1
University of Groningen
MSc. International Business & Management Specialization: International Financial Management
Uppsala Universitet MSc. Economics & Business
The Influence of Offering Additional Financial Services on
the Financial and Social Performance of MFIs in Asia
Author: Roswitha Knol
The Influence of Offering Additional Financial Services on
the Financial and Social Performance of MFIs in Asia
Student Roswitha Knol
S1535013
Admiraal de Ruijterweg 64 I 1056 GM Amsterdam
Roswithaknol@gmail.com
University and University of Groningen
Educational Program Faculty of Economics and Business
MSc. International Business Administration
Specialization: International Financial Management
Uppsala Universitet
Faculty of Economic Sciences
MSc. Economics and Business
University Supervisors Prof. dr. C.L.M. Hermes (first supervisor)
Dr. W. Westerman (second supervisor)
3
Where once the poor were commonly seen as passive victims, microfinance recognizes that poor people are remarkable reservoirs of energy and knowledge. And while the lack of financial services is
a sign of poverty, today it is also understood as an untapped opportunity to create markets, bring people in from the margins and give them the tools with which to help themselves.
The Influence of Offering Additional Financial Services on
the Financial and Social Performance of MFIs in Asia
Roswitha Knol
Abstract
Currently, a growing number of MFIs are offering additional services like microsavings, microinsurance and Business Development Services besides microcredit. The aim of this research is to find empirical evidence on whether offering these additional financial services will have a positive or negative influence on the financial and social performance of MFIs and if a trade-off between social and financial performance is in place. Using a multiple linear regression for a sample of Asian MFIs in the years 2007-2009, evidence is found for relationships between the services offered and the performance. The results indicate that offering additional services does affect the social and financial performance of MFIs. For instance, offering insurance will probably not lead to financial self-sufficiency but it will reach the poorest on average. While a trade-off is verified in some cases, other cases show that synergies are possible as well.
Keywords
5 ABBREVIATIONS
BDS Business Development Services
SPI Social Performance Indicators
MDG Millennium Development Goal
MFI(s) MicroFinance Institution(s)
NBFI Non-Bank Financial Institution
NGO Non Governmental Organization
NI (btd) Net Income (before taxes and donations) ROA (atd) Return On Assets (after taxes and donations) ROA (btd) Return On Assets (before taxes and donations)
ROE Return On Equity
UN United Nations
TABLE OF CONTENT
1. INTRODUCTION 8
1.1 Background 8
1.2 Motivation 8
2. LITERATURE REVIEW 10
2.1 Financial Performance and Social Performance 10
2.1.1 Trade-Off 11 2.1.2 Synergy 11 2.2 Financial Performance 12 2.2.1 Microsavings 13 Compulsory Savings 15 Voluntary Savings 16 2.2.2 Microinsurance 16
2.2.3 Business Development Services 18
2.2.4 All Financial Services 19
2.3 Social Performance 20
Depth of Outreach 21
Breadth of Outreach 21
2.3.1 Microsavings 22
2.3.2 Microinsurance 23
2.3.3 Business Development Services 24
2.3.4 All Financial Services 25
7
3.2 Data 31
3.2.1 Data Collection 31
3.2.2 Limitations 33
4. RESULTS, ANALYSIS AND DISCUSSION 34
4.1 Average Financial Performance 34
4.2 Regression Financial Performance 35
4.3 Average Social Performance 37
Depth of Outreach 37
Breadth of Outreach 38
4.4 Regression Social Performance 39
4.5 Regression per Legal Status 40
4.6 Trade-Off and Synergy 41
5. CONCLUSION 43
6. REFERENCES 45
1. INTRODUCTION Background
Poverty is a major issue worldwide. Since the introduction of the Millennium Development Goals (MDG) by the United Nations (UN) 2000, poverty reduction has become a priority. For instance, MDG 1 is to eliminate extreme poverty and hunger and microfinance is viewed as an important mechanism to realize this goal (www.un.org). Microfinance’s origin lays with Muhammed Yunus in Bangladesh several decennia before the development of the MDGs. Microfinance, and more specific microcredit, was initially founded by Yunus to overcome poverty among women. The shift from microcredit to microfinance comes from the realization that low-income people are beyond needing only lending programs. They also benefit from savings, insurance and Business Development Services (BDS) possibilities (Berg, 2010). Nowadays the focus is not merely on the creation of income, building of assets and the creation of jobs. A broader focus has been accepted which entails the use of insurance, BDS and savings for risk management for the poor (Churchill, 2002). These financial services are delivered to the poor by MicroFinance Institutions (MFIs).
The basic product offered by MFIs is microcredit. Microcredit basically involves a relatively small size loan. Some MFIs offer additional services, for example, BDS, insurance and savings. First, BDS entails several training possibilities for the poor. Second, microinsurance provides low-income people with the opportunity to insure themselves. Third, microsavings gives the poor the chance to put their spare resources in a savings account. Microsavings can be divided in compulsory savings and voluntary savings. Compulsory savings indicates that the client is obligated to save a part of the microloan within the institution whereas voluntary savings are optional. While it must be acknowledged that additional financial services by MFIs are still underdeveloped nowadays, gradually MFIs are gaining interest in offering one or more of these financial services.
Motivation
9
argue that synergies between the two goals are in fact possible (Schreiner, 2001; Morduch, 1999; Meyer, 2002 and Zeller and Johanssen, 2008))
Offering one or more additional financial service like microsavings, microinsurance and/or BDS besides credit could result in a high social performance. A higher social performance could be reached because the varying demands of the poor are better met and thus more people and poorer people can be reached. Reaching a high financial performance by offering multiple services is also possible: providing more than just credit might result in higher profits because for example of economies of scale and scope and retaining customers. Moreover, some risks/costs like information asymmetry may decrease because of a larger knowledge base. Of course some pitfalls must be acknowledged as well. For instance, offering more financial services is likely to encourage moral hazard problems and brings about extra costs; this puts pressure on the financial performance (Hannig, 1999; Morduch, 1999; Matin et al., 2002; Siegel et al, 2001; Ahuja and Guha-Khasnobis, 2005; Churchill, 2002; Brown et al., 2000; Hulme and Mosley, 1998 and Poudyal, 2008). Thus, there are positive and negative potential effects on social and financial performance and offering one or more additional service. The main goal of this research aims at developing empirical evidence on whether MFIs offering additional financial services like microinsurance, microsavings and BDS achieve a higher or lower financial and social performance than credit-only MFIs.
Research concerning the effects of financial services and social and financial performance is limited, especially empirical research in this area is lacking. Therefore, from academic and professional perspective research in this area is necessary and complementary. In this research I will focus on fourteen countries in Asia with data generated from the online database MIXmarket. I will investigate the impact of offering additional financial services on the social and financial performance of MFIs with the following Research Question:
Research Question: ‘Does offering additional services like microinsurance, microsavings and BDS
besides microcredit result in a higher or lower financial performance and a higher or lower social performance for MFIs than offering only microcredit in Asia?’
The remainder of this thesis is structured as follows: Section 2 gives an overview of the literature on the relationship between MFIs offering additional financial services and the expected financial and social performance of these MFIs. Section 2 also presents the associated hypotheses. In the third section the research methods and data collection processes are discussed. The results, analysis and discussion are presented in section 4. Finally, the conclusion is provided in section 5.
2. LITERATURE REVIEW
2.1 Financial Performance and Social Performance
Zeller and Johannsen (2008) state there are three possible operational objectives for MFIs. The first objective entails financial sustainability for an MFI. Financial sustainability describes the ability of an MFI to continue to operate in the long term. It is argued that financial sustainability can only be reached if the MFI is self-sufficient. Self-sufficiency is reflected by an MFI’s ability to cover its costs with its own revenues, and is independent of subsidies and/or donations (Brau and Woller, 2004; Hashemi, 2007; Woller et al., 1999; Morduch, 1999; Churchill, 2002; Zeller and Johanssen, 2008 and Morduch, 2000). The belief that an MFI should be self-sufficient and not subsidy dependent has been depicted as the ‘institutionist view’ (Brau and Woller, 2004 and Woller et al., 1999). Thus, financial self-sufficiency ultimately leads to financial sustainability. Therefore, in this review the two terms will be used simultaneously. The second objective addresses an MFI’s aim on impact on poverty. Impact on poverty discusses the way financial systems influence socio-economic and environmental development in an area (not specifically client oriented) and is predominantly aimed on poverty reduction (Zeller and Meyer, 2002). The last objective focuses on depth and breadth of outreach. Depth of outreach measures whether the poorest of the poor are reached whereas the breadth of outreach looks at the quantity of the poor being reached. MFIs typically started out with a social mission and they still embrace this: they strive for a have a high outreach and a large impact (Brau and Woller, 2004; Meyer, 2002; Woller et al., 1999 and Morduch, 2000). This view is recognized as the ‘welfarist view’. The welfarist view emphasizes that MFIs do not need to be financial self-sufficient in order to be sustainable. More precise, welfarists see donations as equity and donors as social investors who are interested in poverty reduction rather than financial revenues (Brau and Woller, 2004 and Woller et al., 1999).
Figure 1: Operational Objectives (Source: Zeller and Meyer, 2002)
Impact on Poverty Outreach to the Poor Financial Sustainability
Institutional Innovations Frame Conditions: Macro-‐Economic Environment, Policy, Human and Social
11
It is argued by several authors that typically a trade-off needs to be made between these three objectives. In their opinion reaching a high social performance and being financially sustainable cannot be reached simultaneously (Hulme and Mosley, 1996; Conning, 1999; Lapenu and Zeller, 2001; Zeller and Johanssen, 2008; von Pischke, 1996; Cull et al., 2007; Hashemi, 2007; Tucker and Miles, 2004 ad Brau and Woller, 2004).
2.1.1 Trade-Off
Several explanations exist for why MFIs are unlikely to achieve both a high social and a high financial performance. Frequently, the MFI with a focus on financial sustainability does not have a large depth of outreach (Cull et al., 2007). Making high earnings is easier for an MFI if it concentrates on the less poor who less risky and more profitable. This is explained by the fact that the less poor are usually able to provide collateral on the loan and they are less likely to default the loan. Moreover, the less poor usually request higher loans, which generally results in higher profits (Zeller and Johannsen, 2008; Tucker and Miles, 2004; Schreiner, 2001; Morduch, 1999 and Brau and Woller, 2004). Every loan exists of transaction costs. Transaction costs are defined as costs experienced when making an economic exchange (Williamson, 1981; Siegel et al., 2001 and Zeller and Johannsen, 2008). This encompasses for instance, costs of finding and assessing the potential micro client, administering the loan request and other administrations, distributing the loan to the client, making sure every client pays back the loan and collecting the paybacks (Mechler et al., 2006; Shankar, 2006 and Siegel et al., 2001). Transaction costs have a fixed cost component. This fixed cost component make the smaller loans (for the poorer people) relatively more expensive for the MFI compared to the larger loans (Hulme and Mosley, 1996; Matin et al., 2002; Conning, 1999; Zeller and Johanssen, 2008 and Lapenu and Zeller, 2001).
2.1.2 Synergy
of costs, the MFI is more likely to reach financial sustainability. In turn, clients value financial self-sufficiency of the MFI because they feel that they can trust this institution to be sustainable. This is especially important if their savings are held here. This trust could lead to higher customer loyalty and consequently in higher customer retention. A high client retention rate does not only assure financial income for the MFI but the client base may also grow because of positive word of mouth, which provokes a positive effect on the depth and breadth of outreach (Zeller and Johanssen, 2008).
2.2 Financial Performance
As explained above, striving for financial self-sufficiency and ultimately financial sustainability are argued to be important goals for the MFI (Morduch, 1999; Brau and Woller, 2004; Cull et al., 2007; Friebig et al., 1999; Kereta, 2007; Woller et al., 1999; Churchill, 2002; Zeller and Johanssen, 2008 and Morduch, 2000). Financial self-sufficiency is usually measured by the way the MFI is able to cover all its costs with its own revenues, without subsidies and donations (Friebig et al., 1999; Cull et al., 2007; Morduch, 1999 and Yaron, 1992). Gaul (2009) reports that 61% of the MFIs were financially self-sufficient in the year 2007. His sample comprised of the whole MicroBanking Bulletin database, which consisted of 2800 observations over 5 years. As can be seen in table 1, these numbers have been somewhat stable around 60% from 2003 to 2007.
Year: 2003 2004 2005 2006 2007 Financial Self-Sufficient MFIs 59% 67% 61% 57% 61%
Table 1: Percentage of Financial Self-Sufficient MFIs (Source: Gaul, 2009)
Revenues are largely generated through the interest on credits and fees. There are also other sources of revenues for MFIs that could lead to financial self-sufficiency. First, a high client retention rate; a high client retention rate can be established by trust and the ability of an MFI to meet the demands of the customer (Meyer, 2002). Customer retention and financial self-sufficiency are also important nowadays because of the increasing competition between several MFIs. The increasing competition requires MFIs to gain market shares or to reach clients in new markets in order to be sustainable. Another source of revenues can be generated through high repayment rates. Repayment rates represent the number of loans that are repaid to the MFI. Both high retention rates and high repayment rates secure a stable stream of income. Cull et al. (2007) argue that higher repayment rates have a positive influence on the financial performance of the MFI.
13
because the MFI will be left empty handed if the client goes in default. Moreover, the MFI has administration costs. Administration costs usually represent the costs of handling the client base. Another source of costs is the cost of capital. These are costs of attracting external funds that can be lent on to clients. Furthermore, costs could arise because of the insecure and risky environment MFIs operate in. Costs appear for instance because of asymmetric information. Information is unequally distributed between the MFI and the client. Information asymmetry could result in adverse selection. Adverse selection assumes that the bad clients ‘lemons’ drive the good clients ‘cherries’ out of the market (Akerlof, 1970; Finkelstein and Poterba, 2004; Siegel et al., 2001; Zeller and Johannsen, 2008; Mosley, 2009; Brown et al., 2000 and Mechler et al., 2006). More specific, the MFI wants to decrease the risks of defaults and therefore sets high interest rates. This attracts less risk averse clients ‘lemons’ (which usually have a higher default risk and thus are riskier clients) who accept these higher interest rates and the cherries (the risk averse clients) will leave the MFI. The MFI will notice an increase in risky clients and a decrease in less risky clients. This could result in costs for the MFIs because the benefits of diversification of the risk between the risky clients and the less risky clients are dimished. Another consequence of information asymmetry is moral hazard. Moral hazard refers to changes in the behavior of clients in case they are only partially responsible for the consequences of their behavior instead of being fully responsible. This is especially the case for insurance schemes. Clients with insurance are said to be less careful because their insurance will pay for the damage anyway (Holmström, 1979; Kereta, 2007; Arnott and Stiglitz, 1988; Fiebig et al., 1999 and Zeller and Johannsen, 2008). If clients engage in riskier behavior the MFI is facing higher costs than they would if the client was acting as normal. Finally, as discussed above, transaction costs can influence the profits of an MFI. Since the credit, savings and insurance are offered in small amounts and the transaction costs are not, the transaction costs are relatively high for the generally small size of the transactions (Matin et al., 2002; Siegel et al., 2001 and Zeller and Johannsen, 2008).
2.2.1 Microsavings
this irregular income. The poor are involved in savings in several ways. For instance, households keep savings at an MFI, store savings at their neighbor and they act as a deposit facility themselves by safeguarding money for a friend (Collins et al., 2009). It is important to note that not all MFIs in all countries are allowed to offer savings; sometimes this is only possible with a banking license.
A discussion exists between authors regarding whether offering savings has a positive or negative influence on the financial performance of the MFI. Hannig, 1999; Morduch, 1999 and Matin et al., 2002 argue that MFIs providing savings are more likely to reach a higher financial performance than credit-only MFIs for multiple reasons. Firstly, offering microsavings could result in a higher client retention rate, higher repayment rates and fewer dropouts for the MFI if the client also has a loan because the client has an incentive to stay and is more likely to be on time with necessary payments. This incentive comes forth because the client probably feels a stronger relationship with and higher loyalty towards the MFI if his or her savings are deposited at the MFI. Moreover, Zapata (2002) argues that a person who is actually capable of saving is probably more able to repay the loan because there is cash at hand. Secondly, MFIs might also recognize an increase in customers because the needs of the poor are better met (Meyer, 2002; Fiebig et al., 1999 and Nourse, 2001; Schmidt and Zeitinger, 1996; Zapata, 2002; Hulme and Mosley, 1996; Kaboski and Townsend, 2005; Morduch, 1999; Hannig, 1999; Berg, 2010; Hulme and Mosley, 1998; Wright, 2000; Siegel et al., 2001 and Mosley, 2009). Thirdly, economies of scale can reduce the costs made per client because some costs can be split over multiple services.
Furthermore, costs arising because of information asymmetries, moral hazard and adverse selection tend to be lower for MFIs offering deposits than for credit-only MFIs. First, information asymmetries are likely to decrease because a savings account gives the MFI insight in the customers’ consumption behavior. Second, savings serve as an incentive for the client not to fall into risky or negative behavior because their deposit money is stored at the MFI, which could lower the negative effects of moral hazard. And finally, offering savings could lower adverse selection because the MFI is able to diversify riskier (credit) and less risky (savings) clients among different products. More precise, savings will probably attract less risky clients that can be diversified with riskier clients (Armendáriz and Morduch, 2000 and Bruno and Khachatryan, 2011).
15
savings as a form of collateral in the case the client also has a credit account. Besides, the client could use the savings to buy assets that can further be utilized as an additional source of collateral (Morduch, 1999; Berg, 2010; Robinson 1995, and Matin et al., 2002).
On the other hand, offering savings some costs might increase for the MFI. For instance, transaction costs will likely increase for the MFI if it mobilizes microsavings. Since the savings are usually relatively small, the transaction costs can therefore turn out to be relatively high. (Zeller and Johannsen, 2008; Rossel-Cambier, K., 2011; Morduch, 1999; Nourse, 2001; Stiglitz 1986; Siegel et al., 2001 and Fiebig et al., 1999). Other costs due to start up, administrations and operations should be taken into account for an MFI mobilizing savings. Nevertheless, some costs only accompany the first transaction (for instance the loan) and can be passed over when the client wishes to open a savings account. These economies of scope see to it that administration processes, analyzing customer information or other operational costs are only necessary once. This means that these costs can thus be split between the costs for offering credit and costs for offering savings (Fiebig et al., 1999; Wright, 2000; Siegel et al., 2001; Meyer, 2002; Hannig, 1999; Morduch, 1999 and Schmidt and Zeitinger, 1996).
Hypothesis 1a: MFIs that offer microcredits and microsavings are expected to have a higher financial
performance than credit-only MFIs
Hypothesis 1b: MFIs that offer microcredits and microsavings are expected to have a lower financial
performance than credit-only MFIs
Compulsory Savings
(Zeller and Johannsen, 2008; Tucker and Miles, 2004; Schreiner, 2001; Morduch, 1999 and Brau and Woller, 2004).
Hypothesis 2a: MFIs that offer micro credits and compulsory microsavings are expected to have a
higher financial performance than credit-only MFIs
Hypothesis 2b: MFIs that offer micro credits and compulsory microsavings are expected to have a
lower financial performance than credit-only MFIs
Voluntary Savings
Voluntary savings, on the other hand, involve deposits on a flexible and optional basis (Nourse, 2001 and Brau and Woller, 2004). Besides the positive impact of internal financing, collateral and higher repayment rates, economies of scale and scope and retention rate, voluntary savings supplies the MFI with information on its clients. For example, it gives insight in its clients’ consumption pattern. An MFI with voluntary savings will have the advantage over credit-only MFIs that they can specify and design products according to the clients needs (Fiebig et al., 1999). Usually, voluntary savings are offered in different sizes in order to successfully and sustainably offer savings to the clients to increase the financial performance of the MFI (Robinson, 1994; Bruno and Khachatryan, 2011; Kaboski and Townsend, 2005 and Fiebig et al., 1999). Moreover, Churchill (2002) stresses MFIs should offer voluntary savings to achieve a higher customer retention rate and a growing client base, without savings this will not be achieved. Given the fact that clients do not necessarily need to borrow credit to be a customer they can also merely save money, a credit-only MFI does not offer this possibility.
Hypothesis 3a: MFIs that offer micro credits and voluntary microsavings are expected to have a
higher financial performance than credit-only MFIs
Hypothesis 3b: MFIs that offer micro credits and voluntary microsavings are expected to have a
lower financial performance than credit-only MFIs
2.2.2 Microinsurance
17
asset insurance and disaster risk insurance (Siegel et al, 2001; Mechler et al., 2006; Brown and Churchill, 2000).
Microinsurance is still in its infancy. Nevertheless, many authors argue that MFIs can make profits by offering insurances (Siegel et al, 2001; Ahuja and Guha-Khasnobis, 2005; Churchill, 2002; Brown et al., 2000 and Hulme and Mosley, 1998). Insurance-MFIs could make profits because of higher retention rates, less dropouts and higher repayment rates. This is because the insurance demands of the client are acknowledged and the client feels an incentive to stay and repay the loan if they take the insurance product from this MFI (Ahuja and Guha-Khasnobis, 2005; Nourse, 2001 and Mosley, 2009). Moreover, in the first place the premium payments serve as a form of collateral and as a source of direct income (Siegel et al., 2001; Ahuja and Guha-Khasnobis, 2005; Brown et al., 2000; Meyer, 2002 and Mechler et al., 2006). Of course this argument is impaired by the fact this direct income can be substantially effected with high costs if the MFI needs to pay out large sums of insurance to their clients (after, for instance, a natural hazard).
Furthermore, without insurance, the poor are prone to use savings or credit for insurance actions. This could entail several risks for the credit-only MFI (Nourse, 2001; Fiebig et al., 1999; Hannig, 1999 and Matin et al., 2002). A simplified example: a client receives a 100% loan for investments in the microenterprise but only spends 70% on (profitable) investments and uses the rest to insure against risks like sickness. Profits were estimated on 100% by the MFI and consequently a payback of 100% was assumed. Since only 70% was invested the client is likely to default a part of its loan and thus will not be able to pay back the whole loan. In this example this results in costs for the credit-only MFI (Siegel et al., 2001; Hulme and Mosley, 1998; Churchill, 2002; Mishra, 1994 and Mosley, 2009).
evaluating claims and the valuation of losses (Siegel et al., 2001; Ahuja, and Guha-Khasnobis, 2005; Mechler et al., 2006 and Mosley, 2009). Furthermore, providing microinsurance means the MFI needs additional information to accurately predict future trends. This is a costly task for MFIs, especially if inaccurate premiums are set. Credit-only MFIs are not faced with these extra costs (Brown and Churchill, 2000; Brown et al, 2000; Siegel et al., 2001 and Mechler et al., 2006).
Hypothesis 4a: MFIs that offer micro credits and microinsurance are expected to have a higher
financial performance than credit-only MFIs
Hypothesis 4b: MFIs that offer micro credits and microinsurance are expected to have a lower
financial performance than credit-only MFIs
2.2.3 Business Development Services (BDS)
Previously termed as non-financial services, BDS are services that can complement the financial services offered by the MFI. BDS is defined by the Committee of Donor Agencies for Small Enterprise Development (2001) as: ‘Services that improve the performance of the enterprise, its access to markets, and its ability to compete’. The content of BDS in general focuses on management training, including business plan development, marketing skills, financial skills, technical skills and legal and regulatory skills (Sievers and Vandenberg, 2007; Dumas, 2001; Karlan and Valdivia, 2010; Nieman, 2001 and Henry, 2006).
19
Moreover, knowledge gained by training can be complementary for entrepreneurs in order to gain financial success in their business. With training they can learn to take advantage of opportunities they might have overlooked without training (Nourse, 2001; Dumas, 2001 and Edgecomb, 2001). Subsequently, when clients experience the positive effects of the training on their profits in the microenterprise they are likely to move up to larger loan sizes because clients want to expand their business and thus need more capital from the MFI. This can contribute to the profits of the MFI offering BDS because MFIs usually start making profits after the third loan (Meyer, 2002; Karlan and Valdivia, 2010 and Sievers and Vandenberg, 2007). Poudyal (2008) even warns credit-only MFIs for stagnated or decreasing loan sizes due to lack of BDS.
Hypothesis 5a: MFIs that offer micro credits and BDS are expected to have a higher financial
performance than credit-only MFIs
Hypothesis 5b: MFIs that offer micro credits and BDS are expected to have a lower financial
performance than credit-only MFIs
2.2.4 All Financial Services
Collins et al. (2009) proved there is a high demand for additional financial services among the poor. They showed that the poor are involved in several forms of financial activities that keeps them busy daily. Why the poor need other services besides credit is explained in more detail per service in the previous chapters. MFIs offering all the financial services, including microsavings, microinsurance and BDS, and not just one additional service, are probably in the best position to completely fulfill all demands and needs of the client. Fulfilling the needs of clients’ best is also very important in the increasing competitive climate most MFIs are faced with nowadays (Sievers and Vandenberg, 2007 and Karlan and Valdivia, 2010).
Moreover, these MFIs will probably benefit from offering all these services in terms of a high financial performance. For these MFIs the same arguments apply as mentioned in the previous chapters. For instance, the MFI might have more satisfied, remaining, returning and new customers, fewer dropouts, collateral and higher repayment rates (Meyer, 2002; Churchill, 2002 and Nourse, 2001). The benefits of economies of scale and scope are highest for the MFI offering all financial services. Concerning economies of scope, general costs like administration can be diversified between the different services. Whilst for economies of scale the benefits arise from the large customer base of the MFI offering all services (Hashemi et al., 2007; Siegel et al., 2001; Nourse, 2001; Churchill, 2002; Staschen, 2001; Brau and Woller, 2004 and Meyer, 2002).
services compared to credit-only MFIs. When all services are offered a larger knowledge base and a larger degree of information exchange between the client and the MFI can be reched (Siegel et al., 2001; Mosley, 2009; Fiebig et al., 1999; Zeller and Johanssen, 2008; Kereta, 2007; Stiglitz and Weiss, 1981; Brown et al., 2000 and Brown and Churchill, 2000). Offering all financial services could also offset the effects of adverse selection because the MFI could attract riskier and less risky clients more sufficient. The MFI can also diversify the risks between the various products offered. More specific, the MFI could attract less risky clients by offering voluntary savings possibilities and riskier clients by offering insurance schemes. Moral hazard could similarly be lower for MFIs offering all financial service than for credit-only MFIs (Holmström, 1979; Kereta, 2007; Arnott and Stiglitz, 1988; Fiebig et al., 1999 and Zeller and Johannsen, 2008). When the information asymmetry is reduced, moral hazard will decrease because MFIs are more able to oversee the actions of their clients (Simtowe et al., 2006).
Offering additional products also has several disadvantages. For example, with offering additional services, additional costs arise because different administrations are necessary. Similarly, the relatively high transaction costs remain a hurdle for all MFIs. Moreover, the MFI must hire employees specialized in these products or train its current employees in the specific characteristics of these products.
Hypothesis 6a: MFIs that offer micro credits, microsavings, microinsurance and BDS are expected to
have a higher financial performance than credit-only MFIs
Hypothesis 6b: MFIs that offer micro credits, microsavings, microinsurance and BDS are expected to
have a lower financial performance than credit-only MFIs
2.3 Social Performance
21 Depth of outreach
Reaching the poorest of the poor is a justifiable goal for the MFI. Namely, by reaching the poorest, poverty could ultimately be alleviated. Lapenu and Zeller (2001) argue that Asia has the largest depth of outreach worldwide with an average loan size of USD 153, as compared to Latin America (USD 418) and Africa (261) in 1999. They argue that the ‘poorest’ are reached in case the average loan size remaind under the USD 300.
The poorest can only be reached if the MFI specifically focuses on this segment of the market. Nevertheless, serving the poorest is a risky and a very expensive business. These people might not have a home that can be used for collateral and their irregular income could influence their payback capacities. These risks could make offering financial services to the poorest an expensive business. Moreover, it is argued it is difficult to reach the poorest of the poor. One of the reasons why it is difficult to reach the poorest is because they are not able to pay high interest rates (Matin et al., 2002). Reaching the poorest concerns meeting their demands and fulfilling their needs. Meyer (2002), Matin et al. (2002) and Zeller and Johanssen (2008) propose that the demands and needs probably are different for the less poor. They might not just want credit but they preferably ask for additional products like insurance to lower the effects of the high risks they face daily like health issues for which they simply cannot take out a loan. Moreover, the poorest have more issues dealing with their irregular income and thus need resources to smooth their consumption with for instance savings (Rutherford, 2000; Collins et al., 2009; Meyer, 2002 and Sebstad and Cohen, 2001).
Breadth of outreach
The percentage of people living under the national poverty line is also presented (data.worldbank.org). Now we can compare the breadth outreach and the percentage of people living under the poverty line in a country in a specific year. While we must be aware of ‘multiple borrowing’ (i.e., clients
borrowing from two institutions or more at the same time) it can be observed that Bangladesh has the
highest market penetration rate, being over 10% in 2009. This appears consistent with the large amount of its inhabitants living under the national poverty line.
Year 2009 Year 2008 Year 2007 People Living under Poverty line Year
Bangladesh 10.23% 13.84% 14.59% 49.60% 2005 Cambodia 7.50% 7.20% 5.58% 25.60% 2007 China 0.12% 0.00% 0.00% 15.90% 2005 East Timor 1.18% 1.06% 1.33% 37.20% 2007 India 2.30% 1.47% 0.91% 41.60% 2005 Indonesia 0.12% 2.09% 1.66% 29.40% 2007 Laos - 0.08% 0.01% 44% 2002 Malaysia 0.75% 0.61% - 2% 2004 Nepal 2.00% 1.99% 1.68% 55.10% 2004 Pakistan 0.85% 0.98% 0.94% 22.60% 2005 Philippines 2.58% 2.35% 2.09% 22.60% 2006 Sri Lanka 4.48% 4.38% 3.63% 14% 2002 Thailand 0.01% 0.01% 0.01% 2% 2004 Vietnam 8.92% 8.14% 6.85% 21.50% 2006
Table 2: Percentage of People Reached and National Poverty Line per Country
While an MFI might be focusing on a large range of clients, this can be difficult to realize. Besides the before mentioned problems of mediocre infrastructure, the poor living in rural areas and the market penetration, another explanation of why this is difficult might include self-exclusion. That is, the poor just may not be interested in microfinance, find it to risky or do not possess all the necessary information (Meyer, 2002).
2.3.1 Microsavings
Depth of outreach
23
people are argued to have a high need for some type of insurance because of the relatively higher risks they face. Expanding the product range and offering savings will likely attract the poorest who are searching for a safe place to store their money. On the other hand, while the MFI is aiming to develop savings because they wish to involve the poorest in their client base, it can be difficult to actually reach these people. For instance, they might live in remote areas that are not yet reached by MFIs because of poor infrastructures.
Breadth of outreach
As mentioned above, the poor do have resources for savings. Subsequently, the poor are looking for possibilities to store their savings in a safer place (for example an MFI) than their current options (under the mattress). MFIs offering deposits are more likely to meet the actual demand of their clients and will be likely to expand their current client base and thus reach a higher breadth of outreach (Fiebig et al., 1999). Furthermore, clients depositing their savings at the MFI could be potential credit, insurance or BDS clients as well which could result in a larger client base (Morduch, 1999 and Churchill, 2002). Moreover, the deposit-MFI has access to internal funds. They could use these funds to improve their infrastructure and try to reach new clients in remote areas. Rossel-Cambier (2011) emphasizes that savings are the starting point of an MFIs growth strategy and consequently to a large breadth of outreach.
Hypothesis 7a: MFIs that offer micro credits and microsavings are expected to have a higher social
performance than credit-only MFIs
Hypothesis 7b: MFIs that offer micro credits and microsavings are expected to have a lower social
performance than credit-only MFIs
2.3.2 Microinsurance
Depth of outreach
With adding microinsurance to the product range, the social performance in terms of depth of outreach might increase since the MFI will be able to reach poorer and riskier clients (Siegel et al., 2001 and Mosley, 2009). Poudyal (2008) argues that insurance is in fact needed to reach the poorest of the poor, because the demand amongst them is highest.
important indicator for the degree of the interest rate, and in this case the chances of repayment are increased, the MFI could lower its interest rates. Consequently, a lower interest rate attracts poorer people and encourages a higher depth of outreach for an insurance-MFI. It must be noted that while the poorest of the poor are the people most in need of forms of insurance they are sometimes excluded from these programs because MFIs try to protect themselves against the effects of adverse selection (Jütting, 2000). As a solution, to ensure that the poorest can be reached, flexible premiums fitting the flexible income of the client should be offered, since poor people’s income is characterized as uncertain and irregular (Ahuja and Guha-Khasnobis, 2005; Meyer, 2002 and Mechler et al., 2006).
Breadth of outreach
As explained above, microinsurance is a highly demanded product by the poor. Offering microinsurance could thus increase the breadth of outreach of the MFI. Offering insurance to poorer and riskier clients could lower the overall risk of the client and they could become more attractive for the MFI and the other financial services offered. Hence, the MFI will have an opportunity to expand its client base. The possible lower interest rates on credit mentioned above for an MFI offering insurance might not only increase the depth of outreach but possibly also the breadth of outreach. Lower interest rates will more likely attract more potential clients. In comparison, a credit-only MFI can only offer higher, and thus unattractive, interest rates to their clients (Ahuja and Guha-Khasnobis, 2005). The flexible premium insurance scheme mentioned earlier could similarly stimulate the breadth of outreach of the MFI (Ahuja and Guha-Khasnobis, 2005). Nevertheless, the mentioned problems resulting from a mediocre infrastructure, possible clients living in remote areas and low market penetration could negatively influence the breadth of outreach.
Hypothesis 8a: MFIs offering microcredits and microinsurance are expected to have a higher social
performance than credit-only MFIs
Hypothesis 8b: MFIs offering microcredits and microinsurance are expected to have a lower social
performance than credit-only MFIs
2.3.3 Business Development Services
25
argument counts for BDS as for offering any additional service; it will attract new clients with needs for training (Sievers and Vandenberg, 2007 and Karlan and Valdivia, 2010). Moreover, Poudyal (2008) argues the MFI might face dropouts if the clients do not receive training. A client following the training and developing capabilities is more likely to repay the loan and maybe even step up to larger loan sizes because training usually leads to profitable businesses (Nourse, 2001).
Hypothesis 9a: MFIs that offer micro credits and BDS are expected to have a higher social
performance compared to credit-only MFIs
Hypothesis 9b: MFIs that offer micro credits and BDS are expected to have a lower social
performance compared to credit-only MFIs
2.3.4 All Financial Services
Depth of outreach
Meyer (2002), Matin et al. (2002) and Zeller and Johanssen (2008) propose that by broadening the existing product line with insurance, savings and BDS, more and poorer people can be reached because it complies with their needs. One of the reasons why it is hard to reach the poorest is because they are not able to pay high interest rates (Ahuja and Guha-Khasnobis, 2005). MFIs providing more than one financial service might be able to offer lower interest rates than credit-only MFIs. Because of economies of scale and scope an MFI that offers multiple financial services can realize relatively lower costs than credit-only MFIs. These lower costs could result in lower interest rates that probably will attract poorer clients (Meyer, 2002 and Rossel-Cambier, 2011). Another reason for possible lower interest rates could be the MFIs is ability to cover some of its risks with the availability of collateral (Matin et al., 2002). Moreover, with a full product range the MFI is in the best possible state to meet the needs of the poorest and thus to reach them.
Breadth of outreach
Hypothesis 10a: MFIs that offer micro credits, microsavings, microinsurance and BDS are expected
to have a higher social performance than credit-only MFIs
Hypothesis 10b: MFIs that offer micro credits, microsavings, microinsurance and BDS are expected
27 3. RESEARCH DESIGN
3.1 Research Methodology
The topic of this research focuses on whether offering additional financial services like microsavings, BDS and microinsurance has an impact on the financial and social performance of MFIs in Asia. In the following paragraphs there will be respectively elaborated on independent and dependent variables, methods, data collection, data source, sample selection and design of this research.
3.1.1 Variables
The variables consist of independent variables, dependent variables and control variables. There will be an elaboration on these variables in the following section.
3.1.1.1 Independent Variables
The independent variables are the variables that might, based on the literature, have an influence on the dependent variables. The first independent variable is ‘MFIs offering only microsavings besides
microcredit’ (SAVINGS). The second independent variable is ‘MFIs offering only microinsurance besides microcredit’ (INSURANCE). The third is ‘MFIs offering only Business Development Services besides microcredit’ (BDS). The fourth is ‘MFIs offering only microcredit’ (CREDIT-ONLY). The
fifth is ‘MFIs offering all financial services’ (ALL SERVICES). The sixth independent variable is ‘MFIs offering only compulsory savings besides microcredit’ (COMPULSORY SAVINGS). The seventh is ‘MFIs offering only voluntary savings besides microcredit’ (VOLUNTARY SAVINGS). The eight is ‘MFIs offering only savings and insurance besides microcredit’ (SAVINGS AND INSURANCE). The ninth is ‘MFIs offering only savings and BDS besides microcredit’ (SAVINGS AND BDS) and the last is ‘MFIs offering only insurance and BDS besides microcredit’ (INSURANCE AND BDS)
MIXmarket has information on the availability of the financial services within MFIs. In the case of BDS and microinsurance only the availability of the service within the MFI is obtainable and therefore used to answer the hypotheses concerning these financial services. For all the types of savings also the total amounts of the savings are presented in the database.
costs and thus achieve a higher financial performance. Moreover, the authors find that in some economies lower interest rates are charged (government rule) which motivates a higher depth of outreach. The countries (BANGLADESH, CAMBODIA, CHINA, EAST-TIMOR, INDIA, INDONESIA, LAOS, MALAYSIA, NEPAL, PAKISTAN, PHILIPIPPINES, SRI LANKA, THAILAND, VIETNAM) are included in the sample.
Second, the legal status will be added to the sample as a control variable. Based on previous research in Latin America, Tchakoute-Tchuigoua (2010) argues there might be a relationship between the legal status of a MFI and their performance. Nevertheless, the author cannot be strengthen his statement with empirical evidence. It is interesting to put the legal statuses of MFIs in the sample because there might be a link to performance. For instance, banks and NBFIs are more profit-oriented than for instance NGOs. Moreover, rural banks are more actively targeting the poorest. MFI types used in the analysis are Rural Banks (RURAL BANK), Cooperatives / Credit Unions (COOPERATIVES / CREDIT UNIONS), NGOs (NGO), Banks (BANK) and NBFIs (NBFI) (www.mixmarket.org and Zeller and Johanssen, 2008). Rural Banks are owned by their members and mostly supported by international NGOs with a high poverty alleviation focus. Cooperatives / Credit Unions are also owned by their members but typically have a profit focus. NGOs, Banks and NBFIs usually have more financial resources because of their large size and/or their international back up. NGOs, because of their financial status they are known to be active in the poorer strata (Morduch, 1999 and Zeller and Johanssen, 2008).
3.1.1.2 Dependent Variables Financial Performance
The dependent variables are the financial performance and social performance. The financial performance, in USD, is measured in terms of the ROA before taxes and donations (btd), ROA after taxes and donations (atd) and the ROE (Schreiner, 2001; MicroBanking Bulletin, 2010 and www.uncdf.org). These scholars argue that the ROA and ROE are relevant tools because these ratios will make it possible to compare companies in the same industry with different company sizes (size measured in terms of assets and equity respectively). For robustness reasons three measures of financial performance are used. The ROA computes to what extent the MFI has utilized its assets to generate profits. The ROA before taxes and donations (btd) is also used because in this ratio the MFIs results from operations are captured. Returns before taxes and donations reflect the income a MFI can generate from its own operations and thus better reflects the MFIs financial viability. The ROA is calculated as follows:
29
ROA btd (ROA BTD) = (Operating Profits + Taxes – Donations) / Total Assets
The ROE calculates how efficient investments from shareholders are used for generating profits:
ROE (ROE) = Operating Profits / Total Equity
There are some limitations when using the ROA and the ROE. For instance, the financial numbers are static and measured at the end of a booking year on the balance sheet or the income statement. Some factors, such as abnormal income or extraordinary costs in a given year could have influenced these static numbers for one year. To cover this limitation in the sample I will look at three book years in a row: 2007, 2008 an 2009 (Schreiner, 2001 and www.uncdf.org). To answer my hypotheses concerning the financial performance I will analyze whether MFIs offering an additional financial service have a higher average ROA atd, ROA btd and ROE compared to MFIs that do not offer the financial service.
Social Performance
The social performance of an MFI could be measured using several Social Performance Measurement tools. For more information see: Zeller et al. (2003), Seepnetwork (2007) and Hashemi et al. (2007) These tools consist of various indicators, the most commonly used are depth of outreach and breadth of outreach. I will use both depth and breadth of outreach in my research as indicators to measure social performance. The depth of outreach is described as to what extent the MFI reaches the poorest of the poor (Hashemi et al., 2007 and Lapenu and Zeller, 2001). The depth of outreach can be measured by using the ‘average loan size’. The average loan size can be described as the average amount of one loan (Schreiner, 2001 and Lapenu and Zeller, 2001). The average loan size formula goes as follows:
Average Loan Size (LOANSIZE) = Gross Loan Portfolio of an MFI / Total Borrowers of an MFI
The ‘average loan size’ measurement assumes that on average smaller loan sizes means that poorer people are reached by the MFI, because poorer people require smaller loan sizes. A limitation of this measurement is the fact that it is an average of the loan size where the few rich clients influence the average and pull it up (Schreiner, 2001). To answer my hypotheses I will investigate whether MFIs offering an additional financial service have, on average, a lower average balance of the loans than MFIs that do not offer the financial service.
will answer my hypotheses concerning breadth of outreach.
3.1.2 Methods
To answer my hypotheses concerning the financial performance I will compare the average ROA (btd and atd) and the ROE between MFIs offering everything, offering only one specific product, a specific combination between products, or not offering the additional financial services at all and analyze the differences between those means. Concerning the social performance this comparison is made with the average loan size and the total number of borrowers.
Finally, a multiple regression (multivariate analysis) is done. The regression will have to show whether the independent variables (savings, insurance, BDS, credit-only, all services, compulsory savings, voluntary savings) have a significance influence on the dependent variables (financial and social performance). From the regression conclusions could be drawn on whether offering certain services could positively or negatively influence the social and financial performance of a MFI. In the regression, the influence of the independent variables will also be measured independently of each other. The standardized equation for the multiple regression:
!!= ! + !!!!!+ !!!!!+ ⋯ + !!!!"+ !!
Where !! is the dependent variable. !! represents the ROA atd, ROA btd and the ROE when measuring the influence on the financial performance. For the social performance !! represents the depth of outreach (loan size) and the breadth of outreach (total borrowers). !! stands for the independent variables included, I expect that these independent variables have a significant influence on the dependent variable !! (ROA ATD, ROA BTD, ROE, BORROWERS, LOANSIZE). !! in the sample represent: SAVINGS, INSURANCE, BDS, CREDIT-ONLY, ALL SERVICES, COMPULSORY SAVINGS, VOLUNTARY SAVINGS, SAVINGS AND INSURANCE, SAVINGS AND BDS, INSURANCE AND BDS, RURAL BANK, COOPERATIVES / CREDIT UNIONS, NGO, BANK, NBFI, BANGLADESH, CAMBODIA, CHINA, EAST-TIMOR, INDIA, INDONESIA, LAOS, MALAYSIA, NEPAL, PAKISTAN, PHILIPIPPINES, SRI LANKA, THAILAND, VIETNAM. ! is the constant, also referred to as the regression intercept. !! is the regression coefficient and determines the influence of each independent variable !! on the dependent variable and !! is the estimation error. In all cases the significance level is set on 1%, 5% and 10%.
Regression per Legal Status
31
BDS Insurance Savings Total Observations
NGO 65% 48% 44% 415
NBFI 16% 29% 18% 224
Rural Bank 7% 10% 27% 190
Credit Union / Cooperative 9% 9% 7% 53
Bank 3% 4% 4% 29
Table 3: Legal Status and Financial Service Offered
As can be seen in the table, NGOs are largely offering BDS and are also largely active in insurance and savings. NBFIs also largely offer insurance and Rural Banks are active in offering savings. Therefore, a separate multiple regression analysis is done per legal statuses. In this multiple regression analysis it is researched whether a significant relationship exists between a MFI with a certain legal status offering a mix of additional products and the social and financial performance.
First we will construct a model (model 1) consisting of MFIs with all possible combinations of financial services (all the independent variables), one dependent variable and the control variables. This model will be repeated for each dependent variable and will also be done per year (2007, 2008, 2009) to control for possible yearly influences. Second, to control for multicollinearity (two or more independent variables could be correlating), models are constructed with only one independent variable at a time, one dependent variable at a time and the control variables (models 2-6). Finally, models will be constructed per legal status (only: NGO, NBFI, RURAL BANK because there is sufficient data) with the independent variables, the dependent variables and the control variables (models 7-9). For an outlay of the models see appendix 1.
3.2 Data
In the following section the data collection, sample selection and the limitations of the data are described.
3.2.1 Data Collection
Diamonds Description 1 diamond General Information
2 diamonds 1 diamond + outreach information (minimum of 2 successive years) 3 diamonds 2 diamonds + financial information (minimum of 2 successive years)
4 diamonds 3 diamonds + audited financial statements comprising the auditors opinion (minimum of 2 successive
years)
5 diamonds 4 diamonds + due diligence review (minimum of 1 review for 2 years)
Table 4: Description Diamonds (source: www.mixmarket.org)
In total 946 observations are made in the years 2007, 2008 and 2009. Since only the MFIs ranked with 3 diamonds or more are used in the analysis; the total number of observations is 857. Furthermore, it is possible that for 1 MFI data was available and thus used for all three years (2007, 2008, 2009) only two years (for example 2007, 2009) or just one year (2008). Therefore, the data is not equally distributed over the three years. The year 2007 has 315 observations, 343 observations for 2008 and
288 observations for 2009.
Country Obs. MFI 3yrs MFI 2yrs MFI 1yr Country Obs. MFI 3yrs MFI 2yrs MFI 1yr
Bangladesh 89 27 3 2 Nepal 77 16 11 7
Cambodia 44 14 1 0 Pakistan 62 16 7 0
China 18 5 2 1 Philippines 176 47 11 13
East Timor 5 1 0 2 Sri Lanka 42 10 6 1
India 216 57 17 11 Thailand 3 1 0 0
Indonesia 85 15 14 12 Vietnam 35 11 1 0
Laos 2 0 1 0 TOTAL 857 221 74 49
Malaysia 3 1 0 0
Table 5: Observations and number of MFIs divided by country
The data extracted from MIXmarket is focused on MFIs in Asia. Worldwide, more than 85 countries are served by MFIs and 70 percent of all MFIs worldwide are concentrated in Asia. Moreover, MFIs in Asia have the highest share of members worldwide, the highest percentage of outstanding credit and savings worldwide. Asian MFIs also have the highest social performance in terms of depth and breadth of outreach (Lapenu and Zeller, 2001 and Montgomery and Weiss, 2005). Besides the unequal distribution of the observations and the number of MFIs, the number of MFIs and the number of observations in the sample are also unequally distributed among the countries (table 5).
Figure 2 shows the distribution of the data among the independent variables in the sample.
Figure 2: Observations per Financial Service
All services, 117 Credit-Only, 93 BDS, 54 Comp. Savings, 100 Vol. Savings, 103 Savings, 316 Savings and Insurance, 117 Savings and BDS, 99 BDS and
Insurance, 27 Insurance, 34 Observations per Financial Service All services Credit-Only BDS Comp. Savings Vol. Savings Savings
33 3.2.2 Limitations
4. RESULTS, ANALYSIS AND DISCUSSION
In this section the results from the tests will be presented. First the results of the average financial and social performance analysis will be given, analyzed and discussed for MFIs offering the different financial services. Subsequently, the results of the multiple regression will have to show whether the independent variables are significantly related to the financial and social performance of the MFI.
4.1 Average Financial Performance
Table 6 shows and compares the outcomes of the average financial performance determined in ROA atd, ROA btd and ROE of MFIs offering only credit, only microinsurance, only BDS, only microsavings or a combination. The outcomes are ranked in order of the highest average financial performance (1) to the lowest financial performance (10). The ANOVA table (see: appendix 2) explains that the results between the groups (the financial services) for the ROA atd and ROA btd are significant. This indicates that the results in the table are representative and conclusions based on the results regarding whether offering additional services besides microcredit will result in a higher or lower average financial performance can be drawn.
If we compare the financial performance of credit-only MFIs (ROA atd: -0.3%, ROA btd: -0.6% and ROE: 7.4%) with MFIs offering all services (ROA atd: 0.1%, ROA btd: 0.9% and ROE: 15.1%) we can see that the financial performance for MFIs offering all services is higher than for credit-only MFIs. The same counts for MFIs offering microsavings additional to microcredits (ROA atd: 2.2%, ROA btd: 2.1% and ROE: 27.1%), MFIs offering savings and insurance (ROA atd: 1.6%, ROA btd: 1.49% and ROE: 6.05%), MFIs offering savings and BDS (ROA atd: 0.49%, ROA btd: 0.3% and ROE: 16.49%) and MFIs offering BDS and insurance (ROA atd: 1.72%, ROA btd: 0.82% and ROE: 9.72%).
Av. ROA atd St. Dev. Av. ROA btd St. Dev Av. ROE St. Dev.
1 Savings 2.24% 0.05 2.10% 0.06 27.06% 2.16
2 Compulsory Savings 1.85% 0.04 2.16% 0.05 19.12% 0.43
3 Voluntary Savings 1.62% 0.06 1.70% 0.08 43.05% 3.67
4 BDS and Insurance 1.72% 0.01 0.82% 0.01 9.72% 0.06
5 Savings and Insurance 1.6% 0.01 1.49% 0.01 6.05% 0.07
6 All Services 0.09% 0.11 0.87% 0.07 15.09% 0.92 7 Savings and BDS 0.49% 0.01 0.3% 0.01 16.49% 0.07 8 BDS -1.80% 0.13 0.57% 0.13 -8.55% 1.27 9 Credit-Only -0.33% 0.10 -0.61% 0.12 7.35% 0.54 10 Insurance -6.67% 0.23 -3.72% 0.16 -37.56% 2.41 Total 0.68% 0.10 1.03% 0.09 15.78% 1.78
35
financial performance can be reached for MFIs offering voluntary savings to their clients (ROA atd: 1.6%, ROA btd: 1.7% and ROE: 43.1%) compared to offering exclusively microcredits. Furthermore, offering microinsurance will result in a lower average financial performance for MFIs (ROA atd: -6.7%, ROA btd: -3.7% and ROE: -37.6%). Finally, MFIs offering BDS as an additional service will experience a similar average financial performance than credit-only MFIs (ROA atd: -1.8%, ROA btd: 0.6% and ROE: -8.6%).
The average financial performance is highest for MFIs offering only savings, voluntary savings and compulsory savings. This makes sense since these MFIs have the opportunity to use internal financing possibilities that are normally cheaper than external finances. Furthermore, savings secures the MFIs with a source of collateral, this will reduce some of the risks the MFI faces. The collateral likewise generates higher retention and payback rates which stimulates the financial performance as well.
Subsequently, the MFIs offering a specific combination of financial services (‘savings and BDS’, ‘savings and insurance’, ‘BDS and insurance’ and all financial services) all present a positive average financial performance. Notable from this finding is the positive average financial performance of MFIs offering a combination of BDS and insurance given the fact that MFIs offering only insurance or only BDS recognize a negative mean financial performance. Offering credit-only generates a negative average financial performance. As mentioned, BDS MFIs and microinsurance MFIs experience an overall negative financial performance. For microinsurance, the relatively high costs (for example: transaction costs, information asymmetries, adverse selection, moral hazard) overshadow the possible profits. In the case of BDS the results on the financial performance were expected to be higher. Apparently costs for administrations, operations and training employees are to high to make MFIs providing BDS more profitable (or profitable at all).
4.2 Regression Financial Performance
Since all the independent variables in the sample are dummy’s we take one independent variable out. In the case of financial services I take out ‘BDS and Insurance’, for legal status I take out ‘NBFI’ and for country I take out ‘East-Timor’.
predicting the future will likely negatively influence financial performance. Offering insurance is an important yet risky business nowadays, more experience and research is necessary to develop profitable insurance schemes within MFIs.
ROA btd ROA atd ROE
Beta Sign. Beta Sign. Beta Sign.
MODEL 1: !!=0.107 !!=0.114 !!=0.022 N = 831 N = 786 N = 786 Insurance -0.143 0.007a -0.204 0.00a -0.063 0.263 BDS 0.068 0.236 -0.021 0.716 -0.001 0.982 Savings -0.008 0.947 -0.012 0.921 0.082 0.516 All Services 0.003 0.975 -0.065 0.437 0.057 0.517 Credit-Only -0.061 0.396 -0.068 0.357 0.002 0.979
Savings and Insurance -0.003 0.971 -0.015 0.871 0.027 0.779
Savings and BDS 0.006 0.939 -0.026 0.735 0.061 0.453 Voluntary Savings 0.041 0.367 0.056 0.224 0.008 0.871 Compulsory Savings 0.03 0.498 0.008 0.868 -0.007 0.877 MODEL 2 !!=0.082 !!=0.082 !!=0.015 N=831 N=786 N=786 Savings 0.048 0.249 0.093 0.03b 0.055 0.217 MODEL 3 !!=0.097 !!=0.105 !!=0.018 N=831 N=786 N=786 Insurance -0.136 0.000a -0.182 0.00a -0.075 0.052c MODEL 4 !!=0.084 !!=0.076 !!=0.014 N=831 N=786 N=786 BDS 0.066 0.074c 0.002 0.961 -0.025 0.527 MODEL 5: !!=0.083 !!=0.077 !!=0.013 N=831 N=786 N=786 Credit-Only -0.046 0.217 -0.024 0.526 -0.013 0.74 MODEL 6: !!=0.081 !!=0.078 !!=0.013 N=831 N=786 N=783 All Services 0.001 0.975 -0.04 0.287 0.013 0.737 MODEL 7: !!=0.084 !!=0.079 !!=0.014 N=831 N=786 N=786 Compulsory Savings 0.061 0.11 0.054 0.169 0.031 0.45 MODEL 8: !!=0.083 !!=0.081 !!=0.014 N=831 N=786 N=786 Voluntary Savings 0.058 0.177 0.089 0.04b 0.036 0.424
Table 7: Regression Financial Performance
a: significant at 1%, b: significant at 5%, c: significant at 10%
Country and Legal Status Dummy’s: yes (see: appendix 3 and appendix 6)
37
retention rate, higher payback rates, a source of collateral of the client, cheaper internal financing possibilities and economies of scope and scale on its financial performance.
The same applies for offering BDS (model 4) besides credit. MFIs offering only BDS have a significant positive relationship with the ROA btd (0.066). While there is only one significant financial performance indicator we could assume a positive influence of offering BDS on the financial performance in this sample. MFIs providing BDS benefit because of higher retention rates, larger loan sizes assuring higher returns, fewer dropouts, direct income from fees, economies of scale and scope, higher repayment rates and better competitiveness.
Since the data has a panel structure the regressions between financial performance and the financial services are also done per year (see appendix 4). Not many differences are witnessed. The most important finding is that offering only microinsurance is significantly negative related to the financial performance in the years 2007 and 2009 (respectively -0.257 and -0.431) and not in 2008.
4.3 Average Social Performance Depth of outreach
When analyzing the ranked table 8 it shows that on average, the highest depth of outreach will be reached by MFIs offering BDS and insurance (USD 96.19) followed by MFIs offering solely microinsurance (USD 136.86). MFIs providing BDS and credit-only MFIs follow closely with depth of outreaches of respectively USD 167.09 and USD 169.17.
Average Loan Size (in USD) Standard Deviation
1 BDS and Insurance 96.19 9.25 2 Insurance 136.90 43.62 3 BDS 167.09 126.31 4 Credit-Only 169.17 127.71 5 All Services 172.80 149.76 6 Compulsory Savings 183.73 123.78
7 Savings and Insurance 227.90 21.95
8 Savings 256.76 181.95
9 Savings and BDS 281.50 56.09
10 Voluntary Savings 365.90 183.26
Total 210.45 163.76
Table 8: Average Depth of outreach
