Board Composition and its Effect on the Outreach and Financial Sustainability Objectives of Microfinance Institutions in East Africa

(1)

University of Groningen

Faculty of Economics and Business

Master Thesis International Financial Management

Board Composition and its Effect on the Outreach and

Financial Sustainability Objectives of

Microfinance Institutions

in East Africa

Jorrit Steenbeek

Sint Lucasstraat 7

9718LP Groningen

S1778382

j.steenbeek@student.rug.nl

(2)

Master thesis Jorrit Steenbeek Page 2

Acknowledgements

Before you lies the final product of my master study International Financial Management at the University of Groningen. I grew up in East Africa and first came into contact with microfinance during my high school years at the International School of Uganda. I realized there that small loans for poor people to start their own business could make a huge difference in their lives.

In February 2013 I started with this thesis after my interest in microfinance was reignited during the master thesis evening in December 2012.

Hereby I would like to thank a number of people who helped me with this study.

First of all, I would like to thank my supervisor, Prof. Dr. Niels Hermes, for his continuous advice, feedback, constant willingness and patience to answer all my questions, while I was writing my thesis. Next I would like to thank Dr. Neema Mori for allowing me to use the data she and her colleagues collected on microfinance institutions in East Africa; without this data this study would not have been possible. Finally I would like to thank my friends and family. Especially my parents and brother for their continued support and their confidence in me, even when times became tough and deadlines started rapidly approaching.

(3)

Abstract

Aim: This paper aimed to discover whether there is an effect of MFI board diversity on the achievement of the outreach and financial sustainability objectives of MFIs.

Design: A stepwise model was used to analyse data originating from MFIs in East Africa, starting with investigating the possible relationship between board diversity and outreach and financial

sustainability. The second step was to determine whether possible underlying determinants of this relationship were board processes and board tasks. Thus, a linear model following the structure: board diversity, board processes, board tasks and finally MFI performance was used.

Results: A negative and significant relationship between diversity and outreach and financial sustainability was found. No relationship was found between board diversity and board processes. Only a very weak relationship was identified between board processes and board tasks. Finally, no significant relationship was found between board tasks and MFI performance.

(4)

1. Introduction

In the past two decades the microfinance industry has experienced explosive growth. Only 10 million households received the services of microfinance institutions in 1997, this number grew to more than 100 million in 2007 (Assefa et al. 2012). Furthermore, the United Nations hailed 2005 as the International Year of Microcredit and in 2006 Muhammad Yunus, founder and CEO of the Grameen Bank in Bangladesh, received the Nobel Peace Prize.

Despite this impressive growth, there are only few microfinance institutions (MFIs henceforth) which are financially sustainable. There is a high dependence on government subsidies and donations from non-government organizations (NGOs henceforth) and other donors (Chamlee-Wright 2005). Furthermore, there also seems to be a more ominous side to microfinance, as the world vividly discovered in 2010 when news of microfinance related suicides were reported in the Indian state of Andhra Pradesh. Investigation into these suicides showed that Indian borrowers were confronted with annual interest rates of 24 – 30% (Bisaws 2012).

The above paragraphs are not written to praise nor criticize the microfinance industry, it is a relatively young, vibrant and intensely fascinating sector of the economy. The potential to eliminate poverty may one day be within reach. However, this does mean that the microfinance industry needs to make a number of very important choices and decisions. These choices and decisions, made by the people who actually control the MFIs are central to the investigation of this paper. The individuals who manage an MFIs actions are bound by a system of corporate governance structures where many influences of the different stakeholder interests and objectives must be achieved and confronted. Governance of MFIs is important since it may mean the difference between efficient and sustainable operations or not (Mori and Mersland 2011).

1.1 Problem Definition

Stakeholder interests and objectives are amassed and represented in the board of directors. For this reason, this paper focuses solely on the board of directors. In the standard, industrial model, of boards of directors; management, shareholder, and sometimes employee representatives meet to discuss their respective party interests. An MFI board, on the other hand, can be composed of parties representing management, shareholders (if any), investors, donors, customers, employees and independent directors (Hartarska 2005, Varottil 2012). Interaction between board members can become especially

complicated in such a setting.

(7)

Master thesis Jorrit Steenbeek Page 7 Thus, the research question of this paper is defined as:

Does the diversity of an MFIs board of directors affect the outreach and financial sustainability objectives of MFIs in the East African countries Uganda, Kenya and Tanzania?

The relevant literature on corporate governance identifies a number of plausible underlying variables which could explain a possible effect between board diversity and outreach and financial sustainability. First of all the literature identifies three board processes: Conflict (Simmons and Peterson 2000, Jehn and Mannix 2001), communication (Shaw 1981, Daft and Lengel 1984) and agreement (Hrebiniak and Snow 1982, Iaquinto and Frederickson 1997). Furthermore, previous research identifies two important board tasks, namely the monitoring task (Hillman and Dalziel 2003) and the advisory task (Mintzberg 1983, Baysinger and Hoskisson 1990) assigned to the board of directors. If this study finds a relationship between board diversity and the outreach and sustainability objectives of an MFI, it is important to find an answer to the following questions as well:

I. What effect does the level of board diversity have on the board processes: conflict, communication and agreement?

II. What is the relationship between board processes and the monitoring and advisory tasks of the board?

III. Is there an interaction effect between board diversity and board processes, and if so, what effect does this have on the monitoring and advisory tasks of the board?

IV. What effect do the monitoring and advisory tasks assigned to the board of an MFI have on the outreach and financial sustainability objectives of an MFI?

After the effect, if any, between diversity and MFI performance has been identified, this paper moves to attempt to explain what possible factors underlie this effect. Thus, three board processes and two board tasks are identified which could explain the relationship between board diversity and MFI performance measures.

1.2 Relevance

From an academic perspective this research is relevant because very little research has been done on the effect of board diversity on the main objectives of an MFI. Hartarska (2005) was one of the first to look at the possible effect different stakeholders on the board of an MFI have on the outreach and financial sustainability objectives of an MFI.

(8)

Master thesis Jorrit Steenbeek Page 8 and other financial services they need to increase their economic activity and consumption. MFIs have been argued to be a large player in the alleviation of poverty and therefore, play an important role in the increased economic activity of developing countries. Thus, the efficient structuring of a board of directors of an MFI may lead to more efficient microfinance services to the poor.

This paper adds to the current literature regarding corporate governance in microfinance. Differently with respect to previous research, this study uses a factor analysis to combine relevant variables into three single board processes and two board tasks, namely the monitoring and advisory tasks.

This paper is structured as follows. The second chapter intends to define and clarify the concepts used in this paper by reviewing the relevant literature. These concepts include microfinance, MFI performance, board diversity, board processes and board tasks. Furthermore, after the review of the literature this section formulates the hypotheses about the expected relationships between the above concepts. The third chapter will present the sample and research methodology. The fourth chapter will present the results obtained from the data and provide a discussion about these results. The fifth chapter will present the conclusion, limitations and recommendations.

2. Literature Review and Hypotheses

2.1

Microfinance and Microfinance Institutions

Microfinance is a concept devised as a method to alleviate poverty by providing inexpensive loans and credit to the poor, enabling them to carry out income-generating activity (Varottil 2012). Through microfinance the poorest citizens gain access to financial services which they would not have been able to obtain from the formal financial sector because they do not have the necessary assets to put up as collateral for a loan (Armendáriz and Morduch 2010). Furthermore, the screening and monitoring costs and the enforcement of contracts could make the costs for commercial banks too high to make lending to the poor a profitable activity (Hermes and Lensink 2007). For the purposes of this paper the term microfinance embraces the efforts to provide small loans to low-income households, collect savings from low-income households and to provide insurance (microinsurance) to the citizens neglected by the formal financial sector (Armendáriz and Morduch 2010).

According to the latest estimates made by the World Bank, updated in 2012, there were 1.29 billion people living on or below the $1.25 poverty line in 2008. 2.47 billion people had a

(9)

Master thesis Jorrit Steenbeek Page 9 among the poorest people on earth when they joined their respective programs (Maes and Reed 2012). These statistics indicate that the microfinance initiative is an essential part of a developing economy.

The microfinance industry is quite diverse with regard to the organizational types of the firms in the industry. MFIs are structured as NGOs, banks, Credit Union/ Savings and Credit Cooperatives, NGOs transformed into Incorporated MFIs, licensed Mutual-Ownership Banks, licensed Equity Banks and non-bank financial institutions (NBFIs henceforth) (Hartarska 2005, Greuning et al. 1998). The literature on corporate governance focuses mainly on the standard concepts and principles which do not adequately address the unique issues specific to the microfinance sector. Some have even argued “for a paradigm-shift that necessitates examination of corporate governance in microfinance

companies through an altogether different lens”(Varottil 2012). The reason conventional corporate governance theory does not adequately address the MFI board setting is due to magnitude of

importance placed on the agency relationships between managers and shareholders. To this end, this paper only investigates three types of MFI: NGOs, commercial/community banks and non-bank financial institutions.

Among MFIs there are not only different types of organizations, they also differ greatly in the types of loans offered, for example, group loans or individual loans. To illustrate, the Badan Kredit Desa of Indonesia serves clients with an average loan balance of $38, compared to the $101 average loan issued by the Grameen bank in Bangladesh (Morduch 2000). This is not limited to just loans, but also interest rates. The real loan interest rate charged by the Grameen bank is around 20% (Ray 1998) whereas customers of the Badan Kredit Desa may pay a real interest rate approaching 50% per year on 3-4 month loans (Morduch 2000). Furthermore, the Grameen bank lends to groups of individuals without a collateral requirement, whereas the Bank Rakyat Indonesia lends to individuals and does require collateral (Ray 1998). Such differences are apparent due to specific country characteristics and population demographics. Furthermore, an MFI can choose to serve an upper segment of the poor, individuals who are not the poorest of the poor, allowing for larger loans, or an MFI can target the poorest of the poor.

2.2

Corporate Governance and Corporate Governance in the Microfinance

Industry

Corporate governance is a notion which includes a myriad of different dimensions. Depending on the subjective view of the writer there are many different views that can be found in the literature for corporate governance. For instance, from a management perspective corporate governance is about the actions and responsibilities of the board of a company – what boards do and how they are

composed (Thomsen 2008). Then again, from an investor and finance perspective corporate

(10)

Master thesis Jorrit Steenbeek Page 10 Corporate governance can also be taken through the perspectives of lawyers, accountants, politicians, sociologists and psychologists, each differing in their approach.

Traditionally, corporate governance has been investigated through the agency theory lens. In this respect an agency relationship can be defined as a contract under which one or more persons (the principal(s)) engage another person (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent (Jensen and Meckling 1976).

Agency theory holds that the relationship between a firm and its shareholder is sacrosanct.

However, the agency theory construction fractures when it is applied to the microfinance sector. There are two reasons for this. Firstly, MFIs are faced with the important issue of which objective, outreach or sustainability, has greater significance. On the one hand, it should be recognized that the interest of borrowers (the poor) as principal stakeholders is vital. On the other hand, regarding the sustainability of an MFI, generating profits, rewarding investors and very rarely, in the case of listed MFIs,

maintaining stock prices are also essential activities (Hartarska 2005; Mersland and Strom 2008; Mori and Randoy 2011; Mori and Mersland 2011;Varottil 2012).

Secondly, with regard to the board of directors of an MFI, the composition of the board of an MFI differs to that of industrial firms and banks. The board of an MFI is comprised of parties with differing and sometimes conflicting interests. Donors, equity investors such as founders, commercial investors and social investors, debt financiers such as banks and other partners, employees and even customers may all have board representation on the MFI board of directors (Mori and Mersland 2011; Varottil 2012). In other words, an MFIs board is subject to a higher level of interest diversity compared to conventional boards. This may lead to a balancing act regarding the two main objectives of the MFI, outreach and financial sustainability, due to the differing objectives inherent to the different

stakeholders represented on the board. According to agency theory the central objective of the firm is the maximization of shareholder wealth; furthermore, agency theory looks specifically at the

relationship between shareholders and management (Freeman 1984; Thomsen 2008, Varottil 2012). This does not fit with the board model used by an MFI.

In short, governance is about achieving corporate goals. The first goal of an MFI is to reach as many clients as possible in the poorer echelon, the second goal of an MFI is to achieve financial sustainability (Mersland and Strom 2009). However, the existing corporate governance framework surrounding MFIs focuses almost wholly on the relationship between the MFI and its capital providers (Varottil 2012), thus, lacking focus on the relationship between the MFI and its

„bottom-of-the-pyramid‟ (BoP) customers.

(11)

Master thesis Jorrit Steenbeek Page 11 through which customers and employees ensure that their interests will be represented and met. This definition does assume that all parties focus solely on their own interests when represented on the board of an MFI.

The literature has shown that the many different stakeholders on the board of an MFI have either a preference to achieve an outreach objective or a sustainability objective. Thus, the board

representatives of these stakeholders will want the MFI‟s focus to be towards their preferred

objectives. Therefore, a thorough understanding of outreach and financial sustainability is first needed. The next two sections outline the relevant literature written on outreach and financial sustainability.

2.3

Outreach

Two distinct dimensions of outreach are prevalent in the literature: depth of outreach and breadth of outreach. Schreiner (2002) describes depth of outreach in terms of social welfare: “depth is the weight of a client in the social-welfare function”. Similar definitions have been used by Thys (2000), Hermes and Lensink (2011) and Mersland and Strom (2008) which argue that depth of outreach refers to the socioeconomic level of clients. Finally, Navajas et al. (2000) define depth of outreach as “the value that society attaches to the net gain from the use of microcredit by a given borrower”. Thus, depth of outreach essentially attempts to measure how poor MFI clients are and how societal welfare is impacted by providing loans to poor borrowers. When an MFI is serving relatively „richer‟ poor clients depth of outreach is lower than when the MFI is serving relatively poorer poor clients. In the literature there have been a number of proxies which have been used to measure the depth of outreach. Hartarska (2005) uses the average loan size divided by annual GDP per capita to measure depth of outreach. Navajas et al. (2000) argue that poverty is a good proxy for depth and use the Index of Fulfilment of Basic Needs as a measure of depth of outreach. Hermes and Lensink (2011) and Luzzi and Weber (2006), proxy the depth of outreach of an MFI by measuring the average loan balance and percentage of female borrowers.

The literature defines breadth of outreach as simply the number of clients of an MFI, and this measure has been extensively used in previous research as a proxy (Navajas et al. 2000; Thys 2000; Schreiner 2002; Hartarska 2005; Luzi and Weber 2006; Hermes and Lensink 2007; Mersland and Strom 2008; Hermes et al. 2009; Hermes and Lensink 2011). Breadth of outreach is not necessarily associated with clients who are the poorest of the poor, but with all the clients of an MFI. According to Schreiner (2002), breadth matters due to budget constraints; “the wants and needs of the poor exceed the resources earmarked for them”. In other words, there are many poor but there are not enough funds to serve all of the poor (Navajas et al. 2000).

(12)

Master thesis Jorrit Steenbeek Page 12 lower gross revenues can be achieved due to the smaller average loan sizes. This argument is

supported by Cull et al. (2009). Consequently, institutions which seek financial sustainability will direct their efforts away from the poorest of the poor and more towards the „richer‟ poor. This idea is supported by Cull et al. (2007) who found an important trade-off between the breadth and depth of outreach. On the other hand, Navajas et al. (2000) found that Banco Sol in Bolivia was reaching 1400 more of the poorest of the poor than any of the other four MFIs active at the time. Two of these institutions were actively pursuing the objective of deep outreach. Furthermore, Rosenberg (1996) argues that self-sustainable MFIs with wide breadth may be able to serve as many of the very poor clients as MFIs with high depth. Rosenberg (1996) illustrates this using sustainability focused credit unions in Colombia which served more poor clients than depth focused village banks in Costa Rica and Guatemala. Finally, Morduch (2000) argues that breadth cannot be judged by itself, wide breadth may compensate for shallow depth and deep depth may compensate for thin breadth. This indicates that the trade-off between depth and breadth of outreach is not as straightforward as may be specified in other research and that depth of outreach and breadth of outreach can actually be compatible with each other.

2.4

Financial Sustainability

Financial sustainability refers to the ability of an MFI to cover its costs (Hartarska 2005). A different angle to sustainability is taken by Navajas et al. (2000), who define sustainability as “permanence”, arguing that the social goal of an MFI is “not to have a sustainable organization but rather to maximize social value less social costs discounted through time”. Sustainability is usually measured using accounting based measures such as Return on Assets (ROA henceforth) (Hartarska 2005; Bogan 2008; Mersland and Strom 2008; Mersland and Strom 2009; Cull et al. 2009; Assefa et al. 2012), Return on Equity (ROE henceforth) (Conning and Morduch 2011), Operational Self-Sufficiency (OSS henceforth) (Hartarska 2005; Luzzi and Weber 2006; Bogan 2008; Cull et al. 2009; Mersland and Strom 2009; Assefa et al. 2012) and Financial Self-Sufficiency (FSS henceforth) (Cull et al. 2007; Cull et al. 2008; Cull et al. 2009; Conning and Morduch 2011)

(13)

Master thesis Jorrit Steenbeek Page 13 also relatively resistant to financial manipulations by a firm‟s management (Gerhart and Milkovich 1990).

ROE is the least commonly used measure of financial sustainability in the microfinance literature and it is defined as “the firm‟s earnings before extraordinary items and discontinued operations divided by the average common shareholders‟ equity” (Lambert and Larcker, 1987).

OSS is a percentage which indicates whether an MFI has earned enough revenue to cover its costs, in other words, the operational self-sufficiency level of 100% or more, and it is defined by the following formula (Bogan 2008):

𝑇𝑜𝑡𝑎𝑙 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑅𝑒𝑣𝑒𝑛𝑢𝑒

𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝐸𝑥𝑝𝑒𝑛𝑠𝑒 + 𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔 𝐸𝑥𝑝𝑒𝑛𝑠𝑒 + 𝐿𝑜𝑎𝑛 𝐿𝑜𝑠𝑠 𝑃𝑟𝑜𝑣𝑖𝑠𝑖𝑜𝑛 𝐸𝑥𝑝𝑒𝑛𝑠𝑒

OSS is typically used as the standard measure of MFI performance (Bogan 2008, Mersland and Strom 2009). As a measure of sustainability it can be argued that it is more relevant to the context of the microfinance sector since the objectives of MFIs are not solely profit driven like the objectives of corporate firms. Furthermore, OSS is a superior measure compared to ROA because ROA does not necessarily incorporate the value of donations, subsidies and inflation (Hartarska 2005). OSS can be affected by a number of activities by the MFI. Firstly, OSS can be affected by unreported subsidies with regard to operations. Secondly, cost allocation can be prepared in such a way that credit

operations look more sustainable than they actually are. Finally, allocated costs to subsidiaries or when donors meet certain costs, such as payment of consultants, OSS is also affected (Arunachalam 2006a).

FSS is an adjusted measure of OSS; it also measures an institutions ability to create adequate revenues to cover its costs (Cull et al. 2007). It incorporates adjusted figures to account for different types of subsidies so that it approximates the returns of an MFI in the absence of subsidized funding (Cull et al. 2009). FSS has an advantage over other measures such as ROA and ROE because it presents a more inclusive rundown of inputs and outputs (Cull et al. 2007). For an illustration of FSS refer to the appendix, section 7.1. The only difference between FSS and OSS is the inclusion of the adjusted cost of capital.

One shortcoming of FSS is that it only provides a snapshot of the current conditions. It is not able to indicate how flexible the institution is, or which strategies the MFIs management would be able to implement if subsidies are not available anymore (Cull et al. 2009). Furthermore, the first two disadvantages of the OSS ratio are also inherent to the FSS ratio (Arunchalam 2006b).

2.5

Board Diversity

(14)

Master thesis Jorrit Steenbeek Page 14 the board may prefer that the MFI focuses on outreach, whereas equity investors may prefer to achieve financial sustainability. The party in a minority position will have difficulty achieving their goals.

One perspective on diversity highlighted in the literature is based on similarity attraction and social categorization. This perspective suggests that diversity features such age and values, which are prominent among members in a unit, limit social integration, may foster conflicts and turnover and decrease unit morale and performance (Harrison and Klein 2007).

Harrison and Klein (2007) make a distinction between three types of diversity; separation, variety and disparity. Separation illustrates differences among unit members in their position on a horizontal range, in other words, separation reflects the stand point or position and distribution of the members‟ values, beliefs, attitude or orientation. Variety describes differences among unit members from different categories, reflecting access to unique sources of knowledge. Finally, disparity describes the differences among unit members in their portion of a valued resource. The definition of diversity in this paper follows Harrison and Klein‟s (2007) definition of separation. Maximum separation occurs when a unit is divided into two divergent camps. As separation increases from minimum to maximum, unit members grow increasingly polarized in their placement along a horizontal continuum. This does not mean that they are completely different from one another; it means that they split into opposing subunits.

Following Hartarska (2005) this paper argues that there are five types of stakeholders in an MFI board, namely; donors, employees, customers, (equity) investors, and independent directors. This paper also assumes that each type of stakeholder has a preference towards achieving outreach or financial sustainability and thus, maximum separation is achieved.

2.5.1 Donors

In the microfinance community a general consensus prevails on two crucial points, first, donor support will continue to play a vital role in the future of microfinance in the immediate future and, second, sustainability will be a key component for the continuity of the microfinance sector (Chamlee-Wright, 2005). This indicates that donors are extremely important for the microfinance industry. An illustration of the importance of donors to the microfinance industry is provided by Morduch (1999), who shows that if small adjustments were to be made to „standardise‟ Grameen Bank‟s accounting practices it would have resulted in losses of nearly $18 million between 1985 and 1996, rather than the $1.5 million profits. Furthermore, donations are considered as income, but if donations are not taken into account Grameen would have incurred $34 million in losses between 1985-1996.

The importance of donors to the microfinance industry gives them a substantial influence with regard to the actions of an MFI. Donors are good at providing funds, measuring progress, and

(15)

Master thesis Jorrit Steenbeek Page 15 The preferences of donors is to serve the most risky clients, the poorest of the poor (Hartarska 2005). Thus, donor priority in the microfinance industry is outreach. Mori and Mersland (2011) argue that donors are more interested in social performance and that donor objectives are to increase outreach to the poor. With their considerable influence on MFI actions, through board representation, donors will push for increased outreach within an MFI. Hartarska (2005) concludes that replacing a single board member with a donor would increase the service of poorer clients by 35%. However, her findings also indicate that this exchange would lead to 35% fewer clients and 25% lower operating self-sustainability.

An interesting finding by Hartarska and Mersland (2012) is that donor pressure redirects attention from financial sustainability to outreach. In addition Morduch (1999) argues that donor representatives have the ability to raise funds, thus, discouraging practitioners to question how and where money will be spent, invested and lent. Consequently there is also a lower incentive to achieve financial sustainability (Hartarska 2005). With regard to fundraising by donors, Bogan (2008) finds support that reliance on donor funds eliminates the motivation for MFIs to operate efficiently, providing evidence that MFIs are not currently realizing efficiency from economies of scale. An alternative interpretation of this result is that donor representatives generally push for greater outreach at the cost of financial sustainability, which requires relatively high levels of efficiency.

Nevertheless, Mersland and Strom (2009) find that donors do not affect the outreach objective of an MFI. It must be noted, however, that stakeholder representation in the board of directors in Mersland and Strom data were „surprisingly low‟ (Mersland and Strom 2009), indicating that the effect of donor representation was very limited.

Donor representatives on the board of an MFI will advocate for the achievement of the social objective of the MFI, namely outreach.

2.5.2 Employees

The conventional corporate governance literature argues that employees are good at providing inside information about the current status of firm actions to the board of directors (Osterloh and Frey 2005; Andrés-Alonso et al. 2009). Particularly, credit officers are the foundations upon which the MFI is built (Mori 2010; Batillana and Dorado 2010). It can be beneficial for an MFI to have employees on their board since they are better informed about the issues and problems concerning the MFI (Mori and Mersland 2011).

(16)

Master thesis Jorrit Steenbeek Page 16 employee representation on the board and the outreach and financial sustainability objectives of the MFI.

Employees can be expected to be driven by the social goals of an MFI due to the target clients of MFIs (Varottil 2012). Holding a non-profit status where social investors, donors and clients are represented can commit the MFI to a social mission in ways that may help attract socially minded staff, drawn by the chance to create social change and willing to work for lower wages (Conning and Morduch 2011). On the other hand, one can follow the agency theory perspective which argues that there will be fewer agency costs when the interests of employees and shareholders are aligned. This can be seen during the commercialization process of an MFI where the interests of employees and shareholders can be aligned by lowering salaries and allowing the bulk of employee compensation to consist of stock options and shares (Varottil 2012). Evidence from India indicates that generous compensation terms are given to senior MFI employees when an MFI has reached its full

commercialization potential and becomes listed on a stock exchange (Varottil 2012). However, few MFIs are actually listed, therefore, the incentives of stock options are limited to a small number of active MFIs and only the senior employees of these MFIs benefit from these financial incentives (Varottil 2012). This would indicate that only senior employees should be driven by financial incentives and therefore, only senior employees should prefer the MFI to focus on financial sustainability. Varottil (2012), also argues that a drawback of such a compensation system is that it creates a short-term mentality within the MFI, which is not beneficial for the long-term sustainability of the MFI.

Empirical evidence shows that the proportion of employees on the board of directors of an MFI is negative for both social and financial performance measures (Hartarska 2005). The only statistically significant effect Hartarska (2005) finds is with regard to the ROA performance measure. This result indicated that on average the replacement of one director with an employee would lower ROA by 6.6%. Hartarska concludes by stating that MFI boards with a higher proportion of employees have worse financial results. With the above in mind:

Employees on the board of the MFI will advocate for the achievement of the social objective of the MFI, namely outreach.

2.5.3 Customers/Clients

(17)

Master thesis Jorrit Steenbeek Page 17 It is important for an MFI to include representatives of clients on their board of directors so that strategic decisions match the needs of the clients and objectives (Mori 2010). Client

representation on the board of directors of an MFI is unusual (Mersland and Strom 2009;Varottil 2012) because BoP customers may not fully understand their own financial circumstances and abilities since a vast majority of them are uneducated (Mersland 2011; Mori and Mersland 2011). However, Pischke (2002) argues that microfinance customers know what they want and possess the knowledge to make certain decisions. Furthermore, MFIs are increasingly operating as market- (customer-) driven and not only product-driven organizations (Woller 2002).

The evidence on client representation on MFI boards with regard to the effect on outreach and financial sustainability is limited and relatively ambiguous. Assefa et al. (2012) argue that the

significant growth of the microfinance sector has led to increased competition for clients and markets and to a certain degree of saturation of microfinance suppliers.

On the other hand, Hartarska (2005) finds, interestingly, that clients on the board of directors of an MFI affect financial sustainability positively. The results show that operational self-sustainability would improve by 27% when a board member is replaced by a client and this would be at the expense of the depth of outreach. A possible rationale for this result is that client board members feel that the objective of the MFI should first be the service of wealthier borrowers and the poorest of the poor should gain access to the services of the MFI when the MFI is financially sustainable. This argument is supported by the findings of Navajas et al. (2000), who found that Banco Sol, in Bolivia, was pursuing financial sustainability and yet still reached more of the poorest of the poor than institutions actively pursuing deep outreach. Furthermore, Mersland and Strom (2009) argue that customer

representation on MFI boards help provide better information and thus improve financial performance. Most of the evidence points to the idea that client representatives on the board of an MFI have a preference to achieve the financial sustainability of an MFI, therefore:

Customers on the board of an MFI will advocate for the financial sustainability of the MFI.

2.5.4 Investors

(18)

Table 2.1: Capital structure of Different Types of MFI

Type of MFI Deposits Commercial

Borrowing

Equity Donations

Non-Commercial Borrowing Commercial MFIs 71% 13% 13% 2% 1% Non-bank Financial Institutions 21% 28% 18% 23% 11% NGOs 10% 26% 8% 39% 16%

Source: Cull et al. (2008)

Access to capital from investors is advantageous to MFIs because investors address initial funding requirements and they can be relied on to provide ongoing funding in the form of additional capital. Commercial investors, such as private equity funds and venture capital firms provide capital to the firms in their portfolio and they have now recognized the microfinance industry as an attractive investment opportunity (Varottil 2012). However, this means that institutional investors who do invest in the microfinance industry expect to receive financial returns from the MFI.

To counteract the effect of commercialization of MFIs social investors are also investing in the microfinance sector (Varottil 2012). Varotill (2012) argues that there are two types of social investors. The first represents the commercial branches of multilateral institutions or other development

agencies. The second is represented by specific institutions which were established to invest in social business. Social investors are argued to have the advantage that they advocate both social and financial objectives (Varottil 2012).

Generally, all equity investors, with the exception of a few social investors, invest in MFIs to obtain financial returns. Board representation of such parties will likely drive MFI management towards a more profit driven objective, rather than an outreach objective. This argument is

strengthened by the results found by Hartarska (2005), who finds that investor representation on an MFI board promotes financial sustainability. Investors on boards of MFIs will push for financial sustainability since this will give them the greatest chance of receiving back their funds and at least some of their expected returns.

However, Hartarska and Mersland (2012) find that equity investors on the board of an MFI are beneficial for reaching poor clients. They argue that through monitoring, equity investors can

positively affect the performance of the MFI, where performance measures the efficiency in reaching poor clients.

(19)

Master thesis Jorrit Steenbeek Page 19 2.5.5 Independent Directors

Independent directors can be defined as the board members who do not have an affiliation with any of the stakeholders of the MFI (Hartarska 2005), or as directors who are neither direct shareholder representatives nor members of management (CMEF 2005) or “directors who have no pecuniary relationship with the company, its management or substantial shareholder” (Varottil 2012). Usually, banks tend to have a greater proportion of outside directors (Hartarska and Mersland, 2012). The reason to have independent directors on the board of an MFI is because they are expected to monitor and advise better than other directors which have personal interests in the MFI (Hartarska 2005). Anguila (2011) and Varottil (2012) both argue that theoretically independent directors will have an unbiased perspective which is required to fulfil their supervisory task effectively. This is supported by the CMEF (2005). Thus, when there are independent directors on the board of an MFI, one can argue that the MFI will achieve higher performance, both socially and financially, compared to MFIs which do not have independent board members.

The findings with regard to the impact on the relation between the proportion of independent directors and firm performance have been positive and negative (Mayers et al. 1997). Still, empirical evidence points to the findings that the proportion of independent directors has a positive impact on performance in some banks (Hartarska 2005). Similarly, in the conventional corporate governance literature, Adams and Mehran (2003) argue that when there is an increase in the proportion of independent directors on the board of a firm there should be an increase in firm performance since they are more effective management monitors. This argument also counts for MFIs. Hartarska (2005) finds that a more independent MFI board gives a better return on assets. Furthermore, Kyereboah-Coleman and Osei (2008) find board independence to be positively correlated to profitability and outreach of MFIs in Ghana.

Previously conducted research has found that in general, the presence of independent directors leads to improved financial performance. With this in mind, the following assumption is made about independent directors on the boards of MFIs:

The presence of independent directors on the board of an MFI will likely increase the financial performance of the MFI, thus, increasing financial sustainability.

For the purposes of this paper, independent directors are therefore assumed to be advocates of financial sustainability.

2.5.6 Board Diversity and MFI Performance Hypotheses

Collectively the stakeholders represented on the board of directors define a very diverse board. Thus, as will be explained in section 3.2.3, this paper develops a measure for diversity. As the

(20)

Master thesis Jorrit Steenbeek Page 20 a board of directors. This section defines the main hypotheses with regard to board diversity and MFI performance.

Stakeholder theory argues that increased diversity is good for performance. It argues that the organization is the centre of a network of stakeholders, a complex system of exchanging services, information and other resources (Freeman 1984). This system creates a network for the firm and its management which leads to greater efficiency in the allocation of resources and the gathering of knowledge. This theory has been widely used in the conventional corporate governance literature. This paper assumes that stakeholder theory can also be applied to the microfinance industry. With this in mind, the theoretical construct this paper builds on is assembled as shown in figure 1, below. The board process and board task dimensions represent the underlying links which tie board diversity to MFI performance. This paper believes that board processes and board tasks are the underlying factors which may explain the relationship between board diversity and MFI performance.

Figure 1: Stepwise theoretical model1

1_{The theoretical model shows the predicted linear relationships. The direct relationship between diversity and}

MFI performance is represented by the solid arrow, whereas the linear relationship predicted by the underlying determinants are represented by the broken-line arrows. Finally, the supplementary analysis looks at the relationship between board processes and MFI performance, represented by the dotted line.

(21)

Master thesis Jorrit Steenbeek Page 21 Thus, when looking at the relationship between board diversity and MFI performance there are two important hypotheses that can be made:

Hypothesis 1a: diversity will have a positive impact on the financial sustainability of an MFI Hypothesis 1b: diversity will have a positive impact on the outreach achieved by an MFI

2.6

Board Processes

Before hypotheses are made regarding the effect of board diversity on board processes an understanding of board processes is necessary, therefore, this section looks at what board processes are and the relevant theory.

2.6.1 Conflict

Team conflict and cohesion have an important impact on the ability of team members to interact effectively over time (Gersick 1988). According to Tuckman (1965) conflict often occurs early between team members as they express differences in values and perspectives. The models proposed by Gersick (1988) and Tuckman (1965) argue that conflict management may be an important developmental process for teams. Team conflict can be broadly defined as “a process in which one party perceives that its interests are being opposed or negatively affected by another party” (Tekleab et al. 2009). However, previous research has shown that conflict is multidimensional and can be

separated into at least two dimensions, including task conflict and relationship conflict (Pinkley 1990). Task conflict refers to “disagreement among group members about the content of the tasks being performed, including differences in viewpoints, ideas and opinions” (Jehn 1995). Issues which arise over the tasks which challenge a team, or board of directors can lead to frustration and lead to dissatisfaction on the board (Amason and Schweiger 1994). Relationship conflict refers to the „interpersonal incompatibility among members, which typically includes tension, animosity, and annoyance among members within a group‟ (Jehn 1995). Interpersonal problems enhance negative reactions such as anxiety and fear, decreasing an individual‟s satisfaction with the group experience (Jehn 1995).

(22)

Master thesis Jorrit Steenbeek Page 22 processing ability of the group because group members spend their time and energy focusing on each other rather than on the issues at hand. This argument is supported by research done by Evan (1965) and Jehn and Mannix (2001). Furthermore, relationship conflicts limit group members‟ cognitive functioning by increasing the stress and anxiety levels (Jehn and Mannix 2001).

Thus, conflict has been suggested to interfere with team performance and reduce satisfaction because it produces tension, antagonism, and distracts team members from performing the task (Dreu and Weingart 2003). This argument is supported by empirical evidence found in studies done by Wall and Nolan (1986). However, research done by Schulz-Hardt et al. (2002) showed that teams made better decisions when there were preference disagreements in the discussion. Gladstein (1984) argues that conflict tends to be beneficial up to some level and beyond that level the conflict becomes detrimental, implying an inverted U-shape relationship.

Since there are only two MFI objectives in which the represented parties are interested in, an increase in diversity leads to an increase in the representation of one of these objectives, leaving the other representatives in a minority position, therefore reducing conflict. On the other hand, as section 2.5 argued, maximum separation occurs when a unit is divided into two divergent subunits. This paper assumes that the stakeholders represented on the board are either advocates of outreach or financial sustainability and therefore, maximum separation can be achieved. Furthermore, due to the complexity of the composition of the board of directors of MFIs, it is likely that board diversity will create an environment in which conflict is apt to thrive. Thus, the following hypothesis is posited:

Hypothesis 2a: diversity will have a positive impact on the level of conflict in a board

2.6.2 Board Communication

Shaw (1981) describes communication as the heart of group behaviour and argues that for a group to “function effectively, its members must be able to communicate easily and efficiently” (Shaw 1981). Research done by Katz and Kahn (1978) and Daft and Lengel (1984) point towards two dimensions of communication: Frequency, the amount of interaction between team members (Smith et al. 1994) and communication informality, the extent to which top management teams favour less formal communication channels (Smith et al. 1994). An expectation predicted by Smith et al. (1994) states that “informal communication is expected to facilitate the ease and frequent flow of

communication among team members”. This is supported by Daft and Lengel (1984) who quote Mintzberg (1973) when he argues that managers thrive on informal, personal communications. Lott and Lott (1961) found that there is a positive relationship between the amount of group

communication and the cohesiveness of the group which benefits higher quality problem solving by the group

(23)

Master thesis Jorrit Steenbeek Page 23 be increased due to the double bottom line the MFIs inherently set for themselves. The flow of

information between MFI directors and executives is crucial for the successful functioning of the board and, ultimately, the MFI. Indeed, in their study Daft and Lengel (1984) argue that information processing within groups is an explanation of the observed performance of the firm.

In short, this paper defines communication as the extent to which information is exchanged, during- and outside (informal) board meetings, between members of the MFI board.

Thus, when there is greater informal communication and sharing of information between MFI board members the tasks the board members are assigned will be achieved more effectively. The level of communication within a board should depend on the diversity of the board of directors. This paper argues that possible underlying determinants of MFI performance, outreach and financial

sustainability, are board processes. In this respect, stakeholder theory implies that an increase in diversity in a board will lead to increased communication and ultimately to increased performance. Furthermore, an increase in diversity could lead to the formation of a minority sub-group within the board for one of the objectives of the MFI. Consequently, the sub-group which holds the majority can communicate more effectively amongst themselves leading to an increase in the overall level of communication within a board. Thus, with this in mind the following hypothesis is form:

Hypothesis 2b: diversity will have a positive effect on the level of board communication

2.6.3 Board Agreement

Agreement reflects a shared understanding of the decision-making process, which, in turn reduces uncertainty and allows participants to focus on the substance of their decision (Iaquinto and

Fredrickson 1997). Iaquinto and Fredrickson (1997) argue that a lack of general agreement is most likely due to inconsistent perceptions among top management team members.

The literature on agreement focuses on the consensus on firm objectives, the means to achieve these firm objectives and on the effect of consensus within a top management team on organizational performance.

With regard to the effect of agreement on the performance of a firm, previous research is ambiguous. A number of authors have found a negative relationship between top management team agreement and firm performance (Grinyer and Norburn 1977, Woot et al. 1977). On the other hand, Hrebiniak and Snow (1982) found a positive relationship between firm performance and agreement amid top management. This finding is supported by Iaquinto and Fredrickson (1997) who find a positive and significant relationship between TMT agreement and firm performance.

(24)

Master thesis Jorrit Steenbeek Page 24 members who share common experiences have a homogenous understanding of the perspectives amongst the board members. This leads to the expectation that these common experiences reinforce the perceived agreement amongst the board members (Ford and Seers 2006).

Thus, when there is a greater level of board agreement amongst board members it is likely that the board tasks assigned to the MFI board will be achieved more effectively, ultimately leading to

improved achievement of MFI outreach and financial sustainability. Finally, as with communication, agreement should also be dependent on board diversity. Conform with the argument for board

communication, stakeholder theory implicitly argues that an increase in diversity will lead to increased board agreement and thus, to increased MFI performance. Given the assumption of the two

preferences of board stakeholders the relationship between board diversity and agreement is likely to be determined by the objective which the majority of the board of directors advocates. Therefore, the following hypothesis states:

Hypothesis 2c:diversity will have a positive effect on the level of board agreement

2.7

Advisory and Monitoring Functions

This section looks at the two main tasks a board of directors must perform. Furthermore, this section develops the relationship between board processes and board tasks and it develops the relationship between board tasks and the MFI objectives, outreach and financial sustainability.

There are two central functions which a board of directors must perform, the monitoring of management on behalf of shareholders and providing resources (Hillman and Dalziel 2003). The resources provided by board members to a firm can be classified into four categories; (1) advice and counsel, (2) legitimacy, (3) channels for communicating information between external organizations and the firm, and (4) preferential access to commitments or support from important elements outside the firm (Hillman and Dalziel 2003). These elements can be combined and called “board capital” (Hillman and Dalziel 2003). The provision of advice and counsel element can be defined as “providing expertise, including the provision of internal firm information by inside directors” (Baysinger and Hoskisson 1990) and administering advice and counsel (Mintzberg 1983). Resource dependence theorists have linked board capital to firm performance in previous literature.

(25)

Master thesis Jorrit Steenbeek Page 25 by the board in an MFI is very important in an MFI since there are potential costs which stakeholders will have to incur if management were to pursue their personal interests (Hillman and Dalziel 2003). The literature on the advisory and monitoring roles of the board of directors is mainly focused towards the effects on firm performance, where performance is measured using the traditional

measures such as Return on Assets (ROA) Return on Equity (ROE) or Tobin‟s Q (Selling and Stickney 1989, Lang and Stulz 1993). However, one can argue that the outreach and financial sustainability objectives can act as proxy‟s for an MFI‟s performance. Depending on the explicit mission an MFI has defined for itself performance can be measured as outreach or financial sustainability (Hartarska 2005, Luzzi and Weber 2006, Bassem 2009).

As has been indicated before, it is not unusual to find different stakeholders on the board of directors of an MFI (Hartarska 2005). The implication is that the agency theory is not able to

accurately predict what effect the monitoring and advisory tasks assigned to the board of an MFI will have on outreach and financial sustainability. However, the stakeholder theory combined with

resource dependence theory may lead to more accurate hypotheses about the effects of monitoring and advice on outreach and financial sustainability. Resource dependence theory perceives the board of directors of a firm as providers of resources such as legitimacy, advice and links to other organizations (Hillman and Dalziel 2003). This theory argues that a larger board has more resources due to

additional members compared to smaller board (Mori and Mersland 2011). The diversity of the different stakeholders in an MFI may have the consequence of creating a large board.

However, Anguila (2011) argues that boards of MFIs scarcely provide functional oversight and control because of little expertise in microfinance, banking or other relevant sectors. Therefore, MFI boards are not able to provide direction, successfully contribute to MFI strategy or provide useful advice to management (Anguila 2011). This idea is supported by the views of Ayesha et al. (2009) who argue that financial expertise and business management skills are commonly not up to standard at the MFI board and management levels because managers and board member often originate from the NGO sector where the strengths and competencies are mainly focused on social programs rather than finance or business.

The formulation of hypotheses starts by looking at the theoretical relationship between board processes and board tasks. Firstly, one can argue that since the focus is on board tasks the relevant type of conflict is task related conflict. The literature has shown that a higher level of task conflict challenges the board members to become more critical, and express their opinions and interests with better information, thus, making better and more informed decisions. Therefore:

(26)

Hypothesis 3b: board conflict will have a positive effect on the advisory task assigned to the board of directors of an MFI

Communication between board members is important for the achievement of the tasks of the board. Good communication between members means that the complex information processing will occur efficiently and thus, lead to better achievement of board tasks. Thus:

Hypothesis 3c: board communication will have a positive impact on the monitoring task assigned to the board of directors.

Hypothesis 3d: board communication will have a positive impact on the advisory task assigned to the board of directors.

Common experiences shared by board members create a uniform understanding of each other perspectives. This understanding is important to for level of agreement within a board and the

accomplishment of the monitoring and advisory task of the board. Therefore:

Hypothesis 3e:board agreement will have a positive effect on the monitoring role assigned to the board of directors.

Hypothesis 3f: board agreement will have a positive effect on the advisory role assigned to the board of directors.

To make the theoretical story complete, the last part of this section looks at the relationship between board tasks and outreach and financial sustainability. An MFI is the center of a network of stakeholders, each with their own resources which can create benefit for the MFI. Thus, theoretically according to stakeholder theory and resource dependence theory, the greater the diversity of the board of directors, the greater the resources available to the board of directors and therefore, the more competent the directors are at monitoring and advising MFI management. Thus:

Hypothesis 4a: the monitoring by the board of directors will have a positive effect on the outreach achieved by an MFI.

Hypothesis 4b: monitoring by the board of directors will have a positive effect on the financial sustainability achieved by the MFI.

Hypothesis 4c: the advisory task assigned to the board of directors will have a positive effect on the outreach achieved by the MFI.

(27)

Master thesis Jorrit Steenbeek Page 27 2.8

Supplementary Analysis Theory and Hypotheses

The theoretical model described at the end of section 2.5.6 shows that this paper believes a relation to exist between board diversity and MFI outreach and financial sustainability. Furthermore, this paper argues there is a relation between board tasks and outreach and financial sustainability. This leaves one relation missing in the theoretical model, the relationship between board processes and outreach and financial sustainability. A supplementary analysis is done to find the effect of board processes on outreach and financial sustainability. Therefore, this section defines the underlying theory and hypotheses with regard to this relationship, as shown in figure2, below.

Figure 2: Linear Relationship between Board Processes and MFI Performance

2.8.1 Conflict

Section 2.6.1 outlines the theory with regard to conflict and performance, therefore, to briefly summarize; task conflict is associated with effective decision making and relationship conflict is negatively associated with firm performance (Simmons and Peterson 2000). This paper previously argued that conflict would have a positive effect with regard to the assigned board tasks of monitoring and advise. This section looks specifically at the relationship between conflict and firm performance. However, still building on the assumption that task conflict is the likely type of conflict encountered in an MFI board, the following hypotheses are formed:

Hypothesis 5a:board conflict is likely to have a positive effect on the level of outreach of an MFI. Hypothesis 5b: board conflict is likely to have a positive effect on the level of financial

sustainability of an MFI.

2.8.2 Communication

As predicted by Smith et al. (1994), informal communication is expected to assist the ease and frequency of communication among team members. Furthermore, information processing becomes more efficient with good communication amongst board members (Daft and Lengel 1984). These positive effects will have an effect on the performance of an MFI, thus, the following hypotheses are posited:

Hypothesis 5c:board communication will have a positive effect on the level of outreach of an MFI.

(28)

Hypothesis 5d: board communication will have a positive effect on the level of financial sustainability of an MFI.

2.8.3 Agreement

The relevant literature is ambiguous on the effect of agreement on the performance of a firm. When there is a better understanding of the team member perspectives amongst the MFI directors, and who have shared common experiences, there will be a greater perceived level of board agreement (Ford and Seers 2006). With regard to performance, greater agreement between board members increases performance and a lower level of agreement will lead to a lower performance. Therefore, this paper hypothesises:

Hypothesis 5e:board agreement will have a positive effect on the level of outreach of an MFI. Hypothesis 5f: board agreement will have a positive effect on the level of financial sustainability

of an MFI.

3. Data and Methodology

3.1

Data

The dataset describes a questionnaire answered in 2010, by employees of different microfinance institutions in Eastern Africa, specifically Uganda, Kenya and Tanzania, about the processes within the board of directors. A five point Likert scale indicates the level which a respondent believes the question to be applicable to their respective MFI. The respondents were asked to answer a total of 51 questions with respect to the board processes of the MFI.

The full dataset is comprised of 79 MFIs in Eastern Africa with a total of 2,244,103 borrowers and a combined total of $10,503,516,816 assets. A critical strength of the dataset is that it is unique

because it focuses on the processes within an MFI board. This is unique since, to my knowledge, no corporate governance oriented paper in the microfinance literature has looked at the processes within a board of MFI directors.

(29)

Master thesis Jorrit Steenbeek Page 29 3.2

Methodology

This paper uses a number of methods to make inferences about the data including OLS regressions, a factor analysis, interaction2 variables and a diversity index.

3.2.1 OLS

This paper uses ordinary least squares (OLS) regression3 to make inferences about the dataset. The ultimate dependent variables outreach and financial sustainability can be measured using a number of different proxies. For outreach the proxy employed is the average loan size of the

borrowers of an MFI. As the literature indicates, this does mean that the dependent variable is more a proxy for the depth of outreach than for the breadth of outreach. However, this should not have any consequences for the results of the regressions and the inferences with regard to the results of the regression will be made accordingly. For financial sustainability the proxy used is the operational self-sufficiency ratio. It is one of the most used measures in the microfinance literature to measure financial sustainability and therefore a comparison across studies becomes easier.

In the regression results there are two regression models, Model 1 and Model 2,

3.2.2 Factor Analysis

Many of the questions in the questionnaire are closely related and may measure the same underlying dynamic. Therefore, a factor analysis4 is used to determine which questions fit together well and can be combined into a single variable. 3 factors were created to capture the board processes: level of conflict, the level of communication amongst board members and finally, the level of

agreement between board members. Furthermore, two factors were created to measure the board tasks; monitoring and advisory task. Table 7.5, in the appendix, outlines the specific variables which were combined to form their respective factors. To gain an initial idea of which variables could be

combined into different factors a correlation table was created, table 7.7 in the appendix. The variables which correlate highly with a high significance level will likely be combined to form a single factor5.

3.2.3 Diversity

After the creation of the three board processes and the factors representing the two board tasks a diversity index was created. As noted in the literature review, this paper assumes that there are two interests represented on the board of directors, the first is outreach and the second is financial

2

The interaction variable methodology is explained in the appendix, section 7.2.

3

There are generally two regression models which are referred to during the analysis, Model 1 refers to the regression which does not take into account control variables and has 73 observations. Model 2 refers to the regression which does take into account control variables and has only 43 observations.

4_{To optimize the analysis of the available data the factor analysis was done without taking into account the}

missing control variable observations for individual MFIs. This means that there were 73 available MFIs with complete information for the factor analysis. When the control variables were taken into account 43 MFIs were left for further analysis.

5_{For a full specification of the factor analysis please refer to tables 7.5, 7.6, 7.7, 7.8, 7.9 and 7.10 in the}

(30)

Master thesis Jorrit Steenbeek Page 30 sustainability. Table 3.1 shows a short summary of which board member represents either outreach of financial sustainability.

Table 3.1: Board member interests represented on the board of an MFI

Board Member MFI Objective

Donors Outreach

Employees Outreach

Customers Financial Sustainability

Investors Financial Sustainability

Independent Directors Financial Sustainability

Following Goodstein et al. (1994) a Gibbs-Martin index (a.k.a. the Blau Index) was calculated using the following formula:

1 − 𝑃𝑖2 𝑛

𝑖 (Goodstein et al. 1994).

Here 𝑃𝑖2 represents the proportion of a board accounted for by the 𝑖𝑡𝑕 group. The Gibbs-Martin index indicates the extent of concentration of board members ranging from a high

concentration with a single group (index of 0 indicates complete homogeneity), to an extremely low concentration based on an equal distribution of members across a very large number of groups (index of 1). As indicated, there are two possible categories into which a board member can fall; a group which holds an outreach objective as the MFIs central mission, and a group which holds financial sustainability as the MFIs most important objective.

4. Results and Discussion

This section outlines the results of all regressions and also directly discusses these results with regard to the hypotheses and theory.

4.1

Board Diversity and its Effect on Outreach and Financial Sustainability

The Gibbs-Martin index is central to the first regression of this paper. This first part of the analysis looks at what effect board diversity has on the MFI objectives outreach and financial sustainability. The results of this regression6 are shown in table 4.1, below.

6_{Footnote 2 does not apply to this regression. The proxies for outreach and financial sustainability were limited}

(31)

Table 4.1: Board Diversity and its Impact on Outreach and Financial Sustainability

Outreach Financial Sustainability

Model 1 Model 2 Model 1 Model 2

Diversity -55.924 (244.028) -1061.687 (418.049)** -46.400 (18.130)** -103.334 (34.389)*** Age 89.665 (130.163) 27.285 (10.241)** Size 130.237 (36.166)*** 0.662 (2.729) Constant 411.743 (92.756)*** -1545.979 (481.434)*** 116.501 (6.890)*** 49.340 (35.783) N 43 37 43 37 R2 0.001 0.434 0.138 0.337 Adj. R2 -0.024 0.379 0.117 0.286 F-stat 0.053 7.913 6.550 6.603

Standard errors in parentheses ***p<0.01, **p<0.05, *p<0.1

Model 1 shows the regression results without taking into account the control variables and Model 2 does take into account the control variables. This holds for all the other upcoming regressions in the next sections.

Board diversity is found to have a negative effect on outreach, Model 2 shows that this effect is significant at the 5% level. The same negative effect is found with regard to financial sustainability, and both Model 1 and Model 2 show that this negative effect is significant at the 5% and 1% level, respectively. Both these results are contrary to the positive impact board diversity was expected to have on outreach and financial sustainability as predicted by hypotheses 1a and 1b. A possible explanation is that as diversity increases, there is a more equal distribution of stakeholders

representing either outreach or financial sustainability. Since a stakeholder can only take one of these positions, an equal distribution for outreach and financial sustainability may not allow the board to focus on either, therefore, decreasing the MFIs achievement for both. Furthermore, the R2 and adjusted R2 show that the model fits the data relatively well, apart for Model 1 for outreach, and are comparable to similar research results in the microfinance literature (Hartarska 2005).

4.2

Diversity and Board Processes

Now that a negative effect of board diversity on outreach and financial sustainability has been uncovered, the question becomes: What is the underlying dynamic of this relationship? Therefore, this section analyses the relationship between board diversity and the board processes conflict,

communication and agreement.