelation Structure
	CONCFLOWS	CONCFLOWS
	(1)	(2)
OPENKOF	−0.00237***	−0.000498**
	(9.55e−05)	(0.000197)
FINKOF	0.000298	8.05e−05
	(0.000603)	(0.000201)
Log (GDPC)	−0.0714***	−0.0183***
	(0.0129)	(0.00321)
FINDEV	−0.00151***	−0.000349***
	(0.000309)	(0.000133)
EDU	−0.000373*	3.91e−05
	(0.000220)	(0.000163)
Log (WGDP)	−0.218***	−0.233***
	(0.0766)	(0.0120)
Log(POP)	0.0574	0.00967***
	(0.0781)	(0.00191)
Log(INFL)	0.00783*	0.0154***
	(0.00468)	(0.00227)
POLITY2	0.000573	0.00133**
	(0.000941)	(0.000536)
Constant	7.131***	7.809***
	(1.173)	(0.366)
Observations/Countries	623/116	620/113
Within R-squared	0.5596

Source: The author.

Notes: *p-value < 0.1; **p-value < 0.05; ***p-value < 0.01. Robust standard errors are in parenthesis. Estimators: FEDK and FGLS.

In model (3), the variables ‘OPENKOF’, ‘FINKOF’, ‘FINDEV’, ‘GDPC’, ‘POLITY’ and ‘INFL’ have been considered as endogenous. Moreover, we have used two maximum lags of dependent variables as instruments and two maximum lags of endogenous variables as instruments in the regressions.

It is worth mentioning that in addition to estimating model (3) by means of the two-step system GMM estimator over the entire sample, we also examine the net effect of trade openness on the diversification of financial flows for development over the sub-samples of LDCs and NonLDCs. To calculate these net effects, we define the dummy variable LDC, which takes the value 1 if a country is classified as an LDC, and 0, otherwise. This dummy is then interacted with the variable capturing trade openness in model (3). The results of the estimations of these different specifications of model (3) without/and with the LDC dummy and the interaction variable are reported respectively in column (1) and (2) of Table 2. As noted above, for robustness check, the variable ‘OPENNEW’ has been used as the alternative measure of trade openness in model (3). The results of the estimation of this specification of model (3) without/and with the LDC dummy and the interaction variable are reported respectively in column (1) and (2) of Table 3.

Table 2.

Impact of Trade Openness on the Diversification of Financial Flows in Least Developed Countries Versus Non-least Developed Countries

Variables	CONCFLOWS	CONCFLOWS
Variables	(1)	(2)
CONCFLOWS_t−1	0.639***	0.604***
	(0.0142)	(0.0186)
OPENKOF	−0.00152***	−0.00119***
	(0.000251)	(0.000283)
LDC ◊ OPENKOF		−0.00112**
		(0.000473)
LDC		−0.0378
		(0.0308)
FINKOF	0.00175***	0.00157***
	(0.000215)	(0.000227)
Log (GDPC)	−0.0223***	−0.0537***
	(0.00392)	(0.00676)
FINDEV	2.16e−05	−9.55e−05
	(0.000132)	(0.000131)
EDU	0.000464**	0.000253
	(0.000185)	(0.000246)
Log (WGDP)	−0.462***	−1.334***
	(0.141)	(0.164)
Log (POP)	−0.000354	−0.0108***
	(0.00295)	(0.00256)
Log (INFL)	0.000565	0.00102
	(0.00322)	(0.00338)
POLITY2	−0.000416	−0.000294
	(0.000530)	(0.000672)
Constant	14.56***	41.96***
	(4.352)	(5.105)
Observations/Countries	514/116	514/116
Number of instruments	105	106
AR(1) (p-value)	0.0000	0.0000
AR(2) (p-value)	0.7928	0.6776
AR(3) (p-value)	0.5053	0.5011
Sargan (p-value)	0.2968	0.2555

Source: The author.

Notes: *p-value < 0.1; **p-value < 0.05; ***p-value < 0.01. Robust standard errors are in parenthesis. In the two-step system GMM estimations, the variables ‘OPENKOF’, ‘FINKOF’, ‘FINDEV’, ‘GDPC’, ‘POLITY’ and ‘INFL’ have been considered as endogenous. Time dummies have been included in the regressions, but relating results are not reported here. Estimator: two-step system GMM.

Table 3.

Robustness Check—Impact of Trade Openness on the Diversification of Financial Flows

Variables	CONCFLOWS	CONCFLOWS
Variables	(1)	(2)
CONCFLOWS_t−1	0.627***	0.629***
	(0.0140)	(0.0159)
OPENNEW	−0.000496***	−0.000475***
	(0.000101)	(0.000120)
LDC ◊ OPENNEW		−0.000192
		(0.000240)
LDC		−0.0157
		(0.0285)
FINKOF	0.00127***	0.00129***
	(0.000251)	(0.000253)
Log (GDPC)	−0.0271***	−0.0369***
	(0.00357)	(0.00776)
FINDEV	0.000151	0.000139
	(0.000145)	(0.000132)
EDU	0.00101***	0.000881***
	(0.000153)	(0.000145)
Log (WGDP)	−0.573***	−0.812***
	(0.117)	(0.155)
Log (POP)	−0.00623**	−0.00805***
	(0.00250)	(0.00249)
Log (INFL)	0.00410	0.00335
	(0.00301)	(0.00335)
POLITY2	−0.00202***	−0.00172***
	(0.000439)	(0.000504)
Constant	18.07***	25.60***
	(3.620)	(4.849)
Observations/Countries	495/116	495/116
Number of instruments	105	106
AR(1) (p-value)	0.0000	0.0000
AR(2) (p-value)	0.7620	0.7475
AR(3) (p-value)	0.7131	0.7310
Sargan (p-value)	0.4259	0.3040

Source: The author.

Notes: *p-value < 0.1; **p-value < 0.05; ***p-value < 0.01. Robust standard errors are in parenthesis. In the two-step system GMM estimations, the variables ‘OPEN’, ‘FINKOF’, ‘FINDEV’, ‘GDPC’, ‘POLITY’ and ‘INFL’ have been considered as endogenous. Time dummies have been included in the regressions, but relating results are not reported here. Estimator: two-step system GMM.

Empirical Results

As noted in the previous section, Table 1 presents the results of the estimation of the static version of model (3) by means of FEDK and FGLS. Results in both columns (1) and (2) of Table 1 suggest a negative and significant effect (at the 1% level) of trade openness on the concentration of financial flows for development. In other words, over the full sample, greater trade openness is positively and significantly associated with diversification of financial flows for development. Concerning the other variables (control variables), we note across the two columns of Table 1 that financial openness does not influence significantly the concentration of financial flows for development. In the meantime, real per capita income, financial development and a better stance of the world economy drive positively and significantly the diversification of financial flows for development. The positive effect of the real per capita income on the diversification of financial flows for development suggests that advanced developing countries experience a higher level of diversification of financial flows for development than less advanced developing countries. At 5 per cent level, the education level does not influence significantly the concentration (or diversification) of financial flows for development. Similarly, at 5 per cent, the population size, the inflation rate and the institutional quality exhibit a significant effect (although a positive one on the concentration of financial flows for development) only for results displayed in column (2) (results based on the FGLS estimator).

Results displayed across the two columns of Tables 2 and 3 suggest that the variable ‘CONCFLOWS’ exhibits a state-dependent path, which shows that concentration of financial flows of development in period t – 1 is associated with greater concentration of these inflows in period t. In addition, the diagnostic tests confirm the validity of the system GMM approach estimator for carrying out the estimation of the dynamic specification of Equation (1). In particular, the p-values associated with the AR(1) test are lower than 10 per cent, while the p-values associated with the AR(2) and AR(3) tests are, as expected, higher than 10 per cent. Furthermore, the Sargan test of overidentification always displays a p-value higher than 10 per cent level of statistical significance.

Turning now to results in Table 2, we note from column (1) that trade openness exerts a negative and significant effect (at 1% level) on the concentration of financial flows for development, that is, greater trade openness induces a higher diversification of financial flows for development over the full sample. Specifically, a 1-point increase in the indicator of trade openness is associated with a 0.0015-point decrease in the index of concentration of financial flows for development. Results in column (2) suggest that while the coefficient of the variable ‘OPENKOF’ is negative and significant at the 1 per cent level, the interaction term associated with the interaction variable (LDC ◊ OPENKOF) is also negative and statistically significant, but at the 5 per cent level. These two results suggest that greater trade openness induces a higher positive effect on the diversification of financial flows for development in LDCs than in NonLDCs, and the net effect of trade openness on the concentration of financial flows for development is given by −0.0023 (= −0.00119−0.00112). Thus, a 1-point increase in the trade openness indicator is associated with a 0.0023-point decrease in the index of concentration of financial flows for development in LDCs, and a 0.0012-point decrease in the index of concentration of financial flows for development in NonLDCs. Focusing on the results in column (1) of Table 2, we note concerning control variables that greater financial openness and a higher education⁶ level are positively and significantly associated with a higher level of concentration of financial flows for development. Higher development level (higher real per capita income) appears to be a positive driver of diversification of financial flows for development. Financial development, the population size, the inflation rate and the institutional quality do not affect significantly (at least at 10% level) the degree of concentration (or diversification) of financial flows for development. Finally, an improvement is real-world GDP is associated with greater diversification of financial flows for development.

The robustness check analysis of the previous findings is provided in Table 3. Column (1) of Table 3 confirms the finding in Table 2 concerning the effect of trade openness on the diversification of financial flows for development: here, trade openness exerts a positive (negative) and significant effect on the diversification of financial flows for development. In particular, a 1-point increase in the indicator of trade openness (‘OPENNEW’) is associated with a 0.0005-point decrease in the index of concentration of financial flows for development over the full sample. The estimation’s outcome displayed in column (2) of Table 3 shows, in contrast with the findings of column (2) of Table 2, that there is no differentiated effect of trade openness—as measured by the sum of exports and imports of goods and services—on the concentration of financial flows for development in LDCs versus NonLDCs. In particular, a 1-point increase in the indicator of trade openness (‘OPENNEW’) is associated with a 0.00048-point decrease in the index of concentration of financial flows for development in LDCs and NonLDCs alike. Results concerning control variables are similar to those in column (1) of Table 2, with the exception of the population size and the institutional quality, which appear to exert a positive and significant effect on the diversification of financial flows for development (whereas in Table 2, the coefficients of these variables are not statistically significant at the 10% level). Likewise, results of control variables in column (2) of Table 3 are broadly consistent with those of column (1).

Conclusion

The international community adopted in 2015 two development agendas, namely the Agenda 2030 for the SDGs and the Addis Ababa Action Agenda (for financing development). In these two agendas, it called for a substantial mobilization of financial resources in order to meet these goals by the agreed deadline, that is, 2030. At the same time, trade has been recognized in Goal 17 of Agenda 2030 as an important means of the implementation of this Agenda. The current article investigates empirically whether trade openness could be instrumental in driving external financial flows for development in order to help developing countries achieve the SDGs by 2030. In other words, the article examines the impact of trade openness on the diversification of key external financial flows for development, namely development aid inflows, migrants’ remittances inflows and FDI inflows. The analysis relies on a panel data set comprising 116 countries, over the period 1970–2017. The empirical exercise is primarily based on the two-step system GMM approach, and suggests evidence that over the full sample, trade openness exerts a positive and significant impact on the diversification of external financial flows for development. This result also applies to the sub-samples of LDCs and NonLDCs. Interestingly, greater trade openness tends to exert a higher effect on the diversification of financial flows for development in LDCs than in NonLDCs. The key policy message of this study is that for developing countries, openness to international trade could be instrumental in driving external financial flows for development, which could in turn help achieve the SDGs.

Footnotes

Acknowledgements

This article represents the personal opinions of individual staff members and is not meant to represent the position or opinions of the WTO or its Members. The author would like to express his sincere gratitude to the anonymous Reviewers as well as the Editor for their useful comments on an earlier version this article. Any errors or omissions are the fault of the author.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

Definition and Source of Variables

Variables	Definition	Source
CONCFLOWS	Index of concentration of financial flows, based on Equation (3), displayed in the fourth section.	Author’s calculation based on data on ODA net disbursement (ODANET) from the Organisation for Economic Co-operation and Development (OECD) Development Database; data FDI inflows from the database of UNCTAD; and data on remittances inflows (REMIT) from the World Development Indicators (WDI, 2019) of the World Bank.
OPENKOF	This is the main measure of trade openness. It is in fact the de facto measure of trade openness, that is, the de facto trade globalization index developed (see Dreher, 2006; Gygli et al., 2019). It is a composite index of trade in goods, trade in services and trade partner diversity.	See the database and other information online at: https://www.kof.ethz.ch/en/forecasts-and-indicators/indicators/kof-globalisation-index.html
FINKOF	This is the measure of financial openness. It is in fact the de facto measure of financial openness, that is, the de facto financial globalization index developed (see Dreher, 2006; Gygli et al., 2019). It is a composite index comprising the sub-indices of FDI; portfolio investment; international debt; international reserves; and international income payments.	See the database and other information online at: https://www.kof.ethz.ch/en/forecasts-and-indicators/indicators/kof-globalisation-index.html
OPENNEW	This is the measure of trade openness used for robustness check. It is measured by the sum of exports and imports of goods and services measured as a share of GDP.	WDI (2019)
GDPC	Real GDP per capita (constant 2010 USD)	WDI (2019)
FINDEV	Domestic credit to private sector by banks (% of GDP)	WDI (2019)
EDU	This is the indicator of the education level in a country. It is measured by the average of the gross school enrolment primary rate, gross school enrolment secondary rate and the gross school enrolment tertiary rate.	Author’s calculation based on data from the WDI (2019)
POP	Total population	WDI (2019)
WGDP	This is the difference between real-world GDP and the real GDP of the country for which the variable is computed.	Author’s calculation based on data extracted from WDI (2019).
Proxy for the Institutional Quality (POLITY2)	This is an index extracted from the Polity IV database (Marshall & Jaggers, 2018). It represents the degree of democracy based on competitiveness of political participation, the openness and competitiveness of executive recruitment and constraints on the chief executive. Its values range between −10 and +10, with lower values reflecting autocratic regimes, and greater values indicating democratic regimes. Specifically, the value +10 for this index represents a strong democratic regime, while the value −10 stands for strong autocratic regime.	Polity IV Database (Marshall & Jaggers, 2018)

Source:The author.

Standard Descriptive Statistics on the Variables Used in the Analysis

Variable	Observations	Mean	Standard Deviation	Minimum	Maximum
CONCFLOWS	661	0.517	0.161	0.337	0.994
OPENKOF	878	45.938	19.223	5.135	89.714
OPENNEW	797	69.552	34.355	0.199	210.129
GDPC	834	3659.339	4937.315	146.345	36548.960
FINDEV	789	26.439	25.073	0.781	236.640
EDU	879	58.158	23.382	1.871	138.724
POLITY2	865	0.308	6.512	−10.000	10.000
INFL	847	53.389	288.527	−5.234	4651.100
POP	924	3.77e + 07	1.41e + 08	176371	1.37e + 09
WGDP	834	4.60e + 13	1.73e + 13	2.09e + 13	7.48e + 13

Source:The author.

List of Countries Used in the Analysis

Full Sample				Least Developed Countries
Afghanistan	Cyprus	Kenya	Papua New Guinea	Afghanistan	Rwanda
Albania	Democratic Republic of the Congo	Korea	Paraguay	Angola	Senegal
Algeria	Dominican Republic	Kyrgyzstan	Peru	Bangladesh	Sierra Leone
Angola	Ecuador	Lao People’s Democratic Republic	Philippines	Benin	Solomon Islands
Argentina	Egypt	Lebanon	Rwanda	Bhutan	Sudan
Armenia	El Salvador	Lesotho	Saudi Arabia	Burkina Faso	Tanzania
Azerbaijan	Eritrea	Liberia	Senegal	Burundi	Timor-Leste
Bangladesh	Eswatini	Libya	Serbia	Cambodia	Togo
Belarus	Ethiopia	Madagascar	Sierra Leone	Central African Republic	Uganda
Benin	Fiji	Malawi	Slovenia	Chad	Zambia
Bhutan	Former Yugoslav Republic of Macedonia	Malaysia	Solomon Islands	Comoros
Bolivia	Gabon	Mali	South Africa	Democratic Republic of the Congo
Botswana	Gambia	Mauritania	Sri Lanka	Eritrea
Brazil	Georgia	Mauritius	Sudan	Ethiopia
Burkina Faso	Ghana	Mexico	Suriname	Gambia
Burundi	Guatemala	Moldova	Tajikistan	Guinea
Cabo Verde	Guinea	Mongolia	Tanzania	Guinea-Bissau
Cambodia	Guinea-Bissau	Montenegro	Thailand	Haiti
Cameroon	Guyana	Morocco	Timor-Leste	Lao People’s Democratic Republic
Central African Republic	Haiti	Mozambique	Togo	Lesotho
Chad	Honduras	Myanmar	Tunisia	Liberia
Chile	India	Namibia	Turkey	Madagascar
China (People’s Republic of China)	Indonesia	Nepal	Uganda	Malawi
Colombia	Iran	Nicaragua	Ukraine	Mali
Comoros	Iraq	Niger	Uruguay	Mauritania
Congo	Israel	Nigeria	Venezuela	Mozambique
Costa Rica	Jamaica	Oman	Vietnam	Myanmar
Cote d’Ivoire	Jordan	Pakistan	Zambia	Nepal
Croatia	Kazakhstan	Panama		Niger

Source:The author.

Notes

References

Acosta

, Fajnzylber

, & Lopez

J. H.

(2007). The impact of remittances on poverty and human capital: Evidence from Latin American household surveys (Working Paper No. 4247). Washington, DC: World Bank Policy Research.

Adams

R. H.

(2006). International remittances and the household: Analysis and review of global evidence. Journal of African Economies, 15(2), 396–425.

Ajide

K. B.

, & Raheem

I. D.

(2016). The institutional quality impact on remittances in the ECOWAS sub-region. African Development Review, 28(4), 462–481.

Alesina

, & Dollar

(2000). Who gives foreign aid to whom and why? Journal of Economic Growth, 5(1), 33–63.

Alesina

, & Weder

(2002). Do corrupt governments receive less foreign aid? American Economic Review, 92(4), 1126–1137.

Alfaro

, Chanda

, Kalemli-Ozcan

, & Volosovych

(2008). Why doesn’t capital flow from rich to poor countries? An empirical investigation. Review of Economics and Statistics, 90(2), 347–368.

Arellano

, & Bond

(1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58(2), 277–297.

Asiedu

(2002). On the determinants of foreign direct investment to developing countries: Is Africa different? World Development, 30(1), 107–119.

Asiedu

(2006). Foreign direct investment in Africa: The role of natural resources, market size, government policy, institutions and political instability. World Economy, 29(1), 63–77.

10.

Aydas

O. T.

, Metin-Ozcan

, & Neyapti

(2005). Determinants of workers’ remittances: The case of Turkey. Emerging Markets Finance and Trade, 41(3), 53–69.

11.

Bandyopadhyay

, & Wall

(2007). The determinants of aid in the Post-Cold-War era. Federal Reserve Bank of St. Louis Review, 89(6), 533–547.

12.

Berthélemy

J-C

(2006). Bilateral donors’ interest vs. recipients’ development motives in aid allocation: Do all donors behave the same? Review of Development Economics, 10(2), 179–194.

13.

Bhattacharya

(2016). Post-2015 international development agenda in the context of interlocking trade and financing in the LDCs. E15 Initiative. Geneva, Switzerland: International Centre for Trade and Sustainable, Development (ICTSD) and World Economic Forum, 2015. Retrieved from www.e15initiative.org/

14.

Blundell

, & Bond

(1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87, 115–143.

15.

Bond

(2002). Dynamic panel models: A guide to micro data methods and practice (Working Paper No. CWPO9/02). London, UK: Institute for Fiscal Studies. Department of Economics, UCL, CEMMAP (Centre for Microdata Methods and Practice).

16.

Brafu-Insaidoo

W. G.

, & Biekpe

(2014). Determinants of foreign capital flows: The experience of selected sub-Saharan African countries. Journal of Applied Economics, 17(1), 63–88.

17.

Brun

, & Gnangnon

S. K.

(2017). Does trade openness contribute to driving financing flows for development? (Working Paper ERSD-2017-06). Geneva, Switzerland: World Trade Organization.

18.

Buchanan

B.G.

, Le

Q.V.

, & Rishi

(2012). Foreign direct investment and institutional quality: Some empirical evidence. International Review of Financial Analysis, 21, 81–89.

19.

Butkiewicz

J. L.

, & Yanikkaya

(2008). Capital account openness, international trade, and economic growth: A cross-country empirical investigation. Emerging Markets Finance and Trade, 44(2), 15–38.

20.

Chami

, Fullenkamp

, & Jahjah

(2003). Are immigrant remittance flows a source of capital for development? (IMF Working Paper, WP/03/189). Washington, DC: International Monetary Fund. Retrieved from https://www.imf.org/en/Publications/WP/Issues/2016/12/30/Are-Immigrant-Remittance-Flows-a-Source-of-Capital-for-Development-16801

21.

Chami

, Barajas

, Cosimano

, Fullenkamp

, Gapen

, & Montiel

(2008). Macroeconomic consequences of remittances (Occasional Paper No. 259). Washington, DC: IMF.

22.

Cleeve

E. A.

, Yaw

, & Zelealem

(2015). Human capital and FDI inflow: An assessment of the African case. World Development, 74, 1–14.

23.

Collie

R. D.

(2011). Multilateral trade liberalisation, foreign direct investment and the volume of world trade. Economics Letters, 113, 47–49.

24.

Davidson

, & MacKinnon

J. G.

(2004). Econometric theory and methods. New York, NY: Oxford University Press.

25.

Dreher

(2006). Does globalisation affect growth? Evidence from a new index of globalisation. Applied Economics, 38(10), 1091–1110.

26.

Driscoll

J. C.

, and Kraay

A.C.

(1998). Consistent Covariance Matrix Estimation with Spatially Dependent Panel Data. Review of Economics and Statistics, 80(4), 549–560.

27.

Durand

, & Massey

(1992). Mexican migration to the United States: A critical review. Latin America Research Review, 27(2), 3–42.

28.

Edwards

(1992). Trade orientation, distortions and growth in developing countries. Journal of Development Economics, 39(1), 31–57.

29.

Elbadawi

, & Rocha

R. D.

(1992). Determinants of expatriate workers’ remittances in North Africa and Europe (World Bank Policy Research Working Paper 1038). Washington, DC: World Bank.

30.

Foad

(2010). Assimilation and trade between the Middle East, Europe, and North America. Review of Middle East Economics and Finance, 6(2), 74–92.

31.

Fugazza

, & Trentini

(2014). Empirical insights on market access and foreign direct investment. ((UNCTAD Policy Issues in International Trade and Commodities, Study Series, No. 63). United Nations Conference on Trade and Development, Geneva, Switzerland. Retrieved from https://unctad.org/en/PublicationsLibrary/itcdtab67_en.pdf

32.

Gastanaga

, Nugent

J. B.

, & Pashamova

(1998). Host country reforms and FDI inflows: How much difference do they make? World Development, 26(7), 1299–1314.

33.

Giuliano

, & Ruiz-Arranz

(2009). Remittances, financial development, and growth. Journal of Development Economics, 90(1), 144–152.

34.

Gnangnon

S. K.

(2017). Structural economic vulnerability, openness and bilateral development aid flows. Economic Analysis and Policy, 53, 77–95.

35.

Gnangnon (2019). Financial openness and aid for trade in developing countries. South African Journal of Economics, 87(1), 46–64.

36.

Glick

, & Hutchinson

(2005). Capital controls and exchange rate instability in developing economies. Journal of International Money and Finance, 24, 387–412.

37.

Gnangnon

S. K.

, & Iyer

(2017). Structural economic vulnerability, trade policy and FDI inflows. Journal of International Commerce, Economics and Policy, 08(01), 35. doi: https://dx-doi-org.web.bisu.edu.cn/10.1142/S1793993317500053

38.

Gygli

, Haelg

, Potrafke

, & Sturm

J.-E.

(2019, forthcoming). The KOF globalisation index—Revisited. Review of International Organizations. Retrieved from https://doi.org/10.1007/s11558-019-09344-2

39.

Hausmann

, & Fernandez-Arias

(2000). Foreign direct investment: Good cholesterol? (Inter-American Development Bank, Research Department Working Paper 417). Washington, D.C.: Inter-American Development Bank.

40.

Head

, & Ries

(1998). Immigration and trade creation: Econometric evidence from Canada. Canadian Journal of Economics, 31(1), 47–62.

41.

Helpman

(1984). A simple theory of international trade with multinational corporations. Journal of Political Economy, 92(3), 451–471.

42.

Helpman

, & Krugman

(1985). Market structure and foreign trade. Cambridge, UK: The MIT Press.

43.

Higgins

M. L.

, Hysenbegasi

, & Pozo

(2004). Exchange rate uncertainty and workers’ remittances. Applied Financial Economics, 14(6), 403–411.

44.

IMF (2005). Two current issues facing developing countries. World economic outlook, April 2005: Globalisation and external imbalances, world economic and financial surveys. Washington, DC: International Monetary Fund.

45.

In’airat

(2014). Aid allocation, selectivity, and quality of governance. Journal of Economics, Finance and Administrative Science, 19, 63–68.

46.

Johnson

G. E.

, & Whitelaw

W. E.

(1974). Urban-rural income transfers in Kenya: An estimated-remittances function. Economic Development and Cultural Change, 22, 473–479.

47.

Kar

, & Beladi

(2004). Skill formation and international migration: Welfare perspective of developing countries. Japan and the World Economy, 16(1), 35–54.

48.

Kaufmann

, Kraay

, & Mastruzzi

(2010). The worldwide governance indicators methodology and analytical issues (World Bank Policy Research Paper No 5430 (WPS5430)). Washington, D.C.: World Bank.

49.

Kpodar

, & Le Goff

(2011). Do remittances reduce aid dependency? (IMF Working Paper, WP/11/246). Washington, DC: International Monetary Fund.

50.

Lartey

E. K. K.

, & Mengova

(2016). Does institutional quality in developing countries affect remittances? The Journal of Developing Areas, 50(1), 59–76.

51.

Lucas

R.E.B.

, & Stark

(1985). Motivations to remit: Evidence from Botswana. Journal of Political Economy, 93(5), 901–918.

52.

Markusen

J. R.

(1983). Free movements and commodity trade as complements. Journal of International Economics, 14, 341–356.

53.

Markusen

J. R.

(1984). Multinationals, multi-plant economics, and the gains from trade. Journal of International Economics, 16, 205–226.

54.

Markusen

J. R.

, & Venables

A. J.

(1995). Multinational firms and the new trade theory (NBER Working Paper No. 5036). Cambridge, MA: National Bureau of Economic Research.

55.

Marshall

M. G.

, Gurr

T. R.

, & Jaggers

(2018). Polity IV project: Political regime characteristics and transitions, 1800–2017. Vienna, Austria: Centre for Systemic Peace.

56.

Mundell

R. A.

(1957). International trade and factor mobility. American Economic Review, 47(3), 321–335.

57.

Neary

J. P.

(2009). Trade costs and foreign direct investment. International Review of Economics and Finance, 18, 207–218.

58.

Neumayer

(2003). The determinants of aid allocation by regional multilateral development banks and United Nations agencies. International Studies Quarterly, 47(1), 101–122.

59.

Okafor

, Piesse

, & Webster

(2015). The motives for inward FDI into Sub-Saharan African countries. Journal of Policy Equationing, 37, 875–890.

60.

Paine

(1974). Exporting workers: The Turkish case. London, UK: Cambridge University Press.

61.

Prasad

, Rogoff

, Wei

S.-J.

, & Kose

M. A.

(2003). Effect of financial globalization on developing countries: Some empirical evidence (Occasional Paper 220). Washington, DC: IMF.

62.

Ratha

(2013). The impact of remittances on economic growth and poverty reduction. Washington, DC: Migration Policy Institute.

63.

Rauch

J. E.

(2001). Business and social networks in international trade. Journal of Economic Literature, XXXIX, 1177–1203.

64.

Richards

(1994). Trade liberalisation and migration flows: Some evidence from developing countries. In Migration and development: New partnerships for co-operation (pp. 153–161). Paris, France: Organisation for Economic Cooperation and Development.

65.

Rodriguez

, & Wu

(2013). The effect of capital controls and prudential FX measures on options-implied exchange rate stability (Working Paper Series). San Francisco, CA: Federal Reserve Bank of San Francisco.

66.

Roodman

D. M.

(2009). A note on the theme of too many instruments. Oxford Bulletin of Economic and Statistics, 71(1), 135–158.

67.

Schiopu

, & Siegfried

(2006). Determinants of workers’ remittances: Evidence for the European neighbouring region (Working Paper Series 688). Frankfurt, Germany: European Central Bank.

68.

Singh

R. J.

, Haacker

, Lee

K.-W.

, & Le Goff

(2011). Determinants and macroeconomic impact of remittances in sub-Saharan Africa. Journal of African Economies, 20(2), 312–340.

69.

Svensson

(2000). Foreign aid and rent-seeking. Journal of International Economics, 51, 437–461.

70.

Trumbull

W. N.

, & Wall

H. J.

(1994). Estimating aid allocation criteria with panel data. Economic Journal, 104(425), 876–882.

71.

X. V.

, & Daly

K. J.

(2007). The determinants of international financial integration. Global Finance Journal, 18, 228–250.

72.

World Bank (2016). Migration and Remittances Factbook 2016, (3rd ed.). Washington, DC: World Bank. doi: 10.1596/978-1-4648-0319-2.

73.

Younas

(2008). Motivation of bilateral aid allocation: Altruism or trade benefits. European Journal of Political Economy, 24(3), 661–674.

74.

Zaman

K. U.

, & lmrani

N. A.

(2005). Workers’ remittances and import demand in Pakistan. The Philippine Review of Economics, XLII(2), 27–137.