In this section, we explain why MFIs are more informative than commercial banks, which play a key role in the proposed linkage. Then, we provide evidence that shows the impor-tance of MFIs in poverty reduction, while addressing briefly the failure of current policies, which impose direct controls on commercial banks loan decisions, in facilitating credit flow to the poor. Finally, we describe some evidence that supports our theoretical conclusions for the benefits of a linkage between commercial banks and MFIs.
MFIs are More Informative According to Morduch (1999a; 2000), who provided com-prehensive surveys on the topic of rural finance, microfinance often displays patterns and features not commonly found in institutional lending. First of all, since MFIs are mostly located close to rural areas, lenders possess a great deal of information about relevant bor-rower characteristics, such as farming ability, size and quality of landholdings, cropping patterns and risk attitudes. In addition to advantages in location, Sarap (1991) discov-ered in survey data that the guarantee system prevalent in informal credit markets with diverse socioeconomic conditions is very complex. And the moneylender, through a variety of guarantees including collateral and personal relationship, is in a position to screen and monitor the borrowers with negligible cost in such a way that there is hardly any risk of default.
Assessing 1,438 households in six provinces in Indonesia, the country’s largest micro-finance bank judged about 40 percent of poor households as creditworthy, and possessing collateral appeared as a minor determinant of creditworthiness (see Johnston and Morduch, 2008). However, the formal banks may not want to provide financial services to the poor, as the lending brings about problems such as collateral, location, information, and small loan size. Since the MFIs entered the scene more than a decade ago, they have always been local and have based their business on local relationships and were willing to serve the unbanked despite their desire for small size loans.5
In order to exploit the informational advantage that the MFIs have, there is a need to establish a linkage between formal banks and the rural borrower via the MFIs. For example, given the 95-100% loan repayment rate of more than 1 million Self-Help Groups (SHGs) in India, the NABARD has initiated a credit linkage programme6 to reduce transaction costs for both banks and borrowers and thus achieve the objective of providing credit
5Jain (1999) uses a model showing the informational advantage of lenders in the informal sector.
6See www.nabard.org/roles/microfinance/index.htm.
access to the poor.7 The MFIs charge interest rates that are much higher on average than bank interest rates but also show significant dispersion, presenting apparent discretion opportunities (see Ghosh, Mookherjee and Ray, 2001). As the MFIs have limited capital base, they usually charge high interest rate to cover their operational expenses. Currently, MFIs in general charge over 50% on their loans, but this high rate can be lowered for good borrowers as in the case of SHGs, if the MFIs can intermediate their access to the formal loan market. Recently, many MFI’s in India, and ICICI Bank, India’s second largest bank (and largest private bank), have entered into a mutually beneficial strategic partnership agreement to provide microfinance services to the poor, given the MFI’s market knowledge of poor customers and the ICICI bank’s vast financial resources (see Ananth, 2005). In Indonesia, Hamada (2010) has shown that bank loans through linkage programs to MFIs contribute more than bank loans alone, in terms of outreach to the poorest of the poor, in line with our theoretical results.
From the borrower’s perspective, due to lack of access to formal financial services, the poor have developed a wide variety of informal, community-based financial arrangements to meet their financial needs, for example, the Rotating Credit and Savings Association (ROSCA), which typically consists of a group of community members who meet regularly to pool their savings, which is then lent out to one member of the group, who repays it, at which time it is lent out to another group member, and so on until each group member takes a turn borrowing and repaying the pool of savings. Evidence suggests that personal trust between group members and social homogeneity are more important to group loan repayment than general societal trust or acquaintanceship between members (Cassar et al., 2007). It has been well documented that joint liability group lending can help reduce information asymmetries (see Hermes and Lensink, 2007), and outreach is lower in the case of lending to individuals than in the case of group lending (Mersland and Strom, 2009). Also
7A SHG is a homogeneous group of about 20 people, who can be financed by a bank without collateral, if the group has accumulated savings and has a credit history.
Ahlin and Townsend (2007) find evidence that repayment is affected negatively by the joint liability rate and social ties, and positively by the strength of local sanctions and correlated returns. These informal mechanisms (i.e., issuing guarantees or group borrowing) may not be very successful, unless they are brought into operating alongside the MFIs. Banks and financial institutions have been entering the microfinance market in increasing numbers in the recent years (see Figure 4). Therefore it is important to get MFIs to coordinate with self-help groups in order to establish a formal framework for a linkage with mainstream banks and make the bank-MFI linkage sustainable.
F ig u re 4 : S e lf-h e lp Grou p -ba nk L ink a g e P ro g ra m m e in In d ia d u rin g
MFIs Can Play an Effective Role in Poverty Reduction Most observers regard mi-crofinance interventions as poverty reducing, although most mimi-crofinance schemes continue to benefit from external subsidies (see Honohan, 2008). Burgess and Pande (2005) find that state-led bank branch expansion into rural un-banked locations in India significantly re-duced rural poverty, using aggregate data. Besides this state-led financial expansion, the micro-credit movement in the recent years also seems to have revolutionized the bank-ing system of many countries such as Bangladesh by movbank-ing a large segment of the rural population, from the informal to the formal market through access to institutional credit,
thereby establishing the creditworthiness of the poor (Robinson, 2001). New institutions like Bangladesh’s Grameen Bank and Bolivia’s BancoSol have shown that it is possible to secure high rates of repayment while lending to poor households. The key is a series of new mechanisms, most famously ‘group-lending’ with joint liability.8
The group lending or peer-monitored lending schemes, pioneered by the Grameen Bank, introduce joint liability which induces a group formation of low risk borrowers. Banerjee and Newman (1994) have illustrated the working of the peer monitoring effect. Follow-ing the loan disbursement, the incentive system is likely to lead to peer monitorFollow-ing, peer support and peer pressure between the borrowers, thus helping the lending institution to address the moral hazard and enforcement problems. de Aghion and Morduch (2005) have described mechanisms, namely direct monitoring, regular repayment schedules, and the use of non-refinancing threats, to generate high repayment rates from low-income borrowers without requiring collateral and without using group lending contracts that feature joint liability. Such alternative types of group contracts expanded during the 1980s and 1990s, primarily sponsored by the NGOs. Lending to the poor is expensive due to high screen-ing, monitoring and enforcement costs (Karlan, 2007). It is believed that Group lending helps overcome this by harnessing social connections via peer monitoring and enforcing joint-liability loans. Karlan (2007) observes direct evidence that individual relationships deteriorate following a default, and that through successful monitoring, individuals know who to punish and who not to punish after default. Although these social network tools can be successful to have higher repayment and higher savings rates, there are costs following the breakdown of a social network. Hence except for linkage with MFI, other tools can be costly.
Granting small loans to help poor people start businesses became a popular poverty-fighting tool, encouraging private-sector activity (see Hassan, 2002; Hulme and Mosley,
8See Besley and Coate (1995), Ghatak (1999) and Conning (1999). In terms of past evidence, Ghatak (1975) found a positive but a weak link between unorganised and organised credit markets in India.
1996). However, due to capital constraints, MFIs can only lend a small amount of money.
This will not suffice for many growth-oriented entrepreneurs to start any microenterprise.
Madajewicz (2010) argues that individual liability offers the wealthier among poor (credit-constrained) borrowers larger loans even without monitoring, and hence micro businesses funded with individual-based loans as opposed to joint liability contracts grow more. The strategic monitoring efforts of group members can differ in equilibrium due to the asymme-try between members in terms of future profits and due to free-riding problem (van Eijkel et al., 2009). Besides, another practical limitation of group lending in urban settings is that members are less likely to know each other well as to whether a member is safe or risky as one would know in a rural setting. Also under a group lending scheme, a default by one borrower can affect the credit rating of the group as a whole. This might help explain why an individual loan contract for a good borrower can be welfare-enhancing if the MFIs with better information about those individual borrowers can graduate them to the formal bank for a larger loan amount to help fund a small business. The success of many individual cases such as in Bangladesh suggests that, if the creditworthiness of the poor established in the process of MFI lending can be incorporated in the credit decisions by formal commercial banks, there will be more chance for growth-oriented entrepreneurs to get access to loans from commercial banks. This market-oriented approach, as shown by the evidence below, is more effective in poverty reduction than current policies which mainly impose controls on commercial banks’ credit decisions, helping achieve financial sustainability with social outreach. So an MFI not only plays the role of a financial intermediary but also can act as a social intermediary.
Table 1: Mistakes of the Government's Official Priority Sector Policies
Regional Rural Banks (Gramin Banks) in India APS LPS Profits (INR bln)
APS - Ratio of the priority sector loans to total assets LPS - Ratio of priority sector loans to total loans INR bln: billions in Indian Rubee
Failures of Current Interventionist Policies The microfinance industry has devel-oped over the past 30 years, but it is still far from reaching its full potential as the industry stands between increased commercialisation and increased donor aid. The NGOs in mi-crofinance not only face challenges in balancing outreach and financial sustainability, but there is growing evidence of their failure to make an overall impact on poverty reduction.
It has been well documented in the literature that, the subsidized credit programs of the last three decades have failed miserably in giving a helping hand to that segment of the population those who have little access to credit.9 Subsidies distort the market, creating a dependency on subsidies and ensuring that commercial players do not enter the market, effectively pushes the poorest further away from the point of becoming financially viable as entities in their own right. For example in India, in the 1960s and the 1970s policy intervention in the rural sector was rooted in agricultural finance in a manner that credit, often subsidized credit, was necessary to enable small farmers to adopt risky new crop technologies and also to push them over to commercial (as opposed to subsistence) agri-culture. This type of ‘directed’ and ‘subsidized’ credit administered through government
9See Morduch (1999b) for a discussion on failure of subsidized schemes. For different examples, also see www.microsave-africa.com.
owned-commercial banks did not seem to have fully met the financial needs in the agri-cultural sector.10 The loans provided for agricultural activities reflecting the government’s pro-rural policies are called the priority sector loans.11 Table 1 shows how reduction in priority sector loans has contributed to improvement in profits of Regional Rural Banks (RRBs), reflecting the mistakes of the government’s official priority sector policies. In the Figure, APS denotes the ratio of the priority sector loans to total assets, and LPS denotes the ratio of priority sector loans to total loans. The ratios are calculated from annual accounts data of RRBs, obtained from the website of India’s Central Bank.12
The microfinance revolution of the 1990s sparked a major debate between the poverty oriented13 lending, reflecting distributive role of credit policy and the financial systems approach promoting greater financial innovation. The importance of MFIs in poverty reduction is now well documented, but how can we do better? Direct subsidies or donations to MFIs in the form of grants are currently seen in many countries, but the massive aid has not delivered the expected results (Padmanabhan, 2001), as many micro enterprises as part of larger aid and development projects, have turned unviable, making the MFIs donor-dependent.14 Subsidization in many instances takes the form of cheap loans and when the funds for this dry up, the institution is not in a position to carry on. Also microcredit interest rates are high because microlending remains a high-cost operation. The key to
10Government loans for agriculture have existed in India since 1793, and short-term cooperative credit institutions have existed since 1904 and the banking sector expanded substantially from 1955 with re-gional rural banks (RRBs) being formed from 1975 onwards, increasing bank finance for rural households (Premchander, 2003).
11Credit allocation in favour of priority sectors such as agriculture has been the traditional instrument for monetary policy to play a distributive role; but it has the disadvantage of distorting the credit market.
12See www.rbi.org.in.
13See Johnson and Rogaly (1997), Buckley (1997) and Hollis and Sweetman (1998).
14See www.themix.org that gives details of 150 MFI’s all nearly viable, but 11,000 others may disap-pear when subsidies are eliminated. These people will have all the hang-ups of dealing with subsidized institutions and no institutions then will be willing to go near them.
reducing these rates in a sustainable manner is to reduce costs through improved market competition, innovation, and efficiency. On the other hand, commercial banks are also unwilling to serve the unbanked (low-income earners, micro-entrepreneurs and the poor) due to high costs involved for small loans.
Recently, there has been a shift in the focus of MFIs toward commercialization or profitability as a result of international donor pressure to achieve large-scale operations and financial sustainability. This is the very discernible direction for MFI as pushed by the CGAP, emphasizing on market reform, careful regulation and monitoring, and the development of various other products and services such as deposit taking — a move from traditional micro lending to financial services in the broader sense (see Drake and Rhyne, 2002). Besides, as mentioned earlier, there is some evidence that commercial banks are cautiously venturing into this market through some form of a linkage programme.
Following this line, our paper has developed a market-oriented approach of a linkage between banks and MFIs lending decisions, which can provide the necessary scale and outreach in order to overcome the sustainability challenge facing the MFIs and thus can help alleviate poverty. If the MFIs can graduate their creditworthy borrowers to the formal bank, these borrowers can get access to a bigger loan amount to expand their business.
The formal financial sector is predominantly urban-based, and tends to be out of reach of peasant farmers, small-scale entrepreneurs and ordinary households, so the microfinance sector, Bangladesh being a glaring example, fills the gap in the market. Since the formal sector for providing productive credit has found it less attractive to enter this sector because of fears over default risk, strengthening credit delivery mechanisms are important with special focus on the promotion of micro credit ventures in the credit delivery that can aid both borrower selection and project implementation. Unlike many other developing countries, in India the policy environment as well as the institutional structure required to serve the needs of the rural people already exists and seems to offer favorable conditions
for providing credit to the poor.15
Funding for MFIs comes primarily from governments and international development organisations including the World Bank and regional development banks, and MFI related agencies or donors. There is anecdotal evidence in the media reports that local NGOs sub-contracted by a donor-funded microfinance programme are taking bribes from borrowers.
This type of corrupt practices suggests that subsidized credit or donor-funded micro-credit expansion may not be the way forward for financial deepening and future development in the disadvantaged local economies. Directing financial services to micro entrepreneurs therefore requires sustainable rural financial systems ideally being market-based. In the case of India, apart from directed lending through priority sector advances, many commer-cial banks have come forward to support innovative microfinance schemes (Ghosh, 2000).
Thus, there is a need to balance microfinance-oriented market-strengthening policies with institutional initiatives to reduce dependence on subsidized or donor-funded microcredit.