In the field of international development cooperation, Africa is a very important and unique region. Most African countries have had painful experiences with Western imperialism and its negative outcomes and thus have received various types of assistance and foreign aid from former imperial powers and the international community. In Africa, a variety of development policies and experiments have been conducted to eradicate extreme poverty, improve the quality of life, and enhance socio-economic development. Despite decades-long efforts to achieve these goals, however, the continent still has the largest number of least developed countries (LDCs), and development issues are hotly contested. Although tens of billions of dollars have been spent in foreign aid over the last fifty years, the number of people who live under the poverty line increased from 375 million in 2000 to 400 million in 2015, clearly showing the ongoing need for economic and human development.

Among many institutions and donors that aim to help these African countries achieve social and economic development, the African Development Bank Group (AfDBG), a regional development institution which includes the African Development Bank (AfDB), African Development Fund (AfDF), and Nigeria Trust Fund (NTF), is a major actor whose primary mission is to provide financial resources and technical assistance exclusively to African member countries and to enhance cooperation among member countries and with other multilateral and bilateral donors. Because of its exclusive focus on member countries in Africa, the amount of resources transferred from the AfDBG to partner countries remains small compared to those from bilateral donors and multilateral agencies. In terms of the size of aid to Africa (net disbursement) as of 2015, the largest donor is the U.S. (9.32 billion USD), followed by EU Institution (6.25 billion USD). AfDBG provided 2.18 billion USD in 2015, ranking 9th in the world in terms of aid to African countries. Among multilateral agencies, the AfDBG is the fourth-largest donor following the International Development Association (IDA) of the World Bank Group (WBG), EU Institutions, and Global Fund. While the amount of resources provided by AfDBG had been much smaller than those by IDA, World Food Programme (WFP), and UN Development Programme (UNDP) by the end of the 1980s, AfDBG distributed more resources to African member countries than WFP and UNDP in recent years.

Although the financial resources that the AfDBG can mobilize to achieve its goals are still limited, it plays a very important role in the regional and national development of its member countries. Most of all, the AfDBG has a significant comparative advantage as it has a deeper understanding of the challenges to development African member countries face under similar political, social, economic, and cultural circumstances, and it can utilize accumulated knowledge based on decades-long experience and learning. Moreover, unlike extra- regional donors, the AfDBG has more localized and region-specific development goals and distributes resources based on these goals and therefore its resource allocation is of great importance to regional and national development. For instance, the primary goals of the AfDBG are referred to as the High 5s: “Light up and Power Africa,” “Feed Africa,” “Industrialize Africa,” “Integrate Africa,” and “Improve the Quality of Life for the People of Africa.” These goals reflect local development needs and efforts to achieve sustainable development goals (SDGs) at the global level. According to the AfDBG, accomplishing the High 5s would contribute substantially to achieving the SDGs by 2030. Importantly, the AfDBG has development strategies specific to fragile states and LDCs so that these countries can receive financial and technical assistance under more favorable terms and conditions.

In this sense, resource allocation by AfDBG is of practical and theoretical significance in light of its primary goal of national and regional development of African countries. Of greater significance is that there exist substantial variations in terms of the amount of loans and concessional funds to its member countries even among those with similar levels of socio-economic development. So, what affects the AfDBG’s resource allocation to partner countries? This question is largely underexplored in the existing literature. Moreover, few research has been conducted to compare the patterns of resource allocation by the AfDB which provides loans to development projects and AfDF which provides concessional funds. Given the different ways to help African countries achieve development goals, it is worth testing emprically if the factors associated with the AfDB’s resource allocation are in fact different from those associated with AfDF.

This paper empirically examines the divergent patterns of resource allocation by the AfDB and AfDF to partner countries. Along with factors that are emphasized in prior works such as partner countries’ needs and aid allocation by two major donors in Africa—the U.S. and France—, we pay particular attention to the internal governance of the AfDB and AfDF and its possible association with their resource allocation to partner countries. Specifically, our paper takes a closer look at two features of internal governance—whether a partner country is represented as a director on the boards of the AfDB and AfDF and how much the country holds voting power—partly because it has been found that partner countries—even with similar levels of national development and aid from the U.S. or France—receive different amounts of resources from AfDB and AfDF.

Our paper also examines if factors that are associated with resource allocation by the AfDB are different from those by the AfDF as few works have empirically tested it. Although both the AfDB and AfDF pursue the same goals of national and regional development in Africa, each has its own purpose, mandate and mission, and therefore, might have dissimilar priorities when allocating its scarce resources to partner countries. Existing works treat the AfDBG as a single organization and conclude that recipient needs are the AfDBG’s main concern (Neumayer 2003). However, it is of great necessity to disaggregate the AfDBG into two different institutions and to empirically examine if resource allocation by the AfDB and AfDF are associated with similar factors or not.

This paper is organized as follows. The second section examines prior works on resource allocation by development banks in general and the AfDBG in particular. The third section explains trends of resource allocation by the AfDB and AfDF and their internal governance. The fourth section presents the research design, followed by a statistical analysis of the factors that are likely to relate to the resource allocation of the AfDB and AfDF. The final section offers policy implications.

In the aftermath of World War II, developed countries began to transfer public funds, goods, and services—referred to as official development assistance (ODA)—to developing countries. As resources transferred from the Global North to the South amounted to hundreds of billions of dollars, there have been many controversies related to why aid-giving countries provide foreign aid to certain partner countries. Existing works suggest that donor countries extend ODA to developing countries to pursue their own political, diplomatic, military, and economic interests, to help developing countries achieve their economic development, to address the development gap between the Global North and South, or some combination of these reasons (McKinlay and Little 1977; Maizels and Nissanke 1984; Schraeder et al. 1998; Alesina and Dollar 2000). While the motivations of donor countries as aid-providers are still debated, it is widely accepted that most donors strive to achieve multiple goals (Berthélemy 2006; Hoeffler and Outram 2011; Kim and Oh 2012; Jain 2016).

More recently, the questions of why and how much multilateral donors such as multilateral development banks (MDB)—for example, regional development banks (RDBs) and the World Bank (WB)—allocate their resources to certain partner countries has drawn much scholarly attention as well (Dollar and Levin 2006; Weaver 2007; Birdsall 2018). In fact, multilateral donors have significantly contributed to the socio-economic development of partner countries as well as global aid norms. According to the Organization for Economic Co-operation and Development (OECD),¹ there are more than 210 major international organizations and funds in the multilateral system of international development cooperation, and approximately 40% of ODA from member countries of OECD’s Development Assistance Committee (DAC) is extended directly to or through multilateral organizations.

Among these multilateral organizations, RDBs such as the AfDBG, Asian Development Bank (ADB), and Inter-American Development Bank (IADB) play crucial roles in helping developing countries achieve their developmental goals and tackling region-wide challenges to sustainable development with various resources from grants and loans to knowledge. A distinctive feature of RDBs is that they are like commercial banks in that they seek to generate revenues and thus have stricter criteria for extending loans than other aid-providing multilateral organizations. Another unique feature involves the internal governance of RDBs. Each RDB has a board of governors and a board of directors, both of which are comprised of individuals from donor and partner countries, and play significant roles in making key decisions (Strand and Trevathan 2015). In this regard, it is likely that resource allocation by RDBs differs from that by bilateral donors or even other multilateral agencies whose decision-making processes are dominated by donor countries. In other words, more influence on aid allocation decisions by partner countries in RDBs can be expected.

The focal point of prior studies of MDBs’ resource allocation is determinants of aid distribution by the World Bank. These pioneering studies employ an analytic framework in examining the WB’s aid allocation that is similar to that used to analyze the determinants of bilateral aid such as donor interest versus recipient needs. For instance, Frey and Schneider (1986) suggest that the WB considers a variety of economic and geopolitical factors in making decisions to allocate resources to partner countries: per capita income, inflation, debt, political stability, and colonial experience. Other studies pay attention to the influence of the U.S. on decision-making process in the WB. The findings of these studies demonstrate that the WB provides more resources to countries that hold the same positions as the U.S. in the UN General Assembly or countries that import a greater share of U.S. products (Anderson et al. 2005; Fleck and Kilby 2006; Harrigan et al. 2006). This suggests that the WB distributes resources in accordance with the political and economic interests of the U.S., which is the WB’s top shareholder.

In contrast to the interest in the WB’s aid allocation, only a few works investigate what affects resource allocation by RDBs (Neumayer 2003; Ben-Artzi 2005; Jang and Jung 2018). In a study that compares aid allocation by four major RDBs such as IABD, ADB, AfDB, and Caribbean Development Bank, Neumayer (2003) argues that AfDB provides more aid to countries with lower per capita income, lower political freedom, less military expenditures, larger population, and colonial experience. Despite his contribution to the existing literature by empirically examining aid allocation by major RDBs, Neumayer fails to take into consideration the commercial nature of these banks.

Ben-Artzi (2005) and Jang and Jung (2018) take a closer look at the internal dynamics within RDBs. In an attempt to incorporate RDBs’ characteristics as commercial banks in the analysis, Ben-Artzi investigates if RDBs’ decisions are made in consideration of financial interests for their strategic lending or the influence of hegemonic states on the board of directors. She overlooks in her analysis, however, is the role partner countries play in RDBs’ decision-making procedures. More recently, Jang and Jung explore the effects of the ADB’s governance on its aid distribution and find that the ADB provides more aid to the partner countries that are represented by board members, which suggests that board members may exert influence on decision-making in line with the interests of their home countries. These works demonstrate that resource allocation by RDBs is not fully explained by conventional frameworks based on donor interests and recipient needs; internal factors need to be taken seriously to better understand RDBs’ aid allocation patterns.

In sum, despite scholarly interest in aid distribution by RDBs and the importance of Africa in the aid community, the AfDBG’s resource distribution and its determinants have been underexplored, and few scholars have examined its internal decision-making dynamics. More importantly, there is no empirical comparative analysis of the two constituent bodies of the AfDBG—AfDB and AfDF—, each of which has its own board of directors and voting scheme. In addition, unlike other RDBs like the IADB and ADB, in which the U.S. and Japan respectively have hegemonic influences, there is no clear hegemonic power in the AfDB and AfDF (Ben-Artzi 2005). In this vein, this paper takes into careful consideration the decision-making processes of these two important multilateral agencies with organizational mandates for African countries’ development and their impact on resource allocation.

Since African countries became independent from Western imperialism in the 1950s and thereafter, they have struggled for poverty reduction, sustainable economic growth, and social development in the region. The AfDB was established with 23 regional member countries in 1964, supported by the newly-independent countries’ strong will for development and the geopolitical intentions of the U.S to protect Africa against the spread of communism (Culpeper 1994). At first, the AfDB did not allow the participation of non-regional countries in order to maintain autonomy from outside influences, especially by former colonial powers. However, the chronic insufficiency of resources and lack of political support substantially restricted their efforts to build a regional self-help system (Mingst 2015). This led to the establishment of the AfDF in 1972 which, along with the AfDB, comprised the AfDBG to bring in extra-regional support. The AfDF aimed to provide concessional funds, contributing to poverty reduction and economic development of lower-income countries in particular. The AfDF consisted of the AfDB and 27 non-regional member countries. In 1982, the AfDBG started to approve the membership of non-regional countries in the AfDB as well.

The board of governors of the AfDBG holds power and responsibility as its highest decision-making body. The composition rule of the board of governors of the AfDB is identical to that of the AfDF: each member country may appoint one governor and one alternate governor to serve for five years. Powers and responsibilities of the board of governors include: general credit policy, approval of new membership and its terms and conditions, decisions on changes of the authorized capital stock of the Bank, establishment or approval of the administration of special funds, and authorization of general arrangements for cooperation with non-member countries and with other international organizations. Each member country has voting power over major issues. Basically, 625 votes are provided to each member and additional votes are given in accordance with a member’s contribution to the capital stock. Of great importance is that the AfDBG was designed to allow African countries as a group to have more voice than external actors by making the voting powers of the 54 African member countries a majority. The board of governors delegates to the board of directors the power of general operations of the Bank (AfDB 2011). Nevertheless, the board of governors is still a critical institution in that it retains full power and authority over matters delegated to the board of directors.

All partner countries are classified into three categories according to their credit ratings and income levels. The AfDB provides non-concessional loans to countries with relatively high levels of per capita income and creditworthiness to sustain non-concessional financing. On the other hand, the AfDF mainly provides grants and concessional loans to countries with low levels of per capita income and credit ratings. While most countries are eligible for either AfDB’s funds or AfDF’s funds, some countries are eligible to receive funds from both.

Each of the AfDB and AfDF has its own independent board of directors. The AfDB’s board of directors exercises all the powers and responsibilities delegated by its board of governors. These include approving particular direct loans, guarantees, investments, determining the rates of interest for direct loans and of commissions for guarantees, submitting financial reports, and determining the general issues regarding the services of the Bank. The board of directors of the AfDB is composed of 13 directors elected from regional member countries and 7 directors from non-regional member countries. Each director appoints an alternate director who is not of the same nationality. Meanwhile, the board of directors of the AfDF is composed of 14 directors: 7 representing regional member countries and 7 from extra-regional donor countries.

Resource distributions by the AfDBG have increased since 2009 along with increasing assistance to Africa by other multilateral agencies. The resources provided by the AfDBG from 2004 to 2016 amount to 43,274 million UA, of which 64.58% came from the AfDB’s resources and 35.32% from the AfDF. As shown in Figure 1, the AfDB distributed more resources than the AfDF except for 2005, 2006, and 2014. The AfDB has provided a total of 29.4 billion UA (about 45 billion USD) over the last 40 years. AfDF has provided grants and concessional loans to 38 African countries, financed by 29 donor countries as of 2017.

A closer look at the top 10 partner countries reveals different patterns of resource distribution by the AfDB and AfDF. Countries that received the largest amounts of funds from the AfDB are Morocco, South Africa, Nigeria, Tunisia, Egypt, Democratic Republic of Congo, Botswana, Angola, Gabon, and Cameroon (in order from the greatest to lesser amount). These 10 countries received almost 78% of total resources of the AfDB from 2004 to 2016, demonstrating that the distribution of non-concessional loans by AfDB is concentrated in a few countries. On the other hand, the AfDF provided most of their resources to Ethiopia, Tanzania, Democratic Republic of Congo, Kenya, Uganda, Ghana, Nigeria, Cote d’Ivoire, Burundi, and Burkina Faso. The amount of aid allocated to these countries from 2004 to 2016 was 9,092 million UA, which makes up 59.32% of all resources distributed by the AfDF during the same period. This observation demonstrates that there are clearly different patterns of resource allocation by the AfDB and AfDF.

The purpose of this study is to examine the resource allocation of AfDBG in African partner countries by taking into account three sets of variables: internal governance of the AfDB and AfDF, recipient needs, and factors related to extra-regional donors. This paper investigates whether and to what extent the number of board directors who represent a partner country and its voting power affect resource allocation of the AfDBG (Strand and Trevathan 2015), with particular attention paid to the governance structure of the AfDBG. It also compares patterns of resource allocation by the AfDB and AfDF to examine if they have different motivations with respect to their respective purpose and organizational mandate. Finally, it takes into consideration factors that have been stressed in the existing literature on aid allocation by multilateral agencies: recipient needs and major bilateral donors.

Internal Governance: Board of Directors and Voting Power

According to scholars who put an emphasis on the internal governance of MDBs in their aid allocation, it is of great importance to identify the actors who play crucial roles in making decisions in MDBs (Ben-Artzi 2005; Jang and Jung 2018). The first set of hypotheses relates to the internal governance of the AfDB and AfDF.

First, it is likely that membership in the board of directors of the AfDB and AfDF relates to resource flow. In most cases, a RDB’s board of directors is responsible for its general operations and key lending decisions, exercising all the powers delegated by its board of governors. These decisions are also made by boards of directors in the AfDB and AfDF. The Board members’ positions and opinions are reflected in the board’s decisions on loans and grants. Therefore, we suggest that a country is likely to receive more loans or grants if there is a board member who represents the country; otherwise, it would receive fewer amounts of loans or grants.

One may argue that a board member in an international organization, given his or her status as representing the whole organization, would not exercise any biased influences in favor of the country of his or her origin. Despite this expectation, there is plenty of evidence of a significant correlation between the existence of board members who represent partner countries and resource allocation to these countries by the WB and ADB. For example, partner countries which are represented on the board of directors receive more than double the funds from the IBRD (Kaja and Werker 2010). Similarly, the ADB also distributes more resources to those countries that are represented on its board of directors (Jang and Jung 2018). Still, others may stress that decisions in a RDB’s board of directors are made collectively, suggesting the need to examine the process of how the individual opinion of board members turn into group decisions on loans and grants. While we recognize this concern, it still makes sense that a board member is able to negotiate terms and conditions for grants and loans in favor of his or her home country in multiple rounds of board decisions. Thus, a positive relationship between the existence of a board member who represents a partner country and the amount of loans or grants to this country is expected. The following hypothesis is proposed:

H1: A partner country which is represented on the board of directors of the AfDB or AfDF is likely to receive more resources from the AfDB or AfDF than other countries that are not represented on the board.

Second, resources from the AfDB and AfDF to a partner country may be allocated in proportion to its voting power. A key organizational feature of the AfDB and AfDF is that both donor and partner countries as shareholders constitute their boards of governors that are their highest decision-making bodies. Each shareholder retains a given share of votes to exercise in general meetings of the AfDB and AfDF. Therefore, not only donor countries but also partner countries have formal influence on the decision-making processes although general operations are in the hands of the boards’ directors. A member country’s voting power is determined by the share of capital stock of the Bank or Fund held by the country in addition to basic votes equally given to all members. Therefore, we can expect a positive relationship between the voting power of a country and favorable resource distributions to the country.

Some may argue that the influence of voting powers on resource allocation is much weaker than that of the board of directors because decisions made in general meetings do not necessitate actual voting in most cases. However, given the constitutional status of the board of governors, each member country’s voting power is likely to have an impact on how the resources are distributed to partner countries, at least indirectly by way of election of directors and specific policy directives. In addition, small shares of votes by partner countries would not have substantial influence on resource allocation. It may be the case that countries with less voting power may coalesce into a voting bloc to make their voices heard. The relentless efforts by members in international organizations to increase their voting powers even slightly reflect the significance of votes. As such, we expect that voting power of a partner country positively relates to resource allocation to the country and propose the following:

H2: A partner country which holds more voting power in the AfDB or AfDF is likely to receive more resources from the AfDB or AfDF than countries that hold less voting power.

Recipient Needs

Consideration of partner countries’ needs can also affect distribution of the AfDB’s loans and AfDF’s grants to them. The primary mission of RDBs is to eradicate poverty within the region and assist sustainable development in member countries. Unlike many bilateral donors that allocate funds in consideration of their own national interests, RDBs as multilateral organizations can distribute resources to countries based on need, not on narrowly-defined organizational interests. In other words, it is expected that RDBs are less self-interested and more altruistic. They are more likely to allocate funds to countries where outside support is most needed, such as developing countries with a lower GDP per capita and higher infant mortality rate. Africa is a region where recipient needs are more evident than other regions. Especially in sub-Saharan African countries, a low GDP per capita and high infant mortality rates indicate a strong need for international support.

Although the AfDBG plays an important role as a major aid-providing institution for African countries, its two institutions—AfDB and AfDF—are supposed to respond to recipient needs differently. Given its organizational mandate, mission, and purpose, the AfDF is expected to consider partner countries’ needs such as GDP per capita and infant mortality rate more seriously than the AfDB does. Therefore, we hypothesize that the AfDB and AfDF are expected to allocate their resources in consideration of recipient needs in general while the AfDF’s resource allocation is more strongly affected by recipient needs than that of the AfDB.

H3: A partner country in which development needs are more evident is likely to receive more resources from the AfDB or AfDF.

H3-1: A partner country with a lower GDP per capita is likely to receive more resources from the AfDB or AfDF.

H3-2: A partner country with a higher infant mortality rate is likely to receive more resources from the AfDB or AfDF.

H3-3: The Resource allocation of the AfDF is likely to be more strongly affected by recipient needs than that of the AfDB.

Extra-regional Donors: the U.S. and France as Key Players in Africa

Extra-regional actors can also affect resource allocation by the AfDB and AfDF in the country. It is suggested in prior works that the distribution of the WB’s resources is strongly affected by the political and economic interests of the U.S. In a similar vein, a study of the ADB’s aid distribution finds that countries to which Japan, the largest shareholder of the ADB, provides more aid also receive more of the ADB’s aid (Kilby 2006; Lim and Vreeland 2013). It shows that major shareholders in an RDB can exert influence on key decision-making.

Unlike other RDBs, there is no dominant actor in the AfDB and AfDF. However, the U.S. and France are the two major actors in the AfDBG.² The U.S. was the average during the period from 2004 to 2016. France provided about 4,546 million USD on average during the same period. In addition, France keeps a close relationship with African countries that had been under its colonial rule (Ben-Artzi 2005). Although it can be argued that more aid from major donors to a partner country crowds out resources from an RDB, previous studies suggest a positive relationship between aid from major donors and from an RDB, which can be inferred as having a crowd-in effect.

H4: A partner country which receives more ODA from major extra-regional donors is likely to receive more resources from the AfDB or AfDF.

H4-1: A partner country which receives more ODA from the U.S. is likely to receive more resources from the AfDB or AfDF.

H4-2: A partner country which receives more ODA from France is likely to receive more resources from the AfDB or AfDF.

Data and Measurement

We use data from a variety of sources from the AfDBG, WB, and the OECD Creditor Reporting System (CRS) in order to measure key variables. The dependent variables are the total amount of resources approved by the AfDB and AfDF from 2004 to 2016. Instead of using data from the OECD CRS database, the amount of loans and grants are collected from the AfDBG’s annual reports because they do not have missing values.³ Three sets of explanatory variables are included in the analysis: internal governance, recipient needs, and influence of extra-regional donors.

The first set of independent variables relates to the internal governance of the AfDBG measured by the number of board members representing a partner country and the country’s voting power. The AfDB and AfDF are run by their own board of directors that make key decisions on resource allocation. According to the Agreement Establishing the African Development Bank and the Agreement Establishing the African Development Fund ,⁴ 13 out of 20 directors are elected from regional partner countries in the AfDB and 7 out of 14 directors are elected from partner countries in the AfDF. We create a variable based on the number of board members representing a partner country, which has a value of 0 or 1 because there is no partner country represented by more than one board member in either the AfDB or in AfDF.⁵ Another variable related to internal governance is a partner country’s voting power. Each member country’s share of voting power is presented in the AfDBG’s annual reports. Because the focal point of our analysis is the relative influence among African member countries, we exclude the voting power of non-regional member countries and recalculate the values so that the sum of each regional member country’s share of voting powers is a total of 100%.

A second set of explanatory variables is suggested to measure the development needs of a partner country: GDP per capita and infant mortality rate. We use the WB database for the GDP per capita and infant mortality rate. The final set of variables used to examine factors external to Africa is the influence of major extra-regional bilateral donors on resource distribution by the AfDB and AfDF. As the U.S. and France provide more ODA to African countries than other OECD DAC donors, the amount of these two major donations to each partner country is used.

As for control variables, a partner country’s governance and its population are included. In its earlier years, the AfDB approved loans based only on economic indicators of partner countries, similar to commercial banks, in order not to appear as intervening in sensitive domestic affairs. However, as global aid norms regarding good governance diffused and the influence of extra-regional donors grew, the AfDBG began to consider the governance of partner countries (Mingst 2015, 81-83). For instance, “The African Development Bank’s Strategy for 2013-2022” includes “governance and accountability” as one of its five top priorities. Moreover, poor governance in many African countries is another major challenge to poverty reduction and sustainable development. To measure the quality of governance in a partner country, an average value of six dimensions of governance in the Worldwide Governance Indicators (WGI) developed by the WB is employed as a proxy. These six dimensions of a country’s governance include voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption.⁶ Lastly, to control for the size of a partner country, its population is included in the analysis.

For variables such as the amount of resources distributed by the AfDB and AfDF, GDP per capita, and the amount of ODA by the U.S. and France, the logarithm (ln) is used. Table 1 presents a concise description of all of the variables and the sources of data and Table 2 shows summary statistics.

In terms of the number of board members representing a partner country in the AfDB and AfDF, the value is either 0 or 1 because only one director can be chosen from a member country. During the period from 2004 to 2016, partner countries such as Algeria, Cote d’Ivoire, Egypt, Libya, Morocco, and Nigeria were represented in the boards of directors in the AfDB and AfDF every year, but 18 countries had no elected director. The voting power of a partner country as well as values for other independent variables shows substantial variations across countries.

To empirically test the hypotheses, we construct panel data from 2004 to 2016 and adopt multiple regression analyses using a random effect model. As it is important to employ a proper estimation method for reliable empirical analyses, a Hausman test is conducted to find the best-fitting model for our estimation, which suggests a random effect model as the most appropriate. A Fixed effect model is not employed as there are few variations in some variables (i.e., board members). All explanatory variables and control variables are lagged by one year to mitigate the endogeneity problem and examine their effects on the dependent variables. A total of 54 partner countries of the AfDBG are included in the analysis.

5.2 Results and Discussion

The statistical result is shown in Table 3. As explained above, the AfDBG determines which countries can receive funds from the AfDB, AfDF, or both by considering such criteria as GDP per capita and credit ratings. However, the panel data shows that, for example, there are multiple cases where the AfDF allocated financial resources to a country which was not supposed to receive them. Therefore, we include all member countries in the statistical analysis to see if the independent variables affect resource allocation by the AfDB and AfDF regardless of a partner country’s qualification. The result clearly demonstrates different patterns of resource allocation between the AfDB and AfDF. Resource allocation by the AfDB is significantly affected by its own internal governance, while that by the AfDF is affected by factors relating to partner countries.

More specifically, there exists a statistically significant and positive relationship between a partner country’s voting power and the AfDB’s resource allocation to this country, supporting the second hypothesis. This result may be so because a country’s voting power per se is closely related to its relative socio-economic position in the region and therefore a higher credit rating which in turn positively affects RDBs’ loan decisions. However, given that the partner countries’ GDP per capita and infant mortality rates are not significantly associated with the AfDB’s resource allocation, a country’s voting power, regardless of the country’s ability to repay its loans, is clearly associated with the amount of the AfDB’s loans to the existence of a director representing a partner country and resource allocation to the country might be spurious. Follow-up studies are needed to further investigate the relationship.

By contrast, resource allocation by the AfDF is affected by factors related to partner countries such as GDP per capita and infant mortality, which supports the above hypothesis (H3 and H3-3). These results might reflect the fact that the very organizational mandates of the AfDF are to supplement the AfDB’s function and to provide concessional funding to socially and economically vulnerable member countries. In particular, the AfDF is mostly funded by advanced donors external to the region. Therefore, the role African countries play in allocating resources during decision-making procedures such as board meetings and general assembly meetings might be less critical than in the AfDB.

As for external factors, resource allocation by both the AfDB and AfDF is positively related to French aid. This result allows us to infer that aid by France to an African partner country induces the AfDBG’s resources to this country. Although French influence on its former African colonies has substantially declined since their independence, these countries, especially Francophone countries, and France still maintain close relationships. The positive relationship between French aid and the AfDBG’s resource distribution might reflect this historical legacy. However, the possibility that this relationship is spurious should be taken into careful consideration. For instance, France might provide more aid to former colonial countries for redemption of past behaviors or for political and economic purposes.

The result also shows that both the AfDB and AfDF provide more resources to partners with better governance. It suggests that these countries are more likely to perform better in terms of loan repayment and project evaluation, which are major concerns of the AfDB. Also, the AfDF provides more funds to partner countries with better governance with higher expectations of aid effectiveness.

In sum, the AfDB is likely to provide more resources to partner countries with more voting power. In contrast, the AfDF, which provides concessional loans and grants to partner countries, allocates more resources to countries in need. This different pattern of resource allocation between the AfDB and AfDF may seem natural given their respective organizational mandate, mission, and purpose. Still, it is worth noting that this empirical result demonstrates the AfDB and AfDF stick to what they are supposed to do for African countries. Finally, the finding of French influence on the AfDBG’s resource distribution may suggest that its historical legacy in Africa still continues.

The AfDBG, established to promote African countries’ socio-economic development and equitable development in the region, strives to find the best solutions to African problems although its resources are limited compared to those of donor countries and multilateral aid agencies. This paper investigates what factors relate to resource distribution by the AfDB and AfDF. Resource allocation by the AfDB, which resembles commercial banks, is more affected by its own internal governance, while that by the AfDF strongly relates to partner countries’ economic conditions. In addition, aid allocation by France to a partner country has a positive and statistically significant relationship with resource distribution by the AfDBG in the country.

This research contributes to the existing literature on multilateral agencies’ aid allocation by empirically testing what factors exert influence on it. More specifically, it focuses on the possibility that resource distribution is affected by the AfDB’s internal governance, which is in line with the finding of prior work on the ADB. Therefore, the strong association between a partner country’s voting power, regardless of its economic position in the region and therefore ability to repay a RDB’s loans, and resource allocation to the country suggests that a country’s voting power is possibly considered when loan decisions are made by the RDB. Its practical implication is that a developing country which would like to receive more loans from a RDB for national development may need to make efforts to increase its voting power in the RDB by contributing more to it as an investment.

Despite this paper’s findings of a positive and statistically significant relationship between internal governance and resource distribution of the AfDB and between resource allocations by France and the AfDBG, these relationships need to be further investigated to show they are not spurious but causally linked. In order to do so, in-depth interviews need to be conducted to see what kind of non-economic factors are taken into account when making decisions to distribute resources to partner countries. Finally, more analyses over a longer time period and with different ways of operationalizing key variables can be done in follow-up studies. Regardless of these shortcomings and limitations, this paper contributes to studies on international development cooperation by empirically testing the AfDBG’s resource distribution.