Abstract
This study investigates the association between the unique characteristics of microfinance institutions and board structure. The agency and resource dependence theories provided theoretical guidance for this study. Using a panel dataset of 63 microfinance institutions in East Africa, we found that the presence of regulations and international influence is associated with larger boards, while the presence of founders is associated with small boards and less board independence. There is a higher level of board gender diversity in microfinance institutions managed by founders. There is greater diversity of nationalities in microfinance institutions that are internationally influenced. The implications for practice and theory from this study are further discussed.
Introduction
With the coming of age of the microfinance industry, it has recently been highlighted that microfinance providers need to improve their corporate governance structures (CSFI 2011). As suppliers of financial services to poor people, microfinance institutions (MFIs) are under pressure to improve their performance and a better governance structure is one way of achieving this objective (Labie and Mersland 2011). In this article, we address how board structure serves as one such governance mechanism.
In order for board-members to oversee and advise the organisation’s managers (Hillman and Dalziel 2003), they need a structure that supports such roles (Jackling and Johl 2009). However, little is known about MFI boards and what determines their structure. Microfinance policy papers (e.g., CMEF 2005; Conger et al. 1999) suggest how MFI boards should be composed and structured, but we lack knowledge on how these boards are actually structured and what factors determine the choice of a certain structure. This study addresses this knowledge gap by examining the relationship between MFIs’ unique characteristics and their board structure. Specifically, our research question is as follows: What determines the board structure of an MFI?
The unique characteristics of MFIs are related to a number of aspects. First, MFIs are characterised by their combined commercial and social goals. Their objectives are therefore to be financially sustainable while reaching out to poor people (Mori and Mersland in press). These objectives have brought them to the attention of international aid agencies and philanthropic investors (Mersland et al. 2011). Second, the international affiliations of MFIs give the industry another unique characteristic. Mersland et al. (2011), for example, found that the internationalisation of MFIs leads to better organisational performance. Third, MFIs vary in their operational forms. Some MFIs operate as regulated bank-like entities, with a profit motive, others as unregulated non-for-profit ones. Lastly, the microfinance industry is young and largely managed by its founders. For example, in our East African dataset, 54 per cent of the Chief Executive Officers (CEOs) of MFIs are also their founders. These characteristics are our motivation for this study; we argue that they partly determine the MFI board structure.
Steier (1998) found that board structure in for-profit organisations is set in a way that supports the specific demands of the industry. In relation to the microfinance industry, Conger et al. (1999) provide guidelines for a desirable MFI board structure. The authors suggest that MFI boards should be composed of a mixture of members with different skills, especially expertise in social and financial issues. Based on the agency and resource dependence theories, board roles are divided into two main categories: monitoring and advising (Hillman and Dalziel 2003). These theories suggest that the ability of a board to undertake these roles effectively depends on how they are structured (Linck et al. 2008). CMEF (2005) further suggests that MFI boards should be structured so that their members are beneficial to the organisation’s attainment of their dual objective.
We examined the determinants of board structure in East African MFIs (Kenya, Tanzania and Uganda). There are two reasons why we were motivated to study this region. First, Africa is recognised as the least developed continent in terms of microfinance, but has the biggest potential market for it (Consultative Group to Assist the Poor [CGAP] 2010). Within Africa, East Africa is shown to be the most developed region in terms of microfinance, for example, CGAP (2010) reports that 30 per cent of borrowers in Africa are from East Africa. Second, microfinance policy papers indicate that most international funding is channeled into East Africa (CGAP 2010). We expected that this significant influence of international donors and capital providers would push MFIs in the region into having an effective governance and board structure.
We examined the association between three MFI characteristics (regulatory status, international influence and founder management) and three board structure dimensions (board size, board independence and board diversity). Hypotheses were tested using an unbalanced panel dataset of 63 MFIs, obtained through a survey conducted between January and August 2010. The data were analysed using a mixture of analytical procedures. We used random effects estimation as the main analytical tool and regressed single measures of each board structure variable in relation to the MFI variables. To check robustness, we did joint estimations, using seemingly unrelated regression analysis, and to tackle endogeneity challenges, we performed a two-stage least squares regression with instrumental variables.
The results indicate that the structure of MFI boards is indeed determined by the MFIs unique characteristics. We found that large board size was determined by whether the MFI is regulated and whether it is exposed to international influence. These findings suggest that highly influential stakeholders, such as owners or international funders, have the power to be included on boards, which is a way for them to reduce agency costs. Our finding also suggest that Founder-managed MFIs are associated with small boards and regulated MFIs are associated with highly independent boards, while MFIs that are internationally influenced and managed by founders have less independent boards. The results further indicated that MFIs that are internationally influenced are associated with a greater diversity of nationalities, while those managed by founders are associated with higher gender diversity on their boards.
To our knowledge, this is the first study to address the determinants of MFI boards. We see this article as providing a contribution in three areas. The first is in relation to boards and governance in MFIs, as little is known about MFI boards. Labie and Mersland (2011) suggest that governance structure is among the most important factors that help MFIs to offer sustainable financial services to the poor. Our study highlights the factors that lead to a given choice of governance structure and, based on the agency and resource dependence theories, we suggest ways of improving that structure. Second, we contribute to an understanding of board selection and board members’ roles in non-profit and mission-driven organisations. We exhibit the roles played by board members in MFI industry and how they are achieved through a given structure. Third, we contribute to research on corporate governance in emerging markets, as the focus of our empirical study is East Africa.
The article proceeds as follows: The second section presents the background to microfinance in East Africa. The third section presents the theoretical background, followed by the fourth section presenting our hypotheses. The fifth section presents the data and methodology. We present and discuss the results in the sixth section followed by conclusions and implications in the last section.
Microfinance in East Africa
The East African countries chosen for this study were Kenya, Uganda and Tanzania. These countries share significant similarities in their legal and regulatory regimes (La Porta et al. 2008) but are heterogeneous in terms of the development of their national microfinance industry. This combination of cross-country institutional similarity with industry variations makes East Africa a suitable environment for microfinance research, particularly in relation to other English common-law countries.
Microfinance operations in East Africa officially started in the 1990s; however, there were microfinance-related activities before, for example, the Uganda Finance Trust was established in 1984 and the Kenya Female Finance Trust in 1982. The number of microfinance activities and operators has increased dramatically over the last twenty years, CGAP (2010) reports that 3.3 million East Africans were borrowing from MFIs as of December 2009. The current major providers of microfinance services in East Africa can be divided into four institutional forms 1 : banks, non-bank financial institutions (NBFIs), cooperatives and non-profit MFIs, which are commonly referred to as non-governmental organisations (NGOs). Cooperatives have vastly different ownership forms and structures and are therefore not included in this study (Hartaska 2005). For example, cooperative members vote for board members, who must be cooperative members themselves. This implies that the board structure of cooperative MFIs cannot be compared to those of other MFIs.
Another characteristic of the microfinance industry in East Africa is the existence of both domestically and internationally-funded/owned MFIs. MFIs with international partners, or those with international networks, have access to international funding and/or technical assistance in various areas. Commonly, fund providers influence the governance of these MFIs. For example, some funders specify how the board should be organised before they provide funding. Since the region receives extensive international attention, one might expect the governance structure of its MFIs to be more ‘developed’ than that in regions with less international involvement.
The level of government regulation in East Africa is more extensive than in other African regions and this again makes the MFIs of this region particularly interesting. We argue that this adds to the pressure to put an effective governance structure in place. In terms of regulatory regime, for-profit MFIs are regulated by central banking authorities, while non-profit MFIs are not regulated. All MFIs are required to have boards. Furthermore, the boards of all regulated MFIs must satisfy certain requirements. As Table 1 shows, for the boards of regulated MFIs in Kenya, at least two-thirds of the members must not be affiliated to the MFI. In Kenya and Uganda, regulated MFIs are required to have at least five board members. Board diversity is not specified in the regulations, although there are general specifications about the roles, skills and expertise needed on boards.
Requirements for Boards of Regulated MFIs
Despite the guidelines, there is still significant flexibility in terms of how MFI board members are appointed (CSFI 2011). Regulators only approve members who have been appointed by MFIs (CMEF 2005). In Uganda, five of the biggest MFIs became regulated between 2005 and 2007, with the founder remaining the CEO in some cases. In Kenya and Tanzania, similarly, more than half of the MFIs are managed by their founders.
Unregulated non-profit MFIs in East Africa come in a mixture of sizes and vary in terms of whether their founders are local or international. Locally-founded, non-profit MFIs are typically still managed by their local founders and, since they are unregulated 2 these founders have a great deal of influence over who sit on their boards. Internationally founded non-profit MFIs have board structures that vary depending on where the MFI was originally based. It is common to find non-profit MFIs in Tanzania whose board-members are CEOs (or board members) of the same MFI in other countries.
Table 2 compares MFIs in East Africa with MFIs worldwide. The global data were obtained from the rating agencies at www.ratingfund1.org and include MFIs from 73 countries.
Comparing MFIs from Global Rating Reports with Our East African Data
Sources: Global data set; East African Data set.
Note: ** p < 0.05.
Table 2 shows that, on average, MFIs in East Africa are somewhat similar to those worldwide in terms of age, gender diversity and board size. However, other characteristics are significantly different. In the East African industry, 66 per cent of MFIs have international affiliations, a larger proportion than in the global dataset. A larger proportion of East African MFIs, 54 per cent are managed by founders, compared to 38 per cent of the global dataset. Finally, there are significant differences between regulated MFIs in East Africa and those in the rest of the world.
Theoretical Background
The management and finance literature state that boards serve two roles (Hillman and Dalziel 2003). First, they advise managers on the organisation’s strategies, which is referred to as the resource and advisory role. Second, they monitor and control the performance of managers (Hermalin and Weisbach 1991), which is referred to as the monitoring role. Similarly, in MFIs, board members are expected to fulfill a mixture of roles and responsibilities. Campion and Frankiewicz (1999) and Dorado and Molz (2005) suggest that MFI boards undertake the following roles simultaneously. First, the board oversees the various constituencies served by the MFI. Second, it establishes the reputation of the MFI, thereby helping to ensure external legitimacy. Third, the board attracts and mobilises resources and expertise. It oversees the management, including the selection and removal of top managers. Finally, it assists in identifying the MFI’s core mission and works to keep it on track. In addition, the microfinance literature suggests that the board itself is expected to change according to changing organisational priorities and environmental demands (Dorado and Molz 2005). We grouped the first three roles together as resource and advisory roles, grounded in the resource dependence theory (Brown et al. 2011; Pfeffer and Salancik 1978). The final two roles come under monitoring, which was guided by the agency theory (Fama and Jensen 1983; Pant and Pattanayak 2010). We employ the complementary understanding of these theories to underpin our argument since they help to explain board roles and the board structure that is most suitable for the fulfillment of these roles.
The premise of this article, therefore, is that an appropriate board structure is necessary if MFI boards are to carry out their roles effectively. These ideas are highlighted in the microfinance practitioner literature, such as Conger et al. (1999) and CMEF (2005), both of which provide guidelines on MFI board structure in terms of composition, representation and size. Regarding composition, this literature suggests that board members should have social and commercial expertise, strategic and operational abilities and legal and communication abilities. Furthermore, it suggests that these skills can be provided by a diverse group of board members, from both inside and outside the MFI. Outsiders (independent members) are particularly valuable as they provide the boards with an independent mindset and experiences they can share with other members (CMEF 2005). Internally-recruited (dependent board members), like employees, and closely-related stakeholders (i.e., from sister organisations) provide the benefit of MFI-specific knowledge (Dorado and Molz 2005). Regarding board size, Conger et al. (1999) argue that MFIs should balance the need for skills and resources with the logistical advantages of small boards. These suggestions imply that board structure matters in MFIs in order for members to undertake their roles effectively and to enhance the attainment of MFIs’ dual objective. Based on these suggestions, theories and other arguments from past corporate research on board structure (e.g., Jackling and Johl 2009; Linck et al. 2008), we chose to address three dimensions of board structure: board size, board independence and board diversity.
Board Size
Board size is an important dimension, as past research indicates that it influences board effectiveness. The literature suggests that large boards are associated with higher coordination costs and free-rider problems (Jensen and Meckling 1976; Yermack 1996). These agency-theory-based arguments imply that relatively small boards operate more effectively, for example by being better at monitoring the CEO. For instance, Lipton and Lorsch (1992) argue that ‘when a board has more than ten members it becomes more difficult for them all to express their ideas and opinions’. Similarly, Jensen (2010) conjectures that ‘keeping boards small helps improve their performance’. In contrast, resource dependence scholars suggest that large boards are the source of various resources (Hillman et al. 2009; Pfeffer and Salancik 1978). From this perspective, board size is a measure of an organisation’s ability to form environmental links and secure critical resources. With different kinds of expertise more likely available on large boards, CEOs will obtain better advice and resources.
Within the microfinance industry, we argue that board size should be determined in a way that facilitates attainment of MFIs’ dual objective. A large board may enhance MFIs’ access to resources while being weaker at monitoring the CEO. Microfinance policy papers advocate boards that are large enough to provide adequate monitoring and resources (CMEF 2005; CSFI 2011). Hartarska and Mersland (2012) indicate that MFIs with large boards are associated with better social performance. They argue that MFIs can benefit from having large boards although the effect can change over time. Their results are consistent with the literature on banks and non-profit organisations, whose boards are found to be larger than conventional ones.
However, Mersland and Strom (2009) state that larger boards in the microfinance industry may induce members to a free-ride in their monitoring responsibility, allowing CEOs greater independence. Similarly, Hartarska (2005) shows that MFIs with small boards have better performance. The argument is that, small MFIs boards contribute to better results through their ability to monitor management. These studies provide evidence that MFIs can benefit from having both small and large boards, but there are some trade-offs between effective monitoring enhanced by small boards and resource abundance enhanced by large ones. Since MFIs board members fulfill both roles simultaneously, we argue that MFIs with large (small) boards benefit more (less) in terms of resources but are less (more) effectively monitored.
Board Independence
Board independence is important as it affects the ability of the board to monitor and provide resources to the management. The agency theory suggests that independent boards provide better monitoring and can better protect stakeholder interests (Fama 1980). Jensen (2010) further suggests that, because inside (non-independent) board members are likely to be ineffective at monitoring and evaluating the CEO, the only inside member on the board should be the CEO. The resource dependence theory argues that independent boards are more likely to reduce uncertainties and bring linkages and resources such as funding, skills, information and access to key constituents like policy decision makers (Hillman et al. 2009). Raheja (2005) developed a model that takes into account both the monitoring and resource provision roles of the board and found that independent boards are beneficial to organisations.
Alternatively, Harris and Raviv (2008) demonstrate that, when insiders have important information relative to outsiders, insider-controlled boards are preferred. Insiders can be better sources of internal information but weak at monitoring. Baysinger (1990) demonstrated that insiders contribute to boardroom discussions in ways that external board members cannot because the former have knowledge of the organisation’s operations and capabilities.
In a young and entrepreneurial industry like microfinance, various stakeholders value clarity and strategic direction. Roche (2009) suggests that young entrepreneurial organisations need more legitimacy and greater acquisition of resources than mature organisations. Similarly, we argue that most MFIs are at the development stage where the need for resources is great (in line with the corporate governance life-cycle arguments of Filatotchev et al. (2006) and Roche (2009). In addition, MFIs, regardless of their age, need board members that can oversee management and help the MFI mission to be on track (in line with boards’ roles).
This means MFIs can benefit from more independent boards, not only for better monitoring, but also because of the greater variety of resources that can be brought by outsiders (Dorado and Molz 2005). Insiders, due to their specific knowledge, are also important and can better assist CEOs with defining and maintaining a strategic focus for the MFI. However, Hartarska (2005) and Hartaska and Mersland (2012) found that MFIs with boards dominated by insiders are not beneficial. This implies that the level of board independence is important for an MFI and should be determined so that it benefits in terms of resources and monitoring.
Board Diversity
Board diversity relates to a combination of attributes, characteristics and expertise that individual board members contribute to the board’s activities. Similar to Walt and Ingley (2003), we define board diversity as the mix of human capital that a board comprises collectively and draws upon in undertaking its roles. This diversity is made up of age, gender, nationality, culture, religion, professional background, knowledge, technical skills, expertise and experience (Walt and Ingley 2003). Both the agency and resource dependence theories advocate diverse boards (Brown et al. 2012; Walt and Ingley 2003). According to the agency theory, diverse boards are better able to monitor CEOs. The resource dependence theory suggests that diverse boards are better at linking the organisation to external resources such as access to funding and connections with competitors and the market. We focused on diversity in terms of gender (Adams and Ferreira 2009) and nationality (Ruigrok et al. 2007) as these aspects contain a mixture of human capital.
Similarly, microfinance literature advocates diverse boards (CMEF 2005; Hartaska 2005; Strøm et al. 2010). Gender diversity is especially important because females constitute the main market for microfinance (Kyereboah-Coleman 2006; Strøm et al. 2010). The objective of the Microcredit Summit Campaign, which plays a central role in promoting microfinance, is ‘to ensure that 175 million of the world’s poorest families, especially females, have access to financial services’ (www.microcreditsummit.org). Mersland and Strøm (2009) also report that 73 per cent of the customers in their global dataset are female. Similar to Mattis (2000), we suggest that there are benefits to increasing the number of women in leadership and board positions for organisations that market their products to women. Women’s ability to contribute to boards may also be attributable to their different leadership styles. Strøm et al. (2010), meanwhile, suggest that gender-diverse boards are important for the effectiveness of MFIs and not only as a means of promoting democracy and equality.
In addition, past microfinance research highlights that female representation on boards can enhance MFIs’ understanding of female customers’ needs (Mersland and Strøm 2009). Strøm et al. (2010), for example, argue that female leadership and board members create a competitive advantage for MFIs due to their market orientation (i.e., females make up the majority of the MFI market). The authors further explain that female leaders may more easily tap into local information sources and help the MFI to create products that are better suited to its customers. Female leaders may also simply have better business acumen. Therefore, we expect MFI boards to be gender-diverse, so that MFIs can benefit not only from gender equality but also from the expertise, skills and experience of both genders.
Because of the high level of involvement of international organisations in the microfinance industry (Mersland et al. 2011), we see diversity of board members’ nationalities as advantageous to MFIs. Board members from other countries (than the MFIs home country) are likely to possess a diversity of cultural values and preferences (Caligiuri et al. 2004). Diversity of nationalities provides the board with a broader range of information resources, skill sets and cultural capital.
Hypotheses Development
The review mentioned earlier highlights that board size, level of independence and diversity matter and are important dimensions of MFI board structure. The question is what are the factors that determine the choice of a certain board structure? Past corporate governance literature helps us to identify possible factors (Lehn 2009). Factors that have been widely studied are the organisation’s size, legal status and age (e.g., Alonso et al. 2009; Linck et al. 2008). Since these factors are generic and mostly based on research in the corporate sector, we included them as control variables. Our contribution is to examine factors that we argue are unique to the microfinance industry and subsequently we hypothesized and tested the relationship between these factors and board structure dimensions. We focused on three factors: regulatory status, international influence and founder management.
Effect of Regulatory Status
MFIs are either regulated by central banking authorities or are unregulated. Regulated MFIs have defined shareholders and operate on a for-profit basis (CMEF 2005). Unregulated MFIs are NGOs and operate on a not-for-profit basis. Governance literature shows that non-profit organisations have large boards due to the fact that they lack owners with a monetary incentive to monitor their investments (Speckbacher 2008). For example, Alonso et al. (2009) showed that the average board size of a sample of non-profit organisations is eleven members.
Similarly, regulated organisations, such as banks, are shown to have large boards. Studies by Adams and Mehran (2003, 2008), for example, reported that the average size of a bank board is eighteen members. Regulated organisations have large boards due to the fact that they have a number of important constituencies who, according to the agency theory, sit on the boards to protect their interests (Adams and Mehran 2008). Constituents can be shareholders, deposit holders, or other independent board members (who indirectly ‘help’ regulators). Similarly, the resource dependence theory suggests that large boards are beneficial due to the resources that each member brings (Pfeffer and Salancik 1978). Board members, in this case, are a tool for managing external dependencies and reducing uncertainties (Hillman et al. 2000). Furthermore, the regulatory requirement in East Africa for example, (see Table 1) requires boards to have at least five members, but sets no limit. Based on the above arguments, we suggest that regulated MFIs need to appoint more board members than non-regulated MFIs in order to protect shareholders’ interests and increase support for the organisation by including key stakeholders that can be co-opted. Thus:
Hypothesis 1a: Regulated MFIs have larger boards than non-regulated MFIs
Unlike unregulated MFIs, regulated MFIs have defined shareholders who need to protect their interests by having independent boards who will monitor the management. In addition, regulated MFIs have greater flexibility over leveraging and seeking funding from various sources (Hartaska and Nadolnyak 2007). Therefore, regulated MFIs need to have board members who can support and assist them with leveraging and accessing funds. Hillman et al. (2000) further argue that regulated organisations need the expertise and skills of independent members such as business experts, influential members of the community and support specialists, such as lawyers, in order to manage external dependencies. Furthermore, regulated MFIs in East Africa, for example, are required to have non-affiliated board members. Adams and Mehran (2008) examined banks’ boards and showed that they are highly independent. This suggests that regulated MFIs will opt for more independent boards, due to their need to monitor management, attract outside resources and comply with regulatory requirements. Thus:
Hypothesis 1b: Regulated MFIs have greater board independence than unregulated MFIs
Unregulated MFIs focus more on social goals through serving female customers (Cull et al. 2009). The suggestion of Mattis (2000) and Strøm et al. (2010) that industries serving females need to have more females on their boards implies that unregulated MFIs may have more female board members than regulated MFIs. Furthermore, international donors provide more support to unregulated MFIs since they are non-profit organisations and to a large extent focus on social objectives (Mori and Mersland in press; Mersland et al. 2011). When international donors provide funds to MFIs, they monitor the usage of these funds by sitting on their boards (Mori and Mersland in press). Steane and Michael (2001) provide further evidence that the boards of non-profit organisations are more diverse than corporate boards. Since unregulated non-profit MFIs focus more on social goals and have more international donors than regulated for-profit MFIs, we expect their boards to be more diverse. Furthermore, regulated MFIs face regulatory requirements, which prescribe board-members’ skills. This implies that regulated MFIs have less incentive to recruit diverse board-members; they only recruit members who match the skills specified by the regulator. Thus:
Hypothesis 1c: Regulated MFIs have less board diversity than non-regulated MFIs
International Influence
There is a high level of international influence in the microfinance industry, which comes in various forms, such as networks, funding providers and international knowledge transfers (Mersland et al. 2011). CGAP (2010) reported that East African MFIs are strongly influenced by international partners (also see Table 2). International business research shows that the internationalisation of organisations brings about better performance (Lu and Beamish 2004; Mersland et al. 2011). This is achieved through, among other means, the provision of resources by international partners and better governance mechanisms (Ruigrok et al. 2007). Similarly, Sanders and Carpenter (1998) suggest that internationalisation is a way of strengthening access to resources and the interdependence of organisations. Furthermore, Barroso et al. (2011) show a positive association between internationalised organisations and board size. We therefore, suggest that MFIs that are influenced by international partners (who are members of international networks) are more likely to have larger boards, as this facilitates the transfer of resources such as funding, networks, information and skills by their international partners. Therefore:
Hypothesis 2a: MFIs that are internationally influenced have larger boards than those that are not
Typical international partners in the microfinance industry are large international NGOs, such as CGAP, FINCA international, Accion international and so forth (Rhyne 2005). These partners play the role of initiating, funding and owning MFIs. They also participate in selecting MFIs’ board members. Sanders and Carpenter (1998) argue that complexities, such as cultural differences and information asymmetries, arise from being associated with international partners. These complexities have implications for the agency relationships between the MFIs and their partners, enhancing the need for monitoring. In this case, we would expect international partners to require their partner MFIs to have more independent boards, hoping they would provide enhanced monitoring and better access to resources/networks. We further expect that the internationally influenced MFIs to be more exposed to the governance practices of their international partners. Based on the above arguments we suggest the following:
Hypothesis 2b: MFIs that are internationally influenced have greater board independence than those that are not
Since most international networks and supporters of microfinance come from the developed world, they prefer MFIs to have diverse boards, as advocated by the political and social forces in their countries. CMEF (2005) provides guidelines for MFIs that collaborate with international partners. Among other things, they recommend that boards should be diverse in terms of gender and cultural background. As mentioned earlier, international affiliations bring cultural complexities, and board members with diverse backgrounds could help to reduce these complexities and enhance monitoring. Ruigrok et al. (2007) further suggest that increased internationalisation of organisations leads to a higher demand for diverse board members who possess the necessary knowledge and contacts in both local and foreign markets. We therefore expect internationally influenced MFIs (MFIs which are members of international networks) to have more diverse boards, since this could enhance monitoring and facilitate better access to resources.
Hypothesis 2c: MFIs that are internationally influenced have greater board diversity than those that are not
Founder Management
Anecdotal evidence, as in the case of Muhammed Yunus and the Grameen Bank, indicates that founder CEOs are critically important organisational players in MFIs. The importance of founders could be due to the relatively young age of the industry (Boeker and Karichalil 2002). Our East African dataset shows a median MFI age of eight years (see Table 2). The implication is that a large number of founders are still present and they exert a significant influence (Alonso et al. 2009). We argue that founder CEOs have a tendency to treat the organisation as their ‘baby’ and want to hold on to it and control it for as long as possible. This was further exemplified in the recent struggle between the Bangladeshi government and Muhammad Yunus, regarding whether the latter should hold on to a board position within Grameen Bank (Burke 2011).
Corporate research shows that organisations with active founder CEOs have different agency issues and that such CEOs have an influence over the choice of board structure (Boeker and Karichalil 2002; Randøy and Goel 2003). This is due to both their formal position and their history within the organisation. Wasserman (2003) suggests that founder CEOs exert power through board seats and over strategic decision making. The agency theory advocates small boards since they are better monitors of CEOs (Yermack 1996). Monitoring might not be in the interests of founder CEOs, but they opt for small boards in order to exert greater control over board decisions (Daily and Dalton 1993). The literature also provides evidence that founder-managed for-profit organisations are associated with small boards due to their founders’ attachment and need to control the organisation (Certo et al. 2001). We therefore suggest that founder CEOs do not want to lose their control, power and influence and thus opt for small boards.
Hypothesis 3a: Founder-managed MFIs have smaller boards than non-founder managed MFIs
Corporate research in developed countries suggests that inside board members are an important complement to CEOs (Certo et al. 2001). Founder CEOs believe that inside board members can provide organisation-specific information that outside board members may lack (Certo et al. 2001). This is in line with resource dependence theorists, who claim that insiders are valuable for supplying the board with specific information about the firm’s competitive environment (Hillman et al. 2000). However, founder CEOs are likely to be less objective in assessing their organisations. Similarly, as founder CEOs need to maintain their influence and power, they are more likely to opt for more insider board members, whom they can control more easily (Boeker and Karichalil 2002). Daily and Dalton (1993) posit that insiders are unlikely to aggressively monitor and evaluate CEO actions due to their subordinate position in the organisation. Similar to the corporate world, we argue that CEO founders in the microfinance industry want to maintain their power, influence and control and therefore opt for less independent boards. This influence can be seen in Table 2. Nelson (2003) further shows that organisations managed by founders are less independent, which results in less monitoring and a high degree of CEO influence. We thus expect MFIs with founder CEOs to have more insiders on their boards as this facilitates the CEO–board influence and access to organisation-specific resources.
Hypothesis 3b: Founder-managed MFIs have less independent boards than non-founder-managed MFIs
Diverse boards might enhance monitoring since the CEO is less able to control the board’s decision making. In order to maintain his influence, a CEO founder might opt for a less diverse board. As Johannisson and Huse (2000) argue in relation to family-based organisations (many of which can be compared to other founder-influenced organisations), ‘it is the CEO, not the shareholders, who recruit board members, and when you recruit people, you recruit people in your own image, people you trust and can easily communicate with’. This implies that founder CEOs use their connections and invite people from their networks onto their boards. Founder CEOs are also likely to invite members they think will be easy to influence, so that decisions can be made in their favour. Diverse board members will be less dependent on the CEO and could provide contrasting views and discipline the management as needed (Fama and Jensen 1983; Walt and Ingley 2003). Despite the benefits brought by diverse boards, we expect to find that the boards of MFIs managed by their founders are less diverse, as this enables CEOs to maintain their control, power and influence.
Hypothesis 3c: Founder-managed MFIs have less board diversity than non-founder-managed MFIs
Table 3 summarises the stated hypotheses and their expected signs.
Summary of the Hypotheses
Methodology
Sample and Data Collection
The major source of data for this study was a hand-collected survey, conducted by one of the researchers between January and August 2010. The researcher contacted umbrella associations of MFIs in the respective countries and obtained MFI directories. These directories included information on MFI and CEO names, MFI locations and addresses. The researcher identified all of the MFIs listed in the directories, excluding cooperatives. Then, she contacted the CEOs of the identified MFIs and out of 103 MFIs contacted, 49 (47.6 per cent) were willing to participate. The MFIs that were unwilling to participate gave reasons such as the following: the CEO needed permission from the board chair in order to approve our visit (which he/she was not willing to do); others stated that such information was confidential and could not be shared for research purposes.
We visited the headquarters of the MFIs that were willing to participate and the CEOs provided information based on the checklist of questions about governance, board demographics and MFI operations. We further requested audited financial statements for the years 2004 to 2009 from each of the MFIs visited. Additionally, we used the snowball sampling technique (Saunders et al. 2003), whereby, when visiting an MFI, the CEO was asked to provide contact details of another MFI, which could have been on our original list or not. We visited any MFIs that fitted our categorisation (i.e., that were not cooperatives and had audited financial statements).
The websites of some of the MFIs that publish information on governance and audited financial reports were also accessed. To further check the quality of our information, we compared the information from these websites with that obtained from the CEOs. The data collection process led to a total of 63 MFIs (a mixture of local and international), board demographic information and financial information for six years. Some MFIs did not have all the information needed and therefore our dataset is an unbalanced panel of 63 MFIs, with 343 observations in total (Wooldridge 2006).
These MFIs probably represent between a quarter and a third of MFIs in the region. We recognise that, since most of the data were based on MFI directories, it is possible that some were left out and that we have some sample bias. However, we argue that our sample of MFIs represents the major and most established players in the region’s microfinance industry.
Variables and Measures
Dependent Variables
The dependent variables are based on three board structure dimensions and measured at the end of each year. ‘Board size’ is measured as the logarithm of the number of board members (Linck et al. 2008). ‘Board independence’ is measured as the proportion of board members who are not current or former employees (or initiators) of the MFI or members of their families (Wagner et al. 1998; Linck et al. 2008). We measured board diversity using two variables: ‘gender diversity’, defined as the ratio of female members to the total number of board members (Adams and Ferreira 2009) and ‘diversity of nationalities’, measured as the ratio of foreign members to the total number of board members (Caligiuri et al. 2004). Data on these variables were compiled from CEOs’ responses to questions about board members’ demographics and, in some cases, the MFIs’ websites.
Independent Variables
The independent variables are the MFIs’ characteristics. Data on these variables were obtained from the CEOs’ responses. ‘Regulated’ was measured as a binary variable equal to one if the MFI is regulated and zero otherwise. Here, we asked the CEOs whether the MFI was regulated by central banking authorities and if so when it began to be regulated. ‘International influence’ was measured as a binary variable equal to one if the MFI is a member of any international network or organisation that is based outside the MFI home country and zero otherwise (Mersland et al. 2011). The CEO here was asked whether the MFI was a member of any international organisation or network and for what purpose. ‘Founder management’ was measured as a binary variable equal to one if the CEO of the MFI was also the founder of it (Boeker and Karichalil 2002). The data here were obtained from the CEOs, by asking whether or not they were a founding CEO.
Control Variables
As mentioned earlier and in line with previous studies on corporate governance, we included control variables that also affect board structure. Since organisation size relates to possible economy of scale effects, we controlled for ‘MFI size’, measured as the logarithm of total assets (Al Najjar 2011). Age is related to changes caused by organisational life-cycle changes (Filatotchev et al. 2006) and therefore we included ‘MFI age’, measured as the number of years since the MFI began providing microfinance services. We also controlled for ‘legal status’, using a binary variable, equal to one if the MFI is an independent legal organisation or a branch of another organisation abroad and zero otherwise (Mori and Mersland in press). Data for these variables were obtained from the audited financial statements and CEO responses. We also employed ‘country dummy’ and ‘time dummy ‘variables in order to capture variations in economic development and time.
Data Analysis
Since our dataset was an unbalanced panel over 6 years, we could have used either a fixed effects or a random effects model, depending on the time invariance of the variables (Greene 2008). A fixed effects regression should be used to control for omitted variables that differ between cases (MFIs) but are constant over time. It allows the variables to change over time so that one can estimate the effects of the independent variables on the dependent variable (Wooldridge 2006). Some omitted variables may be constant over time but vary between cases and others may be fixed between cases but vary over time. To tackle this challenge, one can use random effects in order to take both situations into consideration and obtain efficient estimators. To choose between the two models, we ran a Hausman test (Hausman 1978) to test the null hypothesis that the coefficients estimated by the efficient random effects estimator are the same as those estimated by the consistent fixed effects estimator. The results (unreported) of this test failed to reject the null hypothesis. We therefore used the random effects model, with the generalised least squares (GLS) estimation methodology as the main analysis technique.
Results
Descriptive Results
Table 4 presents summary statistics of the variables used. On average, an MFI has seven board members; the minimum number of members is 2 and the maximum is 14. The average proportion of independent board members is 37 per cent, which implies that most boards are less independent and dominated by insiders. In terms of board diversity, we see that 25 per cent of board seats are held by females and 23 per cent by foreign members. Regulated MFIs make up 39 per cent of the MFIs in our sample, while 66 per cent are under some form of international influence, either in terms of their main donor/funder or other key relationships. The results also show that 54 per cent of the MFIs are managed by founders. The average age of the MFIs is 9 years and 76 per cent of them have their own legal status. Kenyan MFIs make up 34 per cent of the sample, Ugandan MFIs 30 per cent and Tanzanian 36 per cent.
Table 5 presents the correlations between the variables. First, we looked at correlations between the board structure variables. Although we argue that these variables are separate constructs, it is not surprising that they are significantly related, as they are all measures of different aspects of board structure. For example, the variable for ‘regulated’ MFIs is positively and significantly correlated with ‘board size’ (coefficient: 0.35, p < 0.01), while the variable for ‘founder-managed’ MFIs is negatively correlated with both ‘board size’ (coefficient: –0.31, p < 0.01) and ‘board independence’ (coefficient: –0.23, p < 0.01), as hypothesised. We also see significant correlations between board structure and control variables. For example there is a positive correlation between ‘board size’ and ‘age’ (coefficient 0.49), between ‘board size’ and ‘MFI size’ (coefficient 0.41) and between ‘board independence’ and ‘MFI age’ (coefficient 0.32). Because of these significant correlations, we controlled these variables in the regression models.
Descriptive Statistics
As we prepared for the multivariate tests, we turned to the question of multicollinearity among the independent variables. Correlation coefficients among the independent variables were rather low, with the highest being between ‘regulated’ and ‘legal status’ (coefficient: 0.51, p < 0.01). Kennedy (2008) holds that correlations need to be above 0.7 to detect multicollinearity between the two variables. None of the coefficients were this high but significant coefficients are a warning signal that multicollinearity problems may arise. Therefore, variance inflation factors (VIF) are used to test for multicollinearity. The results, as shown in Table 5, range between 1.27 and 2.21, suggesting that multicollinearity was not a problem in our study (Neter et al. 1996).
Pearson Correlations for Variables in the Regression Model
Source: Field results.
Notes: * p < 0.05; ** p < 0.01.
Regression Results and Discussion
Results of random effects GLS regression estimations for each dimension of MFIs board structure are reported in Table 6. Models 1–4 exhibit results for board size, board independence, gender diversity and diversity of nationalities, respectively. In addition, Table 7 reports the joint test using all variables simultaneously as a robustness check.
MFI Regulatory Status
The first hypothesis concerns the effect of regulatory status on board structure. Table 6, Model 1 provides support for hypothesis 1a, which relates regulation to board size. Regulated MFIs have larger boards than unregulated MFIs (coefficient: 0.09, p < 0.05). Interestingly, this is in line with Adams and Mehran’s (2003, 2008) findings for US banks. Despite the fact that unregulated MFIs have a broader set of stakeholders, regulated MFIs need larger boards to assist them by bringing intellectual knowledge to improve decision making, increase support for the MFI, and in turn boost organisational performance (Jackling and Johl 2009).
Model 2 provides support for hypothesis 1b, which deals with board independence (coefficient: 0.06, p < 0.10). This is in line with the corporate literature that argues that regulated organisations need the expertise and skills of independent board members (Hillman et al. 2000). However, we do not know whether these independent board members ‘really’ add expertise and skills, or whether they merely make the MFI ‘look good’ in the eyes of the regulators.
We found evidence of partial support for hypothesis 1c, stating that regulated MFIs have less diverse boards. In regard to gender diversity, model 3 shows support (coefficient: –0.02, p < 0.10) while diversity of nationalities is not supported in model 4 (coefficient 0.08, not significant). This suggests that regulated MFIs have fewer female board members. One possible explanation could be that regulated MFIs need to comply with regulatory requirements (banking and business expertise) that are less prevalent among female board candidates. This is also in line with studies from the corporate sector in developed countries, that a lack of high-level business experience reduces women’s chances of being appointed to boards (Ruigrok et al. 2007). The evidence, however, contradicts the agency and resource dependence theory, which advocate board diversity as a better tool for organisational performance (Walt and Ingley 2003). Based on these results, we argue that regulated MFIs have less incentive to recruit diverse board members.
Results of Random Effects GLS Regression
Results of Seemingly Unrelated Regression
International Influence
The second set of hypotheses looks at the relationship between international influence and board structure. Hypothesis 2a, addressing the relationship between international influence and board size, is supported in model 1 (coefficient: 0.16, p < 0.01). This suggests that, when international actors from the global north are substantially involved in an MFI, they opt for larger boards. We deduce that they do so in order to provide the funding, skills, networks, technical assistance and information needed for long-term MFI survival (Mersland et al. 2011). Furthermore, as international supporters provide resources to MFIs, they also need to sit on their boards to control the usage of their resources, which therefore enlarges board size.
Contrary to the prediction in hypothesis 2b, model 2 shows a significantly negative effect of international influence on board independence (coefficient: –0.09, p < 0.05). One possible explanation for the lower level of board independence when the MFI has international partners could be that MFI board insiders could be helpful in providing organisation-specific information and enhanced decision making, as argued in the case of the corporate sector by (Wagner et al. 1998). Secondly, the results suggest that MFIs that are influenced by international actors could potentially also be influenced by their CEO founders and, therefore, these CEOs prefer insider-dominated boards. To cross-check this, we re-ran the regression to test whether there is an interaction effect between ‘international influence-founder CEO’ and ‘board independence’. The results show that, out of the 66 per cent of MFIs in our sample that are internationally influenced, 38 per cent are managed by founders. By re-running the model with this interaction variable (unreported), the regression results support the existence of such an effect (coefficient: –0.06, p < 0.05). We therefore conclude that internationally influenced MFIs have less board independence, possibly due to the information advantage of board insiders and the CEO founders’ strong position within MFIs.
Hypothesis 2c, regarding the relationship between international influence and board diversity, is partially supported. Model 3 does not show significant results in regard to gender diversity (coefficient: 0.06, not significant). This is surprising, as we expected international partners to advocate gender-balanced boards, as suggested by the literature and investors from the global North (CMEP 2005; Rhyne 2005). Model 4, however, provides statistically and economically significant results in relation to diversity of nationalities (coefficient: 0.23, p < 0.001). This is in line with the corporate literature, suggesting that international board members help to reduce cultural complexities and enhance monitoring (Aguilera and Jackson 2003).
Founder Management
Hypothesis 3 addresses the relationship between founder-managed MFIs and board structure. Model 1 provides support for hypothesis 3a, that founder-managed MFIs are associated with small boards (coefficient: –0.06, p < 0.10). In line with the findings in the corporate literature (Boeker and Karichalil 2002; Certo et al. 2001), we argue that founder-managed MFIs are associated with small boards due to their founders’ attachment to the organisation and their ability to control the organisation. One possible reason why founder CEOs opt for small boards could be that it is easier for them to influence small boards and be accountable to fewer board members. For example, when a CEO seeks the board’s consent, it is easier for him to speak informally to a small number of individual members.
Regarding the relationship between founder-managed MFIs and board independence, model 2 provides evidence in support of hypothesis 3b (coefficient: –0.07, p < 0.01). One explanation could be that founder CEOs believe that inside board members provide an additional source of MFI-specific information, which outside board members would not. Another possibility could be that founder CEOs prefer less board independence, as this makes it easier for them to maintain their power, influence and control (Boeker and Karichalil 2002). This was also shown by Nelson (2003), who found that corporate organisations managed by founders prefer less board independence, which results in less vigilant board monitoring and greater CEO influence.
Regarding the relationship between founder-managed MFIs and board diversity, model 3 provides evidence contrary to the relationship we propose in hypothesis 3c. We expected that founder-managed MFIs would be negatively associated with gender diversity. However, our findings show a significant positive relationship (coefficient: 0.05, p < 0.10). This suggests that founder-managed MFIs prefer board gender diversity, which could be explained by the founder’s desire to achieve the social goal of the MFI; he/she thus uses female board members’ skills to emphasise female outreach (Mersland and Strøm 2009; Strøm et al. 2010). The high level of openness to female board membership could also be a reflection of the high proportion of female founder CEOs in the industry. It is possible that female founder CEOs want to maintain their influence and find this easier to achieve if they invite their female friends onto their boards. To check this further, we ran a regression with an interaction variable for ‘female-founder CEO and gender diversity’. The unreported results (coefficient: 0.23, p < 0.001) show similar results to those described above, namely that female-founder CEOs are positively associated with board gender diversity.
In line with our prediction, model 4 provides evidence that founder-managed MFIs are associated with less board diversity in terms of nationalities (coefficient: 0.05, p < 0.05). Since diversity of nationalities might bring more independent monitoring of the CEO and potentially more complex and slower decision making, a powerful founder CEO might not see it as in his or her interests to advance international board membership.
In summary, our results show that there are multiple associations between the unique characteristics of MFIs and their board structures. We found evidence that internationally influenced and founder influenced East African MFIs tend to produce a board structure that is more suitable for accessing resources mainly from insiders than for monitoring and accessing resources from outsiders. This is not the case for regulated MFIs. We also showed the significant impact founder CEOs have on board structure, which suggests that these leaders have substantial power within the organisations and with their key stakeholders. These findings might reflect the youth of the industry and organisational life-cycle effects (Filatotchev et al. 2006; Roche 2009) suggesting that as the industry matures the effect might change. As is also evidenced from the effect of our control variables, age is positively associated with board independence and board diversity.
The interaction effect of founder CEOs and MFI life-cycle was further confirmed when we cross-checked and re-ran the regressions (unreported), after dividing the sample into two subsamples based on the median MFI age of eight years. The CEO-founder effect is stronger in the sub-sample of MFIs under eight years old than it is in MFIs over eight years old. This implies that, as MFIs grow older, CEO founders are more likely to step down and MFIs will tend to attract more independent and diverse boards.
Robustness
As part of our robustness checks, we re-ran the models. First, we used panel data estimation with the seemingly unrelated regression (SUR) methodology (Greene 2008; also previously applied to microfinance by Mersland et al. 2011), a procedure that takes account of possible correlations among dependent variables. We also performed a Breusch-Pagan test to check the extent to which the residuals in the SUR regression were independent. The results are reported in Table 7.
A common challenge in corporate governance research is the issue of possible reverse causality. Our second robustness check was to re-run all models using instrumental variables with a two-stage linear regression (2SLS). The methodological concern here was the possibility of an omitted unobservable variable that affects both independent and dependent variables (Adam and Ferreira 2009). The literature shows that many of the variables of interest in corporate governance investigations are not truly exogenous, that is they are determined completely outside the model systems but are endogenous (Hermalin and Weisbach 2003). Carter et al. (2010) suggest that we can use lagged performance variables as instruments even though they are not completely outside the system of equations. We followed their suggestions and use lagged variables of MFI performance. We used lagged operational self-sufficiency 3 and lagged number of customers 4 as our instruments. We tested the correctness 5 of these instruments (Cheung et al. 2007) and ran a 2SLS (Wooldridge 2006). The results are presented in Table 8.
The results in both Tables 7 and 8 are to a large extent similar to the main results of most of the hypothesised relationships (see Table 6). We showed that regulated MFIs are associated with large boards and a lack of board gender diversity, and that internationally-influenced MFIs have larger and less independent boards. Also, international influence has no significant association with board gender diversity, but does have a positive relationship with diversity of nationalities. Overall, our results show that there is a need to address the three dimensions of MFI board structure individually and that our random effects model is efficient.
Results of a Two-stage (2SLS) Regression
In addition to robustness checks explained earlier, we re-run regression using SUR to test whether the board structure aspects have effect on the social performance of MFIs in Table 9. We specifically look at the effect of structure variables in enabling MFIs to reduce loan defaults which is a sign for good loan repayments. We also examine the extent to which these board structure variables enable MFIs to reach out to the poor clients and hence improve their socio-economic status. We use portfolio at risk (PaR 30) for testing the repayment performance and percentage of women clients for testing the outreach to the poor performance. Results as presented in Table 9 show that board structure has effects on the MFI performance. In terms of defaults, we evidence that large boards bring high defaults implying less loan repayment. This provides support for the literature that necessitates small boards to perform better than larger boards (Mersland and Strøm 2009). We further evidence independent boards to bring high repayments and low defaults. This implies that MFIs should value the expertise and experiences brought by independent members regardless of the nature of the microfinance industry.
Results of Seemingly Unrelated Regression between Board Structure and MFI Performance
In terms of improvement in socio-economic status of MFI clients, we evidence a strong relationship between board structure variables and percentage of women clients reached. We observe small, independent, gender and foreign represented boards to have high effect on the MFIs ability to reach out to the poor and hence improve their livelihood. These results provide strong support for the effect of MFIs’ board structure and composition on the performance of MFIs.
Conclusions and Implications
This study examined the determinants of board structure in MFIs. Guided by the agency and resource dependence theories, we examined three unique characteristics of MFIs: regulation status, international influence and founder management. Using a sample of 63 East African MFIs, we showed the relationship between these characteristics and three dimensions of board structure: board size, board independence and board diversity (gender and nationality). We used a mixture of analytical procedures to obtain valid and robust results.
The results showed that there are significant associations between MFI characteristics and board structure. We found that MFIs that are regulated have larger boards, high board independence and less gender-diverse boards. The lower diversity in regulated MFIs might be due to the expertise demanded by the regulators or it could reflect discrimination against females. This needs to be further explored in future studies. Similarly, MFIs that are internationally influenced have larger boards, more international boards, and less independent boards. Past research highlights that MFIs’ international partners establish connections with MFIs in order to provide them with resources, such as funding, and international experience, through technical assistance and advice (Mersland et al. 2011). Founder-managed MFIs show similar characteristics to those demonstrated in the corporate founder-CEO literature (Certo et al. 2001). These results have various implications.
Implications for Theory
In this study, we applied the agency and resource dependence theories as complementary theories in studying the microfinance industry. Past studies on board structure mostly applied these theories in relation to large for-profit organisations in developed countries (Jackling and Johl 2009; Linck et al. 2008). We showed that these theories can be applicable to research on a young entrepreneurial industry like microfinance, about which corporate governance research is still very limited (Labie and Mersland 2011).
To our knowledge, we are among the first to highlight the relevance of the resource dependence theory for examining corporate governance in MFIs, though several studies have used the agency theory (Hartaska 2005; Mersland and Strøm 2009). The resource dependence theory is particularly useful for understanding microfinance, due to the common dual objective of MFIs (financial sustainability and having a social mission), and the need for MFIs to obtain access to key stakeholders (for example fund providers and community leaders). In addition to understand the behaviour and structure of MFIs, there is a need to first understand the context and environment of the microfinance industry as stated by Pfeffer and Salancik (1978: 1).
Agency-theory-based studies commonly suggest that small and independent boards are better at monitoring organisations, particularly their CEOs. We found that CEO founders do not like such ‘controls’ and that the CEOs’ ability to affect the choice of board structure indicates that they are able to exercise power over MFIs. We therefore argue that, despite the commonly emphasised monitoring benefits of having small and independent boards, the social mission attachment of the founder CEO might make ‘strong’ monitoring less necessary, as is the case in founder-managed for-profit organisations (Randøy and Goel 2003).There is still relatively limited corporate governance research from emerging economies and Africa in particular (Rossouw 2005). This study indicates that the agency theory and resource dependency theory are useful for understanding boards in an emerging market context such as East Africa.
Implications for Practice
The CSFI (2011) reported that corporate governance is one of the major challenges facing African MFIs. One highlighted challenge is the lack of board independence. We found this lack in both internationally-influenced and founder-influenced MFIs, but we argue that this lack of board independence is due to the industry’s newness and we suggest that this will need to change as MFIs grow. As the CEO of one East African MFI put it, ‘four years ago, the board’s role was largely advisory. Now, stakeholders expect it to take a more active role, especially in monitoring and risk control. We need to draw more on the specialised expertise of outside members’ (CMEF 2005). MFIs should also be aware that board insiders are a source of organisation-specific information (Wagner et al. 1998) and help with the strategic direction of the MFI. However, we suggest that their presence on boards should not be so overwhelming as to make monitoring difficult.
Another practical implication relates to regulation. Currently, some public policy suggests that MFIs should change from operating as NGOs on a not-for-profit basis to being regulated and operating on a for-profit basis (Hartaska and Nadolnyak 2007). MFIs should be aware that this type of change will have implications for the board structure, not only through regulatory requirements, but also because the owners/trustees (or other legal bodies in charge) will have to focus on recruiting board members who can assist with monitoring and provide access to resources.
In this study, we found evidence of the strong influence of CEO founders on MFI board structure. We urge MFI founders, as their organisations grow, there is a need to make room for independent boards, so that the CEO does not limit the MFI’s access to new people with new ideas (Boeker and Karichalil 2002). There is some evidence in the corporate literature that some founders might stay on too long (Filatotchev et al. 2006; Wasserman 2003), and the owners/trustees of MFIs should be aware of this challenge and be ready to replace CEO founders if necessary.
With the financial crisis in the global North, the ability of international partners to support MFIs might diminish and MFI governance structures should reflect this risk. Specifically, MFIs should seek to recruit board members who provide strong links with multiple stakeholders, so that they can work more independently from past or present major donors/owners.
Limitations and Direction for Future Research
This study has several limitations. First, we used only a few dimensions of board structure. There are other board structure dimensions that we did not consider, and which probably would be useful to consider in research on MFIs. Dimensions such as the education, age and expertise of board members (e.g., in Ruigrok et al. 2007) and CEO duality could be considered in future studies.
The independent variables (regulations, international influence, and founder management) chosen for this study might be explored in more detail: for example, the cultural and institutional distance between the MFI and its international partner/network, the specific background of the CEO founder, the content of the regulations, and finally, a more detailed picture of the MFIs’ customer environment. In addition, there might be other determinants of board structure that should be considered in future studies, for example the nature of the corporate governance environment (such as the legal regime), and the level of financial development of the country.
In this study, we found, surprisingly, that international influence brings less board independence. Corporate governance policies, in general also in the microfinance context, advocate ‘greater’ board independence in order to reduce information asymmetries between owners/trustees and management. Future research should therefore address this paradox.