Impact Assessment of Microfinance For SIDBI Foundation for Micro Credit (SFMC)

Impact Assessment of Microfinance For SIDBI Foundation for Micro Credit (SFMC) Phase 1 Report July 2001 March 2002 By Putting people first EDA Rural Systems Pvt Ltd 107 Qutab Plaza, DLF Qutab Enclave-1, Gurgaon 122 002, India Tel: 6350835, 6356692, Fax: +91 0124, 6352489 E-mail: edarural@nda.vsnl.net.in March 2002 1

Selection Of MFIs The 5 MFIs were selected as representative of the 31 MFIs in SFMC s list of current partners representing different regions, models of microfinance (Self Help Group -SHG, Grameen, Individual Banking and sector/enterprise specific cooperatives), age, and outreach to members and range of services. At each MFI, 2-4 sample areas (villages or urban wards) were selected purposefully to represent a typical area of the MFI in terms of socio-economic context and range of MFI programmes. Within each sample area, a stratified random sample of clients, non-clients and dropouts was drawn using wealth ranking as a basis for stratification. A series of 4-5 FGDs on these different topics was carried out in 9 sample villages and 4 urban wards, involving around 400 participants from the 5 MFIs. Quantitative survey techniques based on a pre-set questionnaire were used to interview a sample of clients, dropouts and non-members. Just over 1,400 questionnaires were completed (250-280/MFI) covering: 1,025 clients, 120 dropouts and 290 non-members. Micro finance models Three main microfinance models have been identified as a parameter of analysis for this study. Within MFIs, however, we have found that the approach is not always clear-cut and some MFIs combine the different models or modify them. Therefore, whilst retaining microfinance model as a parameter of analysis it will be important to differentiate between MFIs and between different kinds of client within an MFI. MFIs The main sampling parameters for this study were: 1 MFI model: Model 1 = SHG, Model 2 = Grameen, Model 3 = Individual banking (IND)/sector cooperatives (SC). 2 Region: R1 South & West, R2 North & East. MFI MODEL REGION SHG GRAMEEN IND-S R1: SOUTH& WEST R1/SHG R1/GRAMEEN R1/IND-S R2: NORTH & EAST R2/SHG R2/GRAMEEN R2/IND-S 2

MFI sample for Baseline 1 MFI MODEL REGION SHG GRAMEEN IND-S R1: SOUTH& WEST SHARE R2: NORTH & EAST ASSEFA RASS IASC LUPIN VWS RGVN SIFFS Sample summary Reports MFIs Clusters Survey Total Clients Dropouts Non-clients Baseline 1 8 ~24 2,250 1,600 115 535 Baseline 2 8 (-10) ~30 2,800 2,000 134 666 Total 16-18 ~64 5,050 3,600 250 1,200 Phase 1: Preliminary findings from the pilot test The sample for the pilot test covered 2 SHG model MFIs, 2 Grameen model MFIs and 1 sector specific organization. The two Grameen MFIs have both rural and urban members. The other three are rural focused. This section presents summary indicative findings relating to some of the research questions and analysis parameters. Further and detailed analysis will form part of the completion of the full Baseline Reports at the end of the next two phases. Coverage and outreach of microfinance Coverage of a microfinance programme is computed on the basis of number of client households compared to the total number of households in sample villages and urban slums. In rural areas, outreach of the sample is to 25% households in smaller villages (<500 households) and 13% in larger villages. 3

Micro finance coverage in sample villages Village size (no of households) No of villages in sample Total number of households Number of households With members* Coverage (%) =/>500 8 8,900 1,196 13 <500 9 2,066 524 25 Total 17 10,966 1,721 16 * In some villages, number of members per household is more than one Source: community listing, group records In urban slums, MFI member households were spread over a large area and the operational area of the MFI was not clearly demarcated. In the five slums of our sample, households with members were 100-250 in slum areas of up to 8,000 households. Outreach of the programme was analysed in terms of economic status of client households. For this, the community in sample villages/slums was divided into five economic categories on the basis of community wealth ranking Wealth ranking categories Non-poor Poor 1 Surplus 2 Self sufficient 3 Borderline 4 Poor 5 Very poor 4

MFI services reach all categories of poor and non-poor categories of households almost in proportion to their incidence in the community: 60% are poor, 40% are non-poor Data on economic status of members by period of time with the MFI is shown in Figure 3.2. It is not possible directly to compare long-term (more than 4 years) members with new (less than 1 year) members since the focus of MFI programmes has tended to change over the years. Nevertheless, it is interesting to see that there is (still) a high proportion of the poor (67%) among long-term clients, with 28% being in the lowest categories of the poor. This suggests little significant shift in economic status for long-term clients. Proportion of total 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Figure 3.2 Wealth ranking of clients by time with MFI 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Outreach to different economic groups MFI Clients [618] HH [1,824] Surplus <=1yr [433] 2-3yrs [190] 4+yrs [265] Self sufficient Borderline Poor Very poor Source: Wealth ranking- 4 MFIs, 12 villages & 1 slum Surplus Self sufficient Borderline Poor Very poor Source: Client wealth ranking- 4 MFIs, 14 villages, 2 slums Use of MFI and group funds 1 MFI clients access loans through two distinct sources: either from the MFI (usually a larger amount, on terms decided by the MFI) or from group funds (these are member savings which are rotated within the group usually in smaller amounts on terms decided by the group). Table 3.2 compares loan use among sample clients of two rural MFIs one from South India and the other from North India. In the southern MFI, MFI funds are used primarily for milch 1 During the pre-test, we found that clients distinguish between loans from group savings and external loans from the MFI 5

animals and agriculture, with limited scope for non-farm enterprise. Group funds, which the SHG members decide amongst themselves, are used mainly to meet household needs. In the northern MFI, in comparison, field staff guides the SHGs on loan use for group loans as for MFI loans. There is also an individual borrower programme (for men) with a focus on nonfarm enterprise. Here both MFI and group credit is used mainly for investment, with a significant proportion for non-farm enterprise Table 3.2 Loan use by MFI clients (% of total loan amount from MFI and SHG) past 2 years Purposes From the: MFI 1: South India MFI 2: North India MFI Group MFI Group Agriculture 20% 17% 1% 6% Animal Husbandry 39% 11% 50% 30% Non-farm enterprise 13% 7% 44% 37% Total investment 72% 35% 95% 73% Housing 9% 5% - - Education 2% 11% - 1% Household assets, lifecycle needs 16% 47% 5% 18% Other (debt redemption) - 8% Total household needs 28% 65% 5% 27% Total loan amount (Rs) 7,58,300 3,06,400 Source: Survey data 2 MFIs, 10 villages, 316 client households MFI/group credit and other credit sources 7,90,100 87,500 a) People s credit needs and sources of credit available are wide ranging (Table 3.3). This is applicable across all areas where MFIs operate and across all economic categories. b) Generally, it is the moneylender who meets the maximum number of types of need, but there are other important sources too. Figure 3.3 compares two situations in the south and north. In the northern example, where this MFI has a general loan product, which can be used for any purpose, the MFI meets as many types of needs as the moneylender. A rural/urban comparison suggests that MFIs may meet a larger number of credit needs in urban areas. The local bank is more evident in the southern sample than in the north. Table 3.3 Credit needs and credit sources Credit needs Investment Household Agriculture- land, inputs, tractor Lifecycle marriage, birth, death, festivals Animal husbandry- dairying, small animals HH needs food, daily needs, emergency Education, housing Fishing motor boat, inputs, nets Emergency delivery, illness 6

Trading, services, and manufacturing Other - debt redemption, bribe, travel Credit sources Formal Semiformal Informal Banks with and without subsidy MFI & MFI linked groups High cost money lenders, finance companies, pawn shops Primary Agri Coop Societies Other NGO/bank groups Medium cost ROSCAs, ASCAs, bishis, chit funds Low cost friends and relatives, store keepers, suppliers, employers Source: Focus Group Discussions - 5 MFIs, 20 FGDs, 155 clients and non-clients Figure 3.3 Which source meets most credit needs? No. of needs 10 9 8 7 6 5 4 3 2 1 0 South India - Area of an SHG model MFI (Rural) Shop MFI Finance co. Bank Goldsmith Neighbours ROSCA Moneylender No. of needs 10 9 8 7 6 5 4 3 2 1 0 North India - Area of a Grameen model MFI Bank Goldsmith MFI Moneylender Neighbours Goldsmith Neighbours MFI Moneylender Rural Source: FGDs 2 MFIs, 6 villages & 2 urban slums Urban The majority (85%) of MFI clients surveyed report at least some borrowing from any source in the past two years; 15% report no borrowings from any source. The proportion of borrowers was lower in the case of the very poor (81%) and surplus category (70%). Data on average borrowings from any source over the past two years is presented in Figure 3.4. This data represents mainly short-term (</=1 year) members and mediumterm (2-3 year) members. It therefore provides a baseline for comparison with average 7

borrowings at the time of the follow-up survey. However, for one MFI, even long-term members show a similar pattern of borrowings. Figure 3.4(a) Rupees '000 45 40 35 30 25 20 15 10 5 0 42,678 Average borrowings past 2 years Group 28,688 MFI 20,581 18,643 16,861 13,469 Surplus (37) Self sufficient (119) Borderline (114) Informal Formal Poor (82) Very poor (68) All Categories (420) Source: Sample survey 3 MFIs, 420 borrower households Note: the arrow indicates debt outstanding at the time of the survey Figure 3.4(b) Distribution of borrowings by lending source 80% 60% 40% 20% 0% Surplus (37) Self sufficient (119) Borderline (114) Poor (82) Very poor (68) All Categories (420) Informal Formal Group MFI Source: Sample survey 3 MFIs, 420 borrower households The figures show: a) Increasing levels of borrowings and indebtedness (in absolute terms) the higher the economic category b) In relative terms, MFI borrowing as a proportion of total borrowing of client households averages 33% for poor clients and 23% for non-poor clients 8

In terms of absolute value, clients from poor households borrowed on average Rs 4,787 from the MFI; clients from non-poor households borrowed on average Rs 7,623 c) Informal sources account for 56% of average borrowings. This proportion is highest (64%) for very poor clients and lowest (33%) for surplus category clients 2 Focus group discussions and an interview with a local moneylenders provides insights into The effects of two (northern) MFI programmes on local financial services: o A local moneylender has reduced his interest rate from 10% to 8% a month, and now offers clients more flexible terms (Grameen MFI, urban slum, north India) o In a village of a (women s) SHG MFI in north India, 5 informal chit fund groups of men have been formed, following the example of the women. These groups have features of ROSCAs and SHGs. In the same village, we were told that moneylenders no longer charge exceptionally high interest rates for emergency loans o In the same village, a discussion with a local moneylender revealed an interesting example of arbitrage: he is a large landowner who borrows a substantial sum from the local bank under the Kisan Credit Scheme against his landholding - and onlends this to his informal clients. (His daughter in law is also an SHG member). Women s empowerment This aspect was covered through focus group discussions, which explored women s own perceptions of empowerment and change resulting from their involvement with the MFI programme. Effects of microfinance programme on women vary with MFI s approach to gender issues. For example, women s groups of the SHG model MFI in the South were found to be much more aware and active about the programme than the Grameen model MFI in the North. The focus in the Southern MFI is on group processes that allow women members to take final decisions at least for the group money giving a sense of ownership, and therefore responsibility. On the other hand, in the North, the focus is on group leaders and there is a high degree of involvement of MFI staff even in decisions related to the group s money. This provides a better opportunity to the MFI to monitor the programme closely, and has resulted in a high degree of awareness and sense of responsibility among the leaders. But the approach has left ordinary members ignorant of their roles and responsibilities. In terms of overall effects of microfinance programme on women, a summary of observations made through FGDs in four MFIs is presented in Box 3.3. 2 9

Box 3.3: Effects of microfinance programme on women Positive Opportunity for savings - in own name Increased mobility within and outside village (group leaders especially) Sense of pride and self-respect: they feel confident can now talk to outsiders Group interaction leads to opportunities for exchange of market information Lobbying for infrastructure (municipal areas) Action against alcohol outlets Negative Group leaders have the main opportunities and can be too much in control There is pressure from the family to go for a loan SC women are sometimes excluded or if part of a group, are not fully informed or treated equitably In MFIs where men are also clients, products for women may be of smaller amounts and at higher interest rates (in one case this was for historical reasons related to the cooperative structure) Source: FGDs 9 villages; 89 women clients Survey data (of 84 dropouts) indicates a mix of reasons why people have left a programme, with no clear differential pattern for poor and non-poor. Reasons for leaving are, in descending order of response frequency: o No time o Cannot save o Did not get loan o Casual migration o Sceptical o The way the group/leaders behave Environmental issues Environmental issues came up in the case of two MFIs. Both are part of an overall trend in a sector, which MFI staff are conscious of and, in one case, are trying to address. In the case of an MFI providing a significant part of its loan portfolio for milch animals, we found that the practice of giving hormonal injections to increase milk yield is widely practiced. This may have safety implications for milk consumers and for the animals. In another MFI area, mechanization of fishing boats over the past 10-15 years has put pressure on the natural fishery resource resulting in reduced catches and the disappearance of some species. This is an issue of central concern to the MFI which has the policy of not providing credit for fishing technologies that it has identified as having the most serious effects, whilst providing credit for non-mechanised and certain types of motorized boats. Nevertheless, there are alternative sources of credit to fishermen, including governmentsponsored cooperatives, which promote and subsidise credit for mechanization and as a result 10

the number of motorized boats is continuing to increase beyond the capacity of the fishery resource. Enterprise support For a number of MFIs, purchase of milch animals constitutes a major proportion of their loan portfolio. This has led to follow up support for milk collection. A large MFI in south India has obtained funds to set up its own dairy programme with milk collection centres, chilling plants and milk marketing. Another MFI in north India has contacted private dairies to start collection centres in villages where it is operating. SFMC support For the 5 MFIs sampled, SFMC loan funds represent from 1 to of funds available to the MFI for lending, based on cumulative figures. These funds are mostly (but not always) earmarked for non-farm enterprise and have often contributed significantly to MFI initiative in exploring alternative approaches to channel credit for this purpose. Such alternative approaches usually include targeting men rather than women. However, in the case of one MFI, not only has SFMC enabled the MFI to strengthen its support to its traditional constituency (of boat owning fishermen) but has also led to expanded outreach and provision of services to new types of clientele: women in fishing communities and boat crew members. 11