The Demerits of Microcredit
University of Groningen
Faculty of Economics
International Economics and Business
Date:
August 2006
Author:
Quirine Royaards
Student number: 1198947
Supervisors:
Prof. Dr. B.W. Lensink
Dr. C.L.M. Hermes
Co-assessor:
Dr. G.J. Lanjouw
PREFACE
This thesis reviews the demerits that arise from microcredit programmes and assesses their significance through a case study with institutional investors in the Netherlands. It is the final part of the Master studies International Economics and Business that I studied in Groningen, the Netherlands.
During my internship at the Permanent Mission of the Kingdom of the Netherlands to the United Nations in New York I first came across the phenomenon of microfinance and was intrigued from the first moment. In March 2006, I worked one week at microfinance institution Crédimujer in San José, Costa Rica. In their jeep, we travelled all over the country to visit many different clients. This experience heightened my interest in the microfinance sector and I decided to study the movement more intently. Though this thesis has a critical point of view, let me assure the reader that I look upon microfinance as a most triumphant phenomenon. However, through this paper, I underscore that the world must not lose itself in its own enthusiasm. In addition, I hope to contribute to the development of the sector and to the correct targeting of the poor.
I would like to thank several people, who have been important to me in writing this paper: my brother-in-law who helped arrange my contact with MFI Crédimujer and, of course, Crédimujer for kindly and enthusiastically engaging me for the wonderful short time I spent there. I am also grateful for the 14 investors whom I interviewed and who thereby contributed greatly to this thesis. Furthermore, my 2 supervisors, Prof. Dr. B.W. Lensink and Dr. C.L.M. Hermes, who supported and advised me through the process. They were quick in reading my drafts and giving me advice, and very helpful at all times. Last, nut not least, I would like to thank my boyfriend, friends and family who dealt with me whenever I was tired or discontented after a long day studying. In the end, however, I am pleased with the outcome and hope the reader will be as well.
ABSTRACT
CONTENT
SECTION 1. INTRODUCTION 7
SECTION 2. MICROFINANCE 9
2.1 What is microfinance? 9
Vicious poverty circle 9
Concept 10
Background 10
Microcredit 11
Other microfinance instruments 12
2.2 Positive and Problematic Aspects of Microfinance 13
Positive aspects 13
Problematic aspects 14
SECTION 3. POVERTY ALLEVIATION 16
3.1 Introduction 16
3.2 Target Group: The Very Poor 16
Concept 16
Inequality among the poor 17
3.3 Microcredit Instruments 19
3.3.1 Individual Loans 19
Debt burden 19
Repeat loans 20
Savings as collateral 20
Replicating microcredit programmes 21
3.3.2 Group Lending 21
ROSCAs 22
Ex-Ante Moral Hazard 22
Ex-Post Moral Hazard 24
SECTION 4. GENDER AND SECTOR DEVELOPMENTS 27
4.1 Gender 27
Concept 27
Loan control 28
Work activity 29
Financial intermediaries 30
4.2 The Microfinance Industry 30
Concept 30
Competition 31
Supervision and regulation 32
Subsidization 34
Commercial banks 36
SECTION 5. CASE STUDY: INTERVIEWS 37
5.1 Methodology 37
5.2.1 Poverty Alleviation 39
Target group: the very poor 39
Microcredit instrument: individual loans 41
Microcredit instrument: group loans 43
5.2.2 Gender and Sector Developments 46
Gender 46
The microfinance sector 47
SECTION 6. CONCLUSION 51
6.1. Conclusion and Recommendations 51
6.2 Limitations and Future Research 54
BIBLIOGRAPHY 57
SECTION 1. INTRODUCTION
In recent years the success of microfinance programs has been broadcasted broadly, captivating not just the entire development aid industry, but also journalists, editorial writers, policy makers and much of the general public. Donors are pledging billions of dollars1 into the development and as a consequence, investment funds, such as the ASN-Novib Fund in the Netherlands, are having trouble locating suitable investment opportunities for the rapidly mounting financial resources. The enthusiasm is due to the triumphs of microfinance. Microfinance provides financial services to segments of the population, mainly the poor and the low income, where it was previously unheard of. As a result it has alleviated poverty and made it more difficult for local moneylenders to exploit the poor with extreme loan conditions, such as excessively high interest rates. Moreover, by targeting females, the programs have, in some instances, contributed to women’s empowerment. Microfinance is an extraordinary development tool by aiding the poor through promoting enterprises instead of giving charity (Khawari, 2004; Littlefield, Morduch and Hashemi, 2003).
However, at times microfinance seems to have earned an almost mythical reputation as the panacea for the global poverty problem and may even be bordering hysteria. It seems that donors and advocates regularly over exaggerate the powers of microfinance without placing attention to occurred mishaps. This is augmented by the numerous studies recounting the great accomplishments as opposed to the relatively few dedicated to finding possible critical areas (Chowdhury, 2005; Hulme and Mosley, 1998; Scully, 2004). Especially the instrument of microfinance from which the movement originated several decades ago, microcredit, could have received more constructive attention than has currently occurred. Therefore, this thesis focuses on the problems surrounding microcredit and, as such, can be used as a first framework for further critical research.
The thesis will be divided in two parts, a literature review and an empirical case study on investors in the Netherlands. The key question is: What is the significance of the demerits surrounding microcredit programs? In order to generate a list of demerits the subsequent sections comprise a literature review which focuses on circumstances where microcredit led to critical outcomes. It first concentrates on instances where microcredit did not lead to poverty alleviation, the ultimate goal of microcredit. Secondly, it addresses the influences of specifically targeting females and the recent changes in the sector on the performance of microcredit. Since these demerits will have been extracted from literature and are subject to regional restraints and other limitations, they remain an overview and do not clarify which occur often and which are particularly problematic. As a means to measure their importance, the second part of this paper is a case study with a sample defined by 12 organizations from the Netherlands investing in microfinance activities. The list of demerits extracted from the literature review will be a tool in assessing the participants’ opinions and experiences. Interviews with each
1
institution will take place and the responds will be compared with one another and with literature. After constructing the list of demerits and discovering the most problematic ones, the paper also recommends for whom among the poor microcredit is most appropriate.
This paper will continue as follows. The next section, section 2, will explain the notion of microfinance, its background and its instruments. Moreover, it will provide a short summary of the many positive developments due to microfinance and explain why it is important to focus on its critical aspects as well. Section 3 will give rise as to why and in which cases do programs not lead to poverty alleviation. It will clarify why the poorest among the poor are not always reached. The chapter also focuses on microcredit instruments and reports the possible negative externalities they might provoke which lead to failed poverty alleviation. Section 4 will deal with the demerits of related developments, such as the specific targeting of women. Influence of recent changes in the sector itself, such as competition, regulation and subsidization, will also be studied. Section 5 will comprise the case study of interviewing 12 institutional investors in the Netherlands. The meetings will be compared and the opinions, experiences and actions of investors regarding demerits of microcredit will be described. Section 6 will conclude the thesis as a whole and identify limitations and future issues.
SECTION 2. MICROFINANCE
2.1 What is microfinance? Vicious poverty circle
“Money, says the proverb, makes money. When you have got a little, it is often easy to get more. The great difficulty is to get that little.” 2
This proverb relates to the ‘vicious poverty circle’ in emerging countries. In these countries most per capita incomes have to be spent to meet current, often urgent, needs. These low incomes lead to low per capita savings, which in turn hinders investment in both human and physical capital. Without new investment, neither productivity nor incomes can be increased, thus perpetrating the vicious poverty circle shown in figure 2.1.
Source: Soubbotina, T. and Sheram, K. 2000, Chapter 6, p3
formal contracts, let alone filling out application forms and preparing a financial plan, are alien to them (Woolcock, 1999). Conventionally, bankers have always seen the deprived as unbankable.
Concept
So are poor people doomed to remain poor? Microfinance offers a promising solution to break the vicious poverty chain. The birth of microfinance was rooted in unsatisfied credit demand. Providing financial access to the poor started in the 1970s and has since achieved some astonishing results. At the outset of its launch, microfinance was built only on microcredit services. Microcredit refers to a means of providing poor families with very small loans to help them engage in productive activities or expand their tiny businesses (FAQs Microfinancegateway.org). In the 1990s this term was gradually replaced by microfinance, indicating that a broader range of services began to play a more prominent role as it was realized that the poor require a variety of financial products. Since then microfinance has turned into a much researched discipline. Yet there is hardly any agreement on a universally accepted definition of microfinance, with researchers and microfinance visionaries divided in their opinions on its range and targeted recipients. As Sriram and Upadhyayula (2002, p3) note, "it appears that what microfinance means is well understood, but not clearly articulated". However, generally, microfinance is an umbrella term that refers to the provision of a broad range of financial services that include loans for businesses and personal use, savings and other deposit products, remittances and transfers, payment services, insurance, financial advisory services and potentially any financial product or service a bank can offer to this market segment. In a much narrower sense, microfinance is often referred to as microcredit for tiny informal businesses of microentrepreneurs. (Christen, Lyman & Rosenberg, 2003). In order to limit the cumbersome repetition of ‘microcredit’, in this thesis the term microfinance refers to microcredit, the offering of small loans or lines of credit to the deprived. Unless otherwise stated, this paper refers only to microcredit in developing countries. Microfinance in rich countries may present different issues.
Background
Badan Kredit Desa in Indonesia, and FINCA Village Banks world-wide.3 The table demonstrates several features of the MFIs, such as the number of clients, average loan amounts and terms, gender percentages, target groups and interest rates. The World Bank has estimated that over 7,000 institutions currently serve more than 16 million poor customers with a global turnover of over 2.5 billion dollars. These institutions include a range of financial service institutions, including development banks, foundations, multilateral banks, MFIs and even commercial banks that have microcredit subsidiaries (Arch, 2005).
In the past years, microfinance has received more attention than ever before. At the United Nations (UN) Microcredit Summit in 2000, the UN stated that microcredit would be a key strategy in reaching the Millennium Development Goal (MDG) of reducing poverty by half by 2015. Last year the UN declared 2005 as the International Year of Microcredit and sponsored research projects and meetings to encourage the use of microcredit. In addition, various major international banks are turning their funds towards this new market ‘niche’ of serving the poor and numerous new MFIs are springing up. Arch (2005) attributes the rising attention partially to the whopping repayment rates of over 97 percent, a growth rate of 30 percent and the view that the poor as a niche can be both profitable and morally satisfactory. Motivations stretch from pure profit on the one hand to social objectives on the other hand. The microfinance sector has matured into one of the most successful and fastest growing industries in the world.
Microcredit
Microcredit instruments can generally be divided into two broad categories of individual loans and group approaches, based on how the MFI delivers and guarantees its loans (Ledgerwood, 1999). Microcredit loans vary from as little as 10 to over 5,000 dollars and are generally used for income generating assets. Clients may start or expand businesses, though it is occasionally used to meet the financial consumption needs of the households. While the financial needs of these segments are similar to those of traditional businesses and consumers, their personal and business characteristics, as well as the relative size of their operations, diverge significantly. In terms of loans, this means reduced and simplified paperwork, formal collateral, and time involved to apply for and receive a loan. As traditional collateral and co-signers may not be available or the costs of legally registering and executing guarantees may be prohibitive relative to loan size, non-traditional collateral often is accepted. Conventionally, the size of a loan is based on the client’s fixed salary, whereas the microloan’s size is proposed after an evaluation of the business and neighbourhood. In addition, a microloan duration is much shorter than usual. Unlike traditional bank loans, the loans are provided at
3
full-cost interest rates that are repayable in frequent (monthly or even weekly) instalments, starting quickly after receiving the loan. Interest rates for microcredit are around 25 percent, but can be as high as 40 to 50 percent in some areas (Wharton, 2005). These high rates are due to the high transaction and administrative costs, such as monitoring, training, group meetings and high inflation rates, that MFIs face when providing miniscule loans.4 Costs can soar even more because MFIs’ field officers usually visit their clients, as opposed to clients visiting the institutions. High costs combined with high inflation drive real interest rates up even higher. Moreover, most microfinance clients have lower levels of education than traditional bank clients and require assistance in the application, information-gathering, and transaction processes of loans. The key differences between conventional loans and microloans are summarized in table B.1 in Appendix B.
Microcredit programmes frequently refer to the innovation of ‘group-lending contracts’, a mechanism of mutual responsibility across borrowers.5 Small amounts are granted to individuals that are part of the group and all group members are jointly responsible for repayment to the lender. This generates a strong cohesion among partners: a reliable personal guarantee instead of a real guarantee. The driver in this mechanism is that the more efficient a group of borrowers is in reimbursing, the higher the amount of the loans. The commitment of the members in demonstrating that they can reimburse the loans grows if it is a condition for future concession of bigger loans and if the default of one member has negative effects on the whole group (Ciravegna, 2005). Group-lending contracts may make a borrower’s neighbours co-signers to loans, which contribute to mitigating problems created by informational asymmetries between lender and borrower. That way, neighbours have incentives to monitor each other and to reject risky borrowers from participation. Hence, repayment can be promoted even in the absence of collateral requirements. According to Chowdhury (2005), the success of the group-lending model has led to its use all over the world, currently reaching over 8 to 10 million people.
Other microfinance instruments
Savings is one microfinance instrument that has had tremendous growing demand in the past years. Extensive evidence from around the world has shown that the poor do save and that they need secure places where they can build usefully large sums of money.6 Indeed, on the part of the poor there might be an even greater demand for secure savings than credit (UNDP, 1999). However, it takes time for the poor to trust that their savings are really secure and available to them at all times. In addition to microcredit and saving services, the poor need insurance to protect themselves from uncertainty such as personal misfortune (e.g. illness and theft) and misfortune that is common to the community (e.g.
4
If, for instance, the transaction and administrative costs were 25 dollars per loan, then for a loan of 50 dollars, this would amount to 50 percent. For a loan of 10,000, however, the costs would only be 0.25 percent.
5
Group lending may also be called ‘solidarity lending’, ‘peer lending, joint liability programmes and so on.
6
drought and flooding). Some MFIs have begun to offer insurance products, but examples are still relatively rare and MFIs have yet to tap the potential for sustainable insurance services (International Food Policy Research Institute, 2002). Both the savings and the insurance instruments are relatively new, as are other microfinance mechanisms like remittances and payment transfers. Therefore, since the vast majority of projects and studies still concentrates on credit only (Dichter, 2006), the focus of this paper will be on microcredit, specifically individual loans and group loans.
2.2 Positive and Problematic Aspects of Microfinance
Although the main focus of this report is on the critical aspects of microfinance, this sub-section will begin by describing some of the basic positive impacts. This is followed by a discussion about the danger of over-enthusiasm for microfinance and the importance of continuing to study the more critical effects.
Positive aspects
According to Datta (2004), empirical studies have shown that there are two main positive effects due to microfinance, namely poverty alleviation and women empowerment. Reducing poverty in particular, especially for the poorest, is a very often cited positive development. Studies (e.g. Littlefield, Morduch and Hashemi, 2003) affirm that access to financial services helps the poor protect, diversify and increase their income. It provides the opportunity to accumulate assets, to reduce vulnerability to shocks (e.g. illness, funeral expenses, dowries and crop failure) and to invest in income-generating activities. Furthermore, Hassan (2002) mentions that microcredit gives the poor dignity and self-esteem that comes from offering the poor hope and the opportunity to advance their lives as well as the lives of their children through their own labour. The quality of their lives is improved through better education, health and housing. The microfinance process is as much about finance as about education. While obtaining credit, borrowers receive schooling on health issues, gender roles, business practice and legal rights (Hassan, 2002; Morduch, 1999). This process improves their performance as microfinance clients, but also has important implications on the overall development of the lower class. Enthusiasts also point to repayment rates of over 95 percent, improvements in education of children, and the potential for programmes to become financially sustainable in the long-term (Woolcock, 1999).
reality that women are more reliable bets for banks worried about repayment as various studies assert (e.g. Cheston, S. and Kuhn, L. 2002), reaching females also has important implications for the social development of the country. Littlefield, Morduch and Hashemi (2003) observe that women become more assertive, more likely to contribute in family and community decisions and better able to deal with systematic gender inequities. Statistical studies provide further evidence that a microloan can boost a woman’s self-confidence and lift her status within the family, as an independent producer and provider of income in the household (Goetz and Gupta, 1996; Pitt and Khandker, 1998). A further social implication is that women have been shown to spend more of their income on their households and tend to be more concerned about their children’s health and education than men. Therefore, when women are helped to increase their incomes, the welfare of the whole family is improved.7 Her Royal Highness Princess Máxima of the Netherlands points out that this pertains as well for nieces, cousins and other relatives, resulting in a multiplier effect.8 Hence, microfinance acts not only as an economic stimulator for small enterprises, but also has far reaching social impact on poor women and their families.
Problematic aspects
Has the widespread enthusiasm for microfinance transformed a noble idea into a panacea?
Due to some of the great accomplishments described in the previous sub-section, it is clear that microcredit works for some. However, it is unclear to what extent it works and for whom, which is the main background to writing this paper. It seems that the hope has bred hype. The celebrated anecdotes are not a substitute for careful statistical evidence culled from large samples. As Chowdhury (2005, p3) writes: “There is possibly too much emphasis on the positive aspects of such schemes, and too little on the possible negative ones. This is somewhat surprising in view of the fact that several of these schemes performed poorly.” According to Buckley (1997) and Dichter (2006), to date, institutional innovation and experimentation have run ahead of serious analyses of client impact, the evidence for which often appears quite anecdotal or inconclusive. In addition, Buckley mentions, it is clear that the whole movement has been sustained by substantial donor finance. There is not much convincing evidence that microcredit leads to strong and thriving local economies or that it makes a real difference in the lives of the poorest, and yet the movement continues to engage in little serious empirical impact study. The few empirical studies executed so far do not always conclude that microfinance leads to poverty alleviation. Armendáriz de Aghion and Morduch (2005), for instance, declare that studies of SEWA Bank in India, Zambuko Trust in Zimbabwe, and Mibanco in Peru found that only in India and Peru borrowers had net income gains. Additionally, a study by Hulme and Mosley (1996, vol. 2) of Bolivia’s BancoSol reports that in any given cohort roughly 25 percent
7
For example, studies of Grameen Bank customers indicated that a 1 percent increase in credit provided to women raised the probability of girls enrolling at school by 1.86 percent and improved boys’ school enrolment by 2.4 percent (Hassan, 2002).
8
showed gains to borrowing, 60 to 65 percent remained the same as before and 10 to 15 percent went bankrupt. Other studies (e.g. Hunt and Kasynathan, 2001; Mayoux, 1997) find that targeting only women can actually have adverse negative effects on women empowerment. It can be questioned whether providing financial access increases welfare for the poor in all cases. This leads to the question of the literature review: for whom and in which cases does microcredit not work?
Apparently, alleviating poverty through banking is an old idea with a chequered past. Poverty alleviation through the provision of subsidized credit was a centrepiece of many countries’ development strategies from the early 1950s through the 1980s, but these experiences were nearly all disasters (Morduch, 1999). Dichter (2006) calls these previous complex approaches to development “band aid solutions” that hardly produced any actual changes and merely responded to the need of helping the poor. The realization that massive amounts of foreign aid invested in large projects did not lead to the expected trickle down effect led to the emergence of target group orientation, where donors started a wave of small, diverse projects making credit available to the poor (Khawari, 2004). The mechanisms for delivering financial access may have changed but the goals remain similar.
SECTION 3. POVERTY ALLEVIATION
3.1 Introduction
One of the most prominent goals of microfinance programmes concerns its impact on alleviating poverty in emerging countries and countries in transition. Studies (e.g. Khandker, 2005; Littlefield, Morduch and Hashemi, 2003) demonstrate that in some cases this goal has indeed been reached. However, other cases show that microfinance did not lead to reduced poverty, especially regarding the very poor. Several MFIs who claim the poorest as their target group are having troubles reaching and including them in their programmes and raising them above the so-called poverty line of 1 dollar a day. In addition, the frequently successful instruments used for microcredit, individual and group loans, have been prone to several struggles as well and have not always led to poverty alleviation either. Impacts depend critically on who the MFI is determined to reach and whether the instrument is adequately adapted. This section will review why and in which cases the poorest are not reached and microcredit instruments fail to work properly and therefore, poverty is not always lessened.
3.2 Target Group: The Very Poor Concept
However, some critical studies (e.g. Tranovich, 2005) question whether alleviating poverty through loan-financed enterprise expansion is more valid for the less poor among the poor, e.g. those who have certain skills and an entrepreneurial talent. The challenge of reaching the very deprived is to develop a set of institutions that can service their varying demands. There is a need for social security mechanisms, for instance, such as employment guarantee schemes, drought relief and food-for-work programmes. Financial access is only one element in this wide range of demands. In addition to the vulnerability of the very poor themselves, MFIs face problems with merely attracting the very poor and providing them this access. Even the famous and celebrated Grameen Bank, for example, has problems with reaching the very poor (Hulme and Mosley 1996, vol. 1). Moreover, the Consultative Group to Assist the Poor (CGAP), a large and well-known microfinance network, has changed its name; where the P first referred to ‘poorest’, it now only refers to ‘poor’.
Inequality among the poor
One problem with targeting the very poor is that it is difficult to reach or include them in microcredit programmes. The root of this problem can be found in adverse selection, which deals with problems of private information about the type of the borrower and the project. The adverse selection problem occurs when lenders cannot distinguish risky borrowers from safer borrowers and are thus unable to discriminate against risky borrowers. Poor loans to risky borrowers lead to excessively high interest rates, which drives worthy borrowers from the market. In addition, the problem is magnified by the extent of microborrowers’ limited liability. Since the poorest hardly have any assets, they suffer from adverse selection (Akerlof, 1970; Armendáriz de Aghion and Morduch, 2005). The moderate and less poor receive microloans and thus have the possibility to surface above the poverty line. However, as Scully (2004) notes, credit programmes seldom reach the poorest. Consequently, microcredit may lift some segments of the population while others fall farther behind, leading to the widening of social inequality among the poor.
Why are the very poor excluded from membership, making it so hard for MFIs to reach them? Simanowitz (2002) distinguishes four varieties of exclusion:
• Self-exclusion: people who exclude themselves because they are convinced they will not be accepted by financial institutions when requesting a certain financial service (Ciravegna, 2005). Poor people lack self-confidence which restricts their capacity to consider the programmes as beneficial to them. In addition, they are unwilling to burden themselves with the risk to be saddled with debt that may force them to sell what few assets they possess and become even worse off than before. According to Bosch (2004) the poor also have fewer social contacts making it more difficult to have information about possible loans. Vossen (2005) points out that people who are chronically underfed, have other priorities than setting up a small business.
liability is applied and prospective borrowers are encouraged to form groups on their own, deselecting any person who may be a bad credit risk (Armendáriz de Aghion and Morduch, 2005; Simanowitz, 2002). The community often has a negative perception of the poor, especially regarding their ability to repay loans, and as a result other members do not want them in their group (Hulme and Mosley, 1996, vol. 1; Kabeer and Mosley, 2004; Marr, 2004). In such cases, the very poor and most vulnerable members of the community are excluded from the programme. In addition, Marr (2004) states that the poorest from any given group leave the group prematurely. Nearly 70 percent of her participants in Peru, who left the programme before its end, were classified as “very poor” according to participants’ own wealth ranking. This phenomenon occurs because the very poor are more likely to fall behind with their payments and are more likely to suffer punishments because of their lesser ability to negotiate sanctions and loan repayments. They are generally not aware of their legal and civil rights, and are more easily intimidated by authority. Furthermore, according to Marr, the new group members that are selected come from less poor backgrounds than those of the original participants. Simanowitz (2002) stresses that there is a challenge for MFIs to ensure a culture which engages the participation of the extreme poor, but also offers support in order to ensure that the vulnerability of the poorest does not lead to their experiencing problems and exiting the programme.
• Exclusion by staff: field officers may have explicit or implicit incentives to reject the poorest, based on minimizing their amount of risk. According to Hulme and Mosley (1996, vol. 1), when credit programmes expand, the incentive structures for staff (bonus payments and promotion prospects) favour a concentration on groups other than the very poor. This is due to a perception that the poorest have more problems in repaying loans and create an increased work burden. For instance, in the case of the Bangladesh Rural Advancement Committee (BRAC)9, the average return on loans of new members admitted by staff was higher than those of successful third-time borrowers. Datta (2004) observes that the target of fulfilling the recovery rate is one of the most important issues facing field officers. As a result, loan officers form groups with the more prosperous members among the poor, thereby providing themselves with more insurance that the loans will be repaid while still selecting borrowers that fit within the definition of the poor.
• Exclusion by design: the design of a microfinance programme may exhibit aspects that act to exclude the poorest. These may include entry fees, minimum loan sizes, compulsory savings, minimum age, having an existing business or providing services from central offices rather than in community-based situations (Kirkpatrick, 2002; Mosley, 2001). Datta (2004) gives an example of the exclusionary criteria permanent residence of borrowers and refers to members of ASA and ActionAid in Bangladesh who lost their membership because they lost their houses due to riverbank erosion. Furthermore, Pretes (2002) and Vossen (2005) point out that many microfinance programmes, especially those in East Africa, focus on urban activities, forgoing the rural areas which are more dispersed and harder to reach and monitor. In urban areas borrowers are concentrated and more easily
9
reached, thus decreasing costs. However, the rural poor comprise most of the very poor, yet they have trouble reaching these urban MFIs.
A related problem regarding increased inequality is that the income generating effects of microcredit appear to be more pronounced on the less poor than on the very poor. Johnson and Rogaly (1997) write that borrowers who already have assets and skills are able to make better use of credit. A borrower that is better of, for instance by having more education, more investment opportunities or increased ability to take risks, can generate a greater increase in income from a microenterprise loan than someone less well-off.Often, the very poor are limited by their low skills and an inability to add value. According to Mosley (2001), a positive impact on poverty drops dramatically when moving down the income scale, which may be due to ineffectiveness of MFIs and the need of other ‘anti-poverty modalities’ when reaching the poorest. In his sample of nearly 200 MFIs in Bolivia over 6 years, only a single case was found where a MFI had managed to reduce extreme poverty.10
3.3 Microcredit Instruments
Reaching the poor and alleviating poverty is often tried through providing microcredit, the main instrument that this paper focuses on. This sub-section will first study the critical appraisals for individual loans and then carry on with group contracts. Please note that the aspects regarding the individual loans can frequently be accounted for group loans as well.
3.3.1 Individual Loans Debt burden
Microcredit loans frequently increase the income stream of the poor as different studies have demonstrated (e.g. Khandker, Saman and Khan, 1998). However, in some cases, microloans are criticized for their possibility to raise the overall debt of the poor, especially when loans are pushed or the possibilities of pay-back are not carefully analysed. In other words, microcredit can make the poor even poorer by giving them a loan they cannot repay. This may force the borrower to sell his few assets in order to maintain repayments, reducing long-term income level. If the borrower has no assets to sell, the debt burden will simply increase. This holds true especially for the poorest among the poor. A statistical study by Ahmed (1997) states that the poor who take in microloans, can fall victim to perpetual debt. Instead of the helping hand that the loan was supposed to be, it becomes an additional burden. Marr (2004) observes that field officers do not distinguish between the inability and unwillingness of borrowers to repay. In such cases, borrowers who are unable to repay because they invested their loans in long-term projects (e.g. agriculture or education) or suffered unexpected shocks
10
(e.g. illness or theft) are classified by field staff as intentional defaulters. Due to this bias, MFI officers advise borrowers to invest their microloans in short-term trading activities in order to repay loans promptly. When borrowers cannot repay, officers advise them to borrow from any source in order to repay loans on time. Marr found that an astonishing 62 percent of the participants of her sample became over-indebted from other sources in order to repay MFI loans. Thus, microcredit can push clients to borrow from their original moneylenders. In such cases, clients fall into a vicious circle making them worse off with an even higher amount of debt. Marconi and Mosley (2004) write that, according to their impact assessment, microfinance debt is increasing the vulnerability of the poorest in Bolivia since its recession in 1999. Impacts of microfinance on income, investment and employment were negative for poorer clients, though positive for clients with sales in excess of 500 dollars. Such results indicate that many of the poor risk falling deeper into the poverty cycle by taking microfinance credit, especially during a recession.
Repeat loans
The feature of receiving new additional loans when a previous loan is repaid is quite a common attribute to microcredit programmes. Bond and Rai (2004), among others, show that the promise of future credit, together with the threat of credit denial in case of default, is one of the most important features in increasing repayment rates. However, other studies note that repeat loans become less interesting to graduated borrowers (i.e. loans taken for a second time or more) in time, because the available loans do not meet their needs and demands, impeding their possibility to grow. When graduated borrowers have successfully lifted themselves above the poverty line, the average loans still available to them and the types of activities that they initially were involved in are less appealing. Wahid (1994) and Khandker (2005), for instance, show that the loan recovery rate with graduated borrowers is relatively lower than with new borrowers. Mosley and Hulme (1998) studied 13 MFIs and concluded that for each MFI the impact of lending on the recipient’s household income tended to increase, but at a decreasing rate. If poverty is to be alleviated at a continuing pace, loans must be adapted to client’s requirements. MFIs must keep close eye on this development as it could have serious implications for repayment ratios.
Savings as collateral
Compulsory savings11 is another form of collateral besides group contracts imposed by MFIs. These funds are contributed by borrowers as a condition of receiving a loan, either as a percentage of the loan or as a nominal amount. They are considered part of a loan product rather than an actual savings product, since they are narrowly tied to receiving and repaying loans. They are frequently used and have led to positive effects since they provide an additional guarantee mechanism for repayment. They also show the ability of clients to manage cash flow and make periodic contributions. However, this
11
form of collateral leads to high dissatisfaction among borrowers. Borrowers often perceive these savings (rightly) as a cost they must pay to take part and gain access to credit. Furthermore, the deposit interest rate paid (if any) is usually lower than the return earned by the borrowers if the savings were put into their business or other investments (Ledgerwood, 1999). The scheme adds substantial, hidden costs to borrowing and the opportunity costs of these funds are therefore quite high. According to Murray and Lynch (2003), the frustration with the lack of interest on compulsory savings was the highest of all loan attributes and requirements surveyed for customers of Uganda’s Women Finance Trust. The study pointed towards another aggravation of compulsory saving as well: clients often have long waiting periods before gaining access to these funds, if they get any access at all before having reimbursed their loans. When funds are only available after full repayment of the loan, this can result in clients borrowing amounts that are in fact less than their accumulated savings. Another scenario occurs when compulsory savings are set too high and clients must borrow from their traditional moneylenders to pay these mandatory savings. All these factors encourage clients to leave and result in MFIs facing high drop out rates. In such cases, compulsory savings do not lead to poverty alleviation.
Replicating microcredit programmes
Features such as repeat loans and compulsory savings have, despite their demerits, facilitated the success of microfinance and helped trigger its implementation all over the world. Though the economic and social environments of these countries differ tremendously, microcredit programmes are replicated in one form or another in many countries. The performance of most such programmes, however, has not been encouraging. As Morduch (1999, p4) points out “the leading programmes came about by trial and error. Once the mechanisms worked reasonably well, standardization and replication became top priorities. Innovation continued only around the edges. As a result, most programmes are not optimally designed or necessarily offer the most desirable financial products.” The vast differences in performance of programmes with similar design features have puzzled practitioners worldwide. It is clear, however, that there is no single microfinance format that will satisfy the varying needs of all different segments of the poor. Not all societies have a similar structure, so microcredit programmes that have not assessed the characteristics and needs of their clients can result in nonviable operations (Ciravegna, 2004). Ledgerwood (1999, p21) summarises the problems encountered with replicating successful programmes by stating that they are due to differences in social environments and lack of local alterations. Replication of microcredit programmes can lead to negative impacts as opposed to the purpose of alleviating poverty without careful feasibility studies.
3.3.2 Group Lending
asymmetries between borrowers and lenders. Godquin (2004) and Ghatak (1999), for instance, comment that joint liability, through peer selection, peer monitoring and peer pressure, is to be related with better repayment rates and mitigates adverse selection. Furthermore, group homogeneity and social interactions and sanctions are also expected to boost repayment performances through greater efficiency due to group dynamics. Frequent repayment schedules and moral hazard combined with peer monitoring are additional important features that induce repayments (Armendáriz de Aghion and Morduch, 2000; Besley and Coate, 1995; Van Tassel, 1999). The current literature contains many studies explaining the success of group models, however, there are several critical features that have not received as much attention as, perhaps, deserved.
ROSCAs
The idea of group lending has not come out of thin air. It originates from rotating savings and credit associations (ROSCAs) which have a long history in developing countries, even predating monetization (Hoff, Braverman and Stiglitz, 1993). They have features similar to group lending and are based on pooling resources with a broad group of friends and acquaintances. ROSCAs can be found worldwide.12 Structures are quite diverse, but the basic element is a group of individuals who agree to regularly contribute money to a common pot, which is allocated to one member of the group each period.13 At certain intervals (week, fortnight, or month) the group meets to collect dues and allocate the proceeds, with past recipients excluded from getting the pot again, until every member has had a turn. ROSCAs thus successfully take the bits of surplus funds that come into households and translate them into a large chunk that can be used to fund a major purchase. Due to its successful history, some microfinance lending groups, for example BancoSol in Bolivia, are reconstituted of older groups that previously existed as ROSCAs (Armendáriz de Aghion and Morduch, 2005; Ghatak and Guinnane, 1999).
Ex-Ante Moral Hazard
Armendáriz de Aghion and Morduch (2005, p43) define moral hazard in lending as “situations where lenders cannot observe the effort made or action taken by the borrower, or the realization of project returns.” Moral hazard problems have a double dimension, separated by ante moral hazard and ex-post moral hazard. Ex-ante moral hazard occurs before the production returns of the agent are realized. This occurs when the financial contract does not provide the right incentives to induce borrowers to undertake the actions or level of effort needed that generate the capacity to repay the loan. Various studies (e.g. Ghatak and Guinnane, 1999; van Tassel, 1999) note that group-lending contracts can
12
ROSCAs go by different names in different regions and countries. Examples include the tontines of rural Cameroon, the arisan in Indonesia, the susu in the Caribbean and juntas or panderas in Latin America (Khawari, 2004).
13
positively circumvent moral hazard by inducing borrowers to monitor each others’ choice of projects, their amount of effort and their output.
Despite the positive results attributed to peer monitoring in group models, one critique however, concerns the assumption that members have accurate information about one another or that they can access this information without costs (e.g. Ghatak, 1999), and regards it as excessively simplistic. Marr (2004) points out that there is little knowledge about each other’s household management and exposure to risk. In other instances, officers form groups composed of people who live in highly dispersed geographical areas and who possess very little knowledge about one another. In Marr’s sample in Peru half the participants had undertaken no peer monitoring at all and none undertook this task to an extent larger than fifty percent.14 This could be attributed to the high costs of monitoring, such as time and monetary expenditures. Various studies regarding peer monitoring do not take these costs into account, which limits them from addressing crucial aspects of optimal peer group design. Armendáriz de Aghion (1999) introduces a framework where peer monitoring costs are explicitly taken into account and finds that individual contracts dominate over group-lending contracts if peer monitoring is exceedingly costly. Typically then, monitoring will be imperfect, opening the way for moral hazard to enter back into the picture.
A further problem arises when joint liability models backfire through exerting an excessive pressure to choose overly safe projects that are not optimal in terms of average profitability. For instance, the potential loss of the group is the cost of a project failure and the cost of future exclusion from borrowing. The latter costs necessarily exist because the no-lending threat must be used to avoid voluntary or strategic default. As a result of this threat, a group may become over-conservative in its choice of projects, preferring safer, less profitable projects (Banerjee, Besley and Guinnane, 1992; Ray, 1998). Madajewicz (2003) discovers that the wealthier among the poor borrowers choose less productive investments when they obtain group loans than when they obtain individual loans. Clients with growing businesses or those who get well ahead of their peers in scale may find that the group contract bogs them down. In some programmes, borrowers obtain smaller loans when access is restricted to group loans than if the loans were individual ones, which can limit the size and profitability of successful businesses (Khawari, 2004; Madajewicz, 2003; Ray, 1998). It is because of these restrictions that the most celebrated of group lending MFIs like the Grameen Bank and BancoSol have started individual lending as well.
A prominent feature of group loans, very often present in group programmes, is group meetings. Though primarily used for financial purposes, they are important social and educational events for
14
clients, where they can share ideas, learn new skills and knowledge or make friends (e.g. Armendáriz de Aghion and Morduch, 2005; Hassan, 2002; Murray and Lynch, 2003). However, attending group meetings can also be costly, particularly where people live far apart. Clients have to travel, leave their households and close their businesses for a while. Armendáriz de Aghion and Morduch (2000, p4) give an example, where “in 2 of the 3 Chinese programmes (…) 8 percent of clients had to walk over an hour to attend meetings. In general, meetings and travel time took, on average, just over one hundred minutes.15 A survey of dropouts from group lending programmes in Uganda and Bangladesh stressed that group meetings come at a very high cost for low-income entrepreneurs. It was found that of all dropouts 28 percent in Bangladesh and 11 percent in Uganda left in part because of the frequency of meetings (Murray and Lynch, 2003).
Ex-Post Moral Hazard
Ex-post moral hazard occurs after the borrower’s production returns are realized. In this case, there is a risk that the borrower strategically defaults because the financial contract fails to provide sufficient incentives to repay. Fellow members may inflict social or economic penalties upon borrowers who have excessively risky projects or put in too little effort, thereby burdening the group with unnecessary risk. One important implication of group lending is that all members in the group face consequences if one member runs into serious repayment difficulties. In such a case, either the other members must pay for the defaulter or all group members are cut off from future borrowing. These sanctions have positive effects and studies (e.g. Ahlin and Townsend, 2003; Besley and Coate, 1995) show that they can induce higher repayment rates.
Sanctions, however, can also have possible negative effects on the performance of the group. Paxton (1996, p3) introduces the “domino effect”: a negative externality that arises when members of a group contract default due to the default of one or more other members. This is because one or more members’ authentic difficulties have destroyed the group’s credit rating. Under individual lending, however, some members would have repaid their loan (Ray, 1998). Godquin (2004) states that Bratton16 had found that group models have better results than individual loans in years of good harvest and worse in drought years when co-members are presumed to default. As potential default risks surge, members become more distrustful of colleagues’ behaviour, cooperative instincts vanish, and the cohesiveness of the group rapidly disintegrates. Default risks may exacerbate to such high levels that the whole group disintegrates and members end up in poorer circumstances than they were in previously (Besley and Coate, 1995; Marr, 2004).
15
The data are from a preliminary analysis of a survey of three programmes completed by Albert Park and Changqing Ren.
16
Another observation is that a borrower who is disposed to strategically default will wait until loan sizes have grown substantially before choosing to renege on the loan contract. Repayment incentives also erode if borrowers can scale up their own resources faster than the bank. Morduch (1999) and Ghatak and Guinnane (1999) find that this incentive increases when repeat loans have a clear end, giving rise to a tendency for borrowers to renege in the final period. Only with significant uncertainty about the end date or with a clear move from one programme to the next (“ad infinitum”), are group incentives successful. A related difficulty arises when a lender is so weak that borrowers suspect it might not exist in the future to provide loans. If borrowers start to believe that the bank could go bankrupt, it is more likely that they default now, since it is not clear whether there will be a future stream of loans (Bond and Rai 2002; Westley and Shaffer, 1996). Armendáriz de Aghion and Morduch (2005, p124) state that “such speculation can trigger a ‘debtor run’ that becomes a self-fulfilling prophecy.” Besley and Coate (1995) show another implication where borrowers collude together against the bank and the whole group attempts to strategically default.
Various studies show that group repayment performances improve where social ties and trust between members are strong (e.g. Floro and Yotopolous, 1991). Groups that trust more and cheat (or default) less will do better, which implies that groups who did not know each other should do worse than the self-selection groups with pre-existing social ties (Armendáriz de Aghion and Morduch, 2005). However, in recent years, this expectation has been put to the test by empirical studies, where results show that social ties may not have positive, but rather insignificant or even negative effects (e.g. Ahlin and Townsend, 2003; Abbink, Irlenbusch and Renner, 2002). Wydick (1999) finds in his research among borrowing groups in Guatemala that participants are sometimes softer on their friends, worsening repayment rates. Hence, in the Guatemalan context, group lending does not appear to function through utilizing previously existing social ties to mitigate moral hazard problems. Godquin (2004) points out that social ties inside a group, proxied by the age of the group, had a significant and negative impact on repayment rates. She explains that as the duration of membership increases, the demands for credit of members evolve differently resulting in tensions inside the group. A second explanation supports Wydick’s theory where joint liability does not work if individuals are unwilling to put pressure on and sanction delinquent borrowers. In various cultures, such as in regions of Malaysia, it is unheard of for people to mind each other’s business or to punish persons for their actions (Hassan, 2002).
may react very strongly by resorting to acts of intimidation, threats and even physical abuse in order to suppress information about their financial affairs (Montgomery et al., 1996). In such cases, joint liability has detrimental effects.
Montgomery17 observed, as mentioned in Armendáriz de Aghion and Morduch (2005), that the larger, village-level group plays a key role in ensuring repayment discipline. Instead of the idea that only the group members are jointly liable for the outstanding loan, field officers warn all members in the village that they will withdraw all loans. Moreover, field staff knows precisely which borrowers are finishing their current loans and are about to apply for a new one. These customers are pressured to help the problem customers, or else the anticipated new loans will be deferred. Hence, it is vital to acknowledge and deal with the complexity of social relationships among joint liability group members and between them and officers. Moreover, according to Morduch (1999), the dynamic incentives tend to function much better in areas where mobility is low. The rationale behind this is that it is difficult to monitor and catch defaulters who can travel across town and establish a credit line with a different agent or programme elsewhere. Since mobility in urban areas is comparatively high, dynamic incentives are more ideal for rural areas. The MFI Bank Rakyat Indonesia, for instance, had more problems securing repayments in urban programmes than in rural ones, which may be due to greater urban mobility (Morduch, 1999). Evidence is needed to demonstrate whether joint-liability-based mechanisms can work in more urban environments where even though social enforcement mechanisms are weaker, people might still have information about their peers (Ciravegna, 2005; Ghatak, 1999).
In general, Armendáriz de Aghion and Morduch (2000, p17) state the problem of group lending clearly by writing that “group lending is just one part of a set of related mechanisms employed to aid loan repayment rates. Moreover, it is not obvious that it is joint liability in group lending that drives the results, and not other aspects like public repayments, facilitation of education, and participation alongside neighbours.” It must be realized that the empirical data on group lending lags behind theory and that the data so far suggests important challenges to the generally optimistic tenor of the theoretical research on group contracts.
17
SECTION 4. GENDER AND SECTOR DEVELOPMENTS
Apart from the main idea of microfinance alleviating the poor, the movement has led to further implications. One such development is the specific targeting of women in order to attain women empowerment. Moreover, regarding the sector itself, there have been developments in regulation, subsidization and an increase in the number and variety of actors. Though all implications have had certain positive effects, each of them has their demerits.
4.1 Gender Concept
To many, microfinance is all about women. The leading programmes of Grameen Bank and BancoSol have a clear focus on women and renowned microfinance networks such as the Women’s World Banking or NGOs such as Pro Mujer reinforce this association. All organisations, with hardly any exceptions, declare that women are more reliable and have better pay-back ratios than men (Bosch, 2004). According to the Microcredit Summit 2000 nearly 80 percent of the microfinance borrowers are women. Nowhere is this percentage more striking than in Bangladesh where the Grameen bank currently serves over 6 million borrowers, of which the vast majority, 96 percent, is women. Professor Yunus, director of the Grameen bank, claims that he prefers female borrowers, because, due to women’s loyalties to their families, they have a more long-term vision, more self-sacrifice and attach more importance to the rest of the family than men do.18 These loyalties also make women less mobile, which makes it easier to organize the frequent group meetings (Bosch, 2004). MFIs explain their special interest in women entrepreneurs in that they often make up the poorest segments of society, have fewer economic opportunities and are generally responsible for child-rearing, including education and health. Given their vulnerable position, many MFIs seek to empower women by increasing their economic position in society (UNDP, 1999). These rationales explain the great amount of microfinance programmes that specifically targets women borrowers.
Section 2.2 briefly mentioned the positive social impacts among the poor population when targeting women. Yet, this was only half of the story. There seems to be no unambiguous evidence that targeting women constantly leads to an increase in women empowerment. Many evaluations of microfinance programmes have assumed that high take-up and repayment levels indicate positive impact on women and have not investigated further. Some only have anecdotal information from participatory consultations and gender workshops. Mayoux (1997) comments that studies of empowerment come to different conclusions, even about the same organization, because they vary in their definitions of empowerment and investigation methods.19 Many programmes have had positive as
18
Speech from Professor Yunus at the Microfinance Symposium Vlissingen, 12 May 2006.
19
well as negative impacts on women. Within schemes, impacts often vary significantly between women in different productive activities and from different backgrounds. Only catering to female clients in order to solve gender issues is not sufficient. Where culturally permitted, the ultimate goal should take into account the needs and constraints of both women and men when designing and delivering finance (Boros, Murray and Sisto, 2002). Empowerment of women is not only created by microfinance programmes, but depends on social and cultural factors as well (e.g. Forsythe, Korzeniewicz and Durrant, 2000; Malhotra and Mather, 1997). Despite the positive results of studies regarding women empowerment and microfinance, this sub-section will focus on the demerits of targeting women.
Loan control
A current debate focuses on whether women who receive a microloan are obliged to hand it over to their husband or another male in the household. A renowned study investigating this issue of loan pass-through was carried out by Goetz and Gupta (1996).20 They found that, although 95 percent of borrowers were female, in only 37 percent of cases did female borrowers retain significant control over loan use by these criteria. Hence, the majority of women borrowers in the programmes studied did not control either the loans received or the income generated from their microenterprises. While this is not necessarily negative when viewing the household as an economic unit, it can be an extra burden for women if they are held responsible for repayment without any say in the management or use of the funds. Goetz and Gupta also noted that men’s appropriation of loans is common and increases with loan size. The high degree of male control of loans can postpone the appearance of the positive social externalities, expected from increasing women’s control over household income. Even worse, it can undermine household survival strategies where men invest loans badly, forcing women to mobilize repayment funds from resources which would otherwise be used for consumption or savings purposes. In this scenario, women have responsibility but without control (Inter-Parliamentary Union, 199821; Mayoux, 1997) 22. Creevey, Ndour and Thiam’s study in Guinea (1995) and Holvoet’s research (2005) in South India find that none of their participants perceive any change in the decision-making in the household due to credit delivery. Women may have more say in matters directly related
20
They conducted qualitative studies of 275 loans (22 of these to men) across four organizations: BRAC’s Rural Development Programme, Grameen Bank, TMSS (Women’s NGO: Thangemara Mahila Sebuj Sengstha) and RD-12 (Governments Rural Poor Programme). They used a five-point index of "managerial control" over loans as their indicator of empowerment, selecting mostly female borrowers as their participants. At one end of their index were women who were described as having "no control" over their loans: these were women who either had no knowledge of how their loans were used or else had not participated in the activities funded by the loan. At the other end were those who were considered to have exercised "full" control, having participated in all stages of the activity funded by the loan, including the marketing of the product.
21
The article (1998, p6) mentions several studies that come to similar conclusions:
“a study of 151 Grameen Bank loans to women found that 12 percent surrendered the entire loan to male family members;
another study in Bangladesh discovered that of 140 loans made by ACTIONAID to women, about 50 percent were used for men's productive activities;
an assessment of K-REP confirmed that men try to control income from women's enterprises;
according to a study of women borrowers in the Grameen Bank, 10 of 40 women in the sample were passing on all or most of their loans to male family members under circumstances that gave them little control over the use of this capital.”
22
to the loan use, but they are not able to translate this into increased involvement in domains of household decision-making. Goetz and Gupta (1996) observed another scenario where men were unwilling to repay the loans of their spouses, leading to an intensification of tensions within the household, spilling over into violence. In addition, violence against women was also exacerbated by the frustration of husbands when the wife delayed or failed to access credit. Rahman (1999) too found that 70 percent of Grameen borrowers in his survey acknowledged escalated violence in the household as a result of women’s involvement with microfinance. He explained that microfinance may aggravate tensions, because men feel increasingly threatened in their role as primary income earners in traditional societies. A related problem, demonstrated by Hunt and Kasynathan (2001), is that NGO staff underestimates the degree of control that male family members have over loans provided to women. Few mechanisms are in place to monitor who controls loans, or their impact on wider social aspects such as violence against women, divorce and polygamy. Moreover, they found, that women are more likely to control loans if their husbands are absent, or if the money is used for a ‘traditional’ female activity. Regarding male support, in some cases, especially in Africa, women’s increased independence is temporary and leads to withdrawal of the husband’s support. Anxiety is expressed that small increases in women’s income can lead to a decrease in male contribution to certain types of household expenditure (Cheston and Kuhn, 2002; Mayoux, 1997 and 1999). Moreover, Mayoux (1999) discovers in her survey in Africa that women empowerment is very region and household specific and relies on inflexible social norms and traditions. Loans to women are thus not always a step forward in the alleged road to women empowerment. Rankin (2002) even concludes that microcredit may entrench, not challenge, the gender division of labour and power.
Work activity
A further demerit of microfinance to women is that the businesses created by microloans result in too excessive work loads, leading to ill health and exhaustion. The involvement of women in microcredit programmes can be negative due to a higher cost in terms of intensified demands on women’s labour (Cheston and Kuhn 2002; Goetz and Gupta, 1996). Mayoux (1997) too found that women enterprises often led to small increases in access to income at the cost of heavier work loads and repayment pressures. In addition, women often employ daughters and daughters-in-law as unpaid family employees thereby increasing their workload as well. Moreover, the need to have children work at home could decrease schooling levels, though other studies (e.g. Glenn, 2004) point out that microfinance leads to an increase of children’s education. Child labour may also imply an increased demand for children, increasing fertility rates (Morduch, 1999; Pitt et al. (1999).
sweater knitting that do not disrupt practices of isolation and seclusion within their households.” In their survey, Cheston and Kuhn (2002, p4) express concern that the “poverty-focused products tend to be the women’s products, with women keep being clustered in lower return businesses.” The gender gap increases when women keep investing in traditional activities that are not skill-intensive, while men become more skilful. The rationale behind the phenomenon is that unskilled women have very few income-generating opportunities they can invest in outside their households, besides the more traditional practices of which they have relatively more knowledge (Armendáriz de Aghion and Morduch, 2005). These developments hold back the women empowerment that microfinance is supposed to enforce.
Financial intermediaries
In order to have impact on women’s lives, MFIs should closely monitor usage of the credit and keep in mind the local social and cultural gender issues. There is evidence that MFIs specifically target women solely because of their high repayment rates. It is often perceived that the small credit amounts used in microfinance seem to suit women better than men and because in some countries women are less mobile than men, they have fewer tendencies to take the money and run. Yet, when it turns out that the loans are actually controlled by men, it may be that organizations use women as financial intermediaries to show donors that the schemes are efficient and sustainable. Focusing on ensuring high repayment rates, agencies lose sight why individual clients encounter problems with repayments, constraints brought about by gender, culture, tradition and the law (Pankhurst, 2002). Holvoet (2005) stresses the importance of recognising the difference between microfinance programmes that use women’s groups as financial intermediaries only and those that consider groups to be agents of social change. Female participation rates in microfinance programmes should thus not be assumed as an indicator of female empowerment (Hulme and Mosley, 1996, vol. 1). In a context of increasing donor funds being channelled into microfinance programmes, it is important to put into perspective the assumption that any microfinance programme empowers women.
4.2 The Microfinance Industry Concept
The increasing attention in recent years on microfinance has had repercussions on the industry. The number of suppliers has grown sharply, resulting in increased competition and supply in the microfinance market. These factors in turn have influenced repayment behaviours. Moreover, subsidies to MFIs have escalated, which has led to the current discussion whether MFIs should become self-sustainable or whether it is more important that they reach a large number of poor people. There has been a growing debate about the necessity of institutional oversight of formal and semiformal providers.23 This is mainly due to the magnitude of the market, the complexity of the
23
