Performance and Sustainability of Self-Help Groups in India : A Gender Perspective

Applied Economic Research (NCAER), New Delhi, India. This paper uses the primary data collected for a project on the impact and sustainability of the SHG Bank Linkage Programme carried out at NCAER during the period 2007–2008. The authors thank the Deutsche Gesellschaft für Technische Zusammenarbeit Rural Financial Institutions Programme for funding this project. The authors would also like to thank Palash Baruah for assistance with the data analysis and two anonymous referees for their useful comments and suggestions. All errors and omissions are the authors’ responsibility. The views are those of the authors and not of the institution to which they belong.


I. BACKGROUND
In the pursuit of economic development and planning, microfinance programs1 were engineered by a few well thinking planners2 to generate income and employment and alleviate poverty especially in the developing countries.The approach is accepted by the World Bank and other financial institutions as an important tool for poverty eradication and enhancement of living standards, particularly those of women.Moreover, microfinance has come to be regarded as a supplementary development tool that widens the financial service delivery system by linking a large rural population with formal financial institutions through self-help groups (SHGs). 3A number of impact analyses have already established that these microfinance programs contribute to the achievement of several of the Millennium Development Goals (MDGs). 4n India, the National Bank for Agriculture and Rural Development (NABARD), a pioneer in agriculture credit activities in the country, was the first to notice the potential of an SHG bank linkage program (SBLP).In 1996, NABARD launched a nationwide pilot project to link the SHGs to the banks.In 1998, NABARD partnered with the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) Rural Financial Institutions Programme-which aims to improve the access of rural populations to sustainable, quality financial services-in support of the SHG bank linkage program.Since this partnership began, microfinance has been steadily spreading through the SHGs in India.The total number of bank-linked SHGs in India grew from 4,757 in March 1996 to 1,374,917 in March 2007.A report by NABARD (2008) suggests that by March 2008, the SBLP covered more than 70 million poor households across India.About 5,009,994 SHGs were maintaining savings accounts with the banking sector and about 79.6 percent of these SHGs were all-female.India's southern region had the largest percentage of SHGs having bank savings accounts (48.0 percent), followed by the eastern region (21.4 percent), central region (12.9 percent), western region (9.4 percent), northern region (4.2 percent), and northeastern region (4.1 percent).Among the three types of banks involved in promoting and disbursing loans to SHGs in India, the commercial banks had the largest share of SHG savings accounts (56.1 percent), followed by regional rural banks with 27.7 percent and cooperative banks with 16.2 percent.The total number of 3,625,941 SHGs in March 2008 had outstanding bank loans of US$4,250 million5 and about 68 percent of the banks reported more than 80 percent recovery of loans from SHGs.
The progress of SHGs in India has been monitored by NABARD, other financial institutions (e.g., MYRADA 2002), and individual researchers (e.g., Harper 2002, Kropp andSuran 2002).In the case of India, a number of studies have examined the impact of the SBLP on various socioeconomic parameters of SHGs. 6Most of them have assessed the general impact of the SBLP but have not distinguished its impact from a gender perspective.The first study of the impact of the SBLP on SHGs that could be considered as an impact study was carried out for NABARD by Puhazhendi and Satyasai (2000).The study used a multistage stratified random sampling method to assess the impact of microfinance on Eventually, they approach a bank and leverage their accumulated savings for higher loans, which they intermediate within the SHG.socioeconomic7 conditions of 560 household members in 223 SHGs located in 11 states.They found that the average value of assets per household (including consumer durables and livestock) increased by 1.72 times in the post-SHG period from the level of Rs6,843 during the pre-SHG period.Average household saving, a meager Rs460 during the pre-SHG period, increased by 214 percent in the post-SHG period.The share of households among the SHGs living below the poverty line (42 percent) decreased to about 22 percent in the post-SHG period.With regard to employment performance, the study found that employment increased by 17 percent between the pre-and post-SHG periods.On empowerment, they found that involvement in SHGs had significantly contributed to the selfconfidence of the participating women.
A study by Puhazhendi and Badatya (2002), also carried out for NABARD, assessed the impact of the SBLP on SHG members in the Indian states of Orissa, Jharkhand, and Chhattisgarh.The study followed a multistage random sampling method.The findings suggest an increase in household savings and assets for the SHG members after they formed the group.The average loan amount per member increased significantly by 123 percent between the pre-and post-SHG periods.About 83.3 percent of the groups had promptly repaid the loans and only 16.7 percent had repaid late.The net annual income of SHG households increased by 23 percent after joining the SHGs.Employment was found to have increased by 34 percent on average in each SHG household.The share of SHG member households below the poverty line fell from 88 percent to about 75 percent.
A more recent study on the impact of the SBLP on SHGs reported that the net household income of SHGs registered a significant growth of 6.1 percent per year between the pre-and post-SHG periods (NCAER 2008).Net change in the value of consumer durable assets per household was Rs4,329 and the assets recorded a high annual growth of 9.9 percent between the two periods.Average savings (financial and physical savings) per household registered an annual growth of 14.2 percent between 2002 and 2006.The average loan amount per household registered an annual growth rate of 20.5 percent between the pre-and post-SHG periods.The study also reported that about 93.0 percent of households took out loans in the post-SHG period as compared to 46.5 percent during the pre-SHG period.Loan repayment was also very good as 96.4 percent of households made regular repayments on their loans.One of the important findings of the study is that the share of households living below the poverty line decreased from 58.3 percent to 33.0 percent between the two periods.Regarding the impact of the SBLP on social empowerment of women, the study found that about 92.0 percent of households reported that women were more empowered socially after joining an SHG.
To sum up, the broad findings of the impact studies so far suggest that the average value of assets and savings of SHG member households increased significantly during the post-SHG period.Average annual net income of SHG member households increased significantly and this considerably contributed to poverty reduction.Employment per household also increased.There was remarkable improvement in social empowerment of SHG members in terms of self-confidence, decision making, and communication.
Some other studies such as Nair (2005), Moyle, Dollard and Biswas (2006), and Chakrabarti (2004) have assessed more specific issues such as the role of SHG federations in the sustainability of SHGs and the economic and personal empowerment of women, as well as the role of microfinance in poverty eradication.The major findings of these studies are that SHG federations play a critical role in improving the sustainability of SHGs through financial and organizational support.Besides promotion of savings among SHG members in general and savings for education, housing, marriage, and festivals in particular, they also provide loans to SHGs at low interest rates.As far as organizational support is concerned, federations employ their own resources in promoting new SHGs and have been able to reduce the cost of promotion of SHGs as compared to other agencies such as banks and nongovernmental organizations (NGOs).A specific study on women's empowerment (Moyle, Dollard and Biswas 2006) found that a large share of female SHG members reported significant development of their self-confidence and work efficiency despite the challenges they face due to the work and responsibilities involved with being an SHG member.
The present study differs from earlier studies in several ways.As mentioned earlier, not many studies have given due attention to gender analysis of SHGs.The preliminary analysis of the present study suggests that all-female SHGs are performing much better than all-male SHGs in social and economic activities.Therefore, gender analysis of SHGs is an important issue from the policy prospective because we need to understand why a particular group of SHGs is performing better and getting more benefits out of financial outreach than others.Another important issue is sustainability of SHGs, which has not been thoroughly examined in the literature.Sustainability of SHGs is critical because many SHGs, particularly all-male ones, are unable to keep on functioning in the long term.Some of the factors causing this are mismanagement of money borrowed from the bank and lack of accountability and commitment toward expanding the group activities.Our objective is to explore these two critical issues of SHGs from a gender perspective, and identify best practices to help in improving SHG sustainability and eradicating poverty.
The remaining structure of the paper is as follows.Section II deals with the sample design and the methodology of the sample selection process.Section III covers descriptive analysis of the performance of SHG groups.Section IV discusses the issue of the sustainability of different SHG types using an econometric tool.Section V concludes the paper and offers policy suggestions.

II. SAMPLE DESIGN AND METHODOLOGY
The study is based on information obtained from a primary sample survey conducted in six states in India: Andhra Pradesh and Karnataka (southern region), Maharashtra (western region), Orissa (eastern region) Uttar Pradesh (central region) and Assam (northeastern region) (for details, see Appendix 1).These states together cover 77 percent of the bank-linked SHGs in India as of March 2002.Therefore, the total sample size covered in this study could be considered to represent all of India.From the gender perspective, the total number of bank-linked8 SHGs have been classified into three types: SHGs with only male members (all-male SHGs); SHGs with only female members (all-female SHGs); and SHGs comprising both male and female members (mixed SHGs).A description of the sample is given in Table 1.The analysis is based on a total sample of 961 SHGs, out of which 95 are all-male, 811 are all-female, and 55 are mixed.A multistage sample design was adopted for selecting the sample SHGs to be interviewed in the survey.In order to assess the performance of SHGs from the gender perspective, comparisons between the base year (2002) and the reference period (2006) were made (for details, see Appendix 1).A thirteen-page questionnaire was used to collect qualitative data, using the focus group discussion method, and quantitative data, using the records maintained by SHGs.In order to maintain the consistency of data and assess the validity of the information with the field workers or the NGOs, cross questions were used to obtain reliable information on the pre-SHG situation.All the economic parameters for pre-SHG (before bank linkage) and post-SHG (after bank linkage) were measured at reference year prices.

III. PERFORMANCE OF SHGS
This section examines the performance of the three types of SHGs under the SBLP.Before describing the analysis, we present an overview of the sample SHGs.The overview includes such important indicators as the number of years of bank linkage, number of literate members, distance of SHGs from banks and nearest towns, and number of members belonging to disadvantaged classes.The performance indicators are defined and analyzed in terms of the development of each SHG and its members in various socioeconomic parameters after the group became linked to the bank.The performance indicators include financial management practice, savings activities, loan recovery, and improvement of technical skills.

A. Profile of the Sample SHGs
The category-wise profile of the sample SHGs shows that the average number of members per SHG in the sample is around 13 (Table 2).Among the SHG types, the average number of members in all-female SHGs is higher than in all-male and mixed SHGs.The average number of years of bank linkage of SHGs shows that all-female SHGs in the sample have been operating for longer than other types of SHGs.This suggests that the performance of all-female SHGs is good or that they are well organized in terms of adhering to the rules and regulations set by the banks or NGOs.The survey results show that all-female SHGs have been bank-linked for longer than other SHG types.The number of years of bank linkage is also a critical factor to judge the performance of SHGs.The chances of becoming defunct are very high for an SHG in the initial stages of its existence.The average number of years of bank linkage for all types of SHGs is 5.5.It varies slightly among different types of SHGs; all-male SHGs had the shortest length of bank linkage with an average of 5.1 years.But the bulk of SHGs in the sample had been bank-linked for more than five years.Out of the 961 SHGs, 56.4 percent had been in existence for 3-5 years and 43.5 percent for more than 5 years.The percentage of SHGs that had been banklinked for more than five years was considerably higher in all-female SHGs as compared to all-male SHGs.The distances of the SHGs from banks as well as the nearest towns are important considerations from the point of view of the development of the SHGs, as the income-generating activities chosen by the SHG or its members depend on the availability of the market.While the average distance of the SHGs from the bank was 5.5 kilometers, the distance from towns was more than 13 kilometers (Table 2).The average distance from banks for allfemale SHGs was 5.7 kilometers, which is greater than the average distance from banks for other types of SHGs.So, despite female SHGs being farther from banks than other SHGs in the sample, they have been in existence for a longer time.The number of years of bank linkage or socioeconomic performance of SHGs could be important reasons of such an outcome.These issues will be discussed further in Section IV.

B. Distribution of Sample SHGs by Member Characteristics
The distribution of the sample SHGs by various socioeconomic characteristics of the members is presented in Table 3.The members of the sample SHGs were reported to be young, with an average age of about 37 years.In an ideal SHG, the members should have similar social and financial backgrounds.This contributes to easier interaction and smoother communication among members, facilitating equal opportunity of selfexpression within the group.In terms of distribution of SHG members by social groups, the percentage of SHGs with members from the schedule caste, schedule tribe, and other backward classes totaled 42.4 percent.The percentage of SHGs with only schedule caste and schedule tribe members was substantially higher in all-female SHGs than in all-male SHGs.In the other backward classes category, the percentage was higher in all-male SHGs than in all-female SHGs.SHGs with members only from the general category were found to be much less common than the other categories.This suggests that people from the disadvantaged classes, who tend to live below the poverty line, are more involved in self-help group programs regardless of gender.
In order to study the percentage distribution of SHGs by members' literacy levels, the SHGs were divided into three categories: those in which all the members were illiterate, those in which up to 50 percent were literate, and those in which more than 50 percent were literate.9A negligible percentage of allfemale SHGs (0.6 percent) fell into the first category.While a higher percentage of all-female SHGs (51.4 percent) fell into the second category, a lesser percentage of all-female SHGs (48.0 percent) fell into the third category.Interestingly, illiteracy among the all-male SHG members was found to be nil.
SHG members tend to pursue multiple income-generating activities to sustain their livelihood.We have divided the various income activities into three broad categories: farm, non-farm, and mixed activities.The largest proportion of SHGs (68.7 percent) had members involved in mixed activities.This trend was shown for all types of SHGs by gender.A greater proportion of all-male SHGs were involved in farm activities, while the non-farm activities showed the reverse trend, with a greater proportion of involvement on the part of all-female SHGs.

C. Financial Management Practices
In order to carry out monetary transactions, SHGs must maintain a number of documents.Table 4 looks into the financial management practices observed by the SHGs in the sample.The books of accounts of SHGs were, to a considerable extent, maintained by literate group members.This percentage was observed to be highest for all-female SHGs (81.0 percent), followed by mixed SHGs (74.5 percent) and all-male SHGs (71.6 percent).The same was seen with regard to maintenance of the SHG passbooks as well as passbooks of individual members.The role of NGOs or other representatives in maintaining the documents of SHGs was small in all three types of SHGs.A very low percentage of SHGs rarely update the book of accounts (9.5 percent), SHG passbooks (10.5 percent), and members' passbooks (11.2 percent).

D. Savings Activities of SHG Members
Improved savings activities through SHGs are one of the most important objectives of the SHG bank linkage program.Each member in an SHG pays certain membership fees.For any delay or late payment, SHGs may also charge a certain amount as a penalty.After collection of a substantial amount of money through these fees, SHGs deposit the funds in a bank and become credit linked.One of the basic principles of SHGs is that each member saves a small amount that helps the group in getting bank loans at lower interest rates, which is particularly beneficial for those who are otherwise ineligible for bank loans.This is primarily intended to inculcate and strengthen the habit of saving in SHG members.Table 5 depicts the frequency and amount of savings activities by SHG members.Notes: Rs45 = US$1.T-value between all-female and all-male SHGs, and all-female and mixed SHGs is statistically significant at the 1% level, which implies that their mean value is significantly different.T-value of average amount of own funds available at the time of bank linkage and 2006 is statistically significant at 1% level for all three categories of SHGs.Source: Authors' calculations based on survey data.
Saving on a monthly basis seems to be the favored practice; more than three fourths of the SHGs prefer monthly savings.The average mandatory saving per member in the sample SHGs was Rs35.6 in 2006.There was, however, a great deal of variation in monthly saving habits among the three types of SHGs, ranging from about Rs30.9 for mixed SHGs to Rs36.1 for all-female SHGs.The average amount of savings per year per member in 2006 was Rs457.8.Mixed SHGs, with Rs380.0,recorded the lowest yearly savings as compared to allfemale SHGs with Rs467.6 and all-male SHGs with Rs418.6.Regarding the availability of SHGs' own funds, a significant improvement was reported in 2006 as compared to the amount at the time of bank linkage.However, the improvement did not vary much among different types of SHGs.

E. Financial Sustainability of SHGs
The financial sustainability of an SHG depends mainly on its repayment patterns.Only when SHGs make timely repayments to banks do they gain access to bigger loans.The ability to make timely repayments to banks, in turn, depends on the repayment patterns of the individual members.As discussed earlier, some SHGs impose a penalty on delinquent members, as peer pressure alone is not always enough to ensure that members repay on time.Table 6 summarizes the survey results on financial sustainability of the SHGs.
The percentage distribution by loan recovery reveals that 73.7 percent of sample SHGs reported 100 percent repayment by members.Among the three types of SHGs, 75.1 percent of all-female SHGs reported 100 percent repayment by members, which is considerably higher than all-male SHGs (56.8 percent) but lower than mixed SHGs (81.8 percent).
The financial performance of SHGs in 2006 is also indicated by loan amount outstanding.The average outstanding amount for all types of SHGs in 2006 was Rs28,779.The amount was greater in all-female SHGs (Rs35,237), followed by all-male SHGs (Rs30,657) and mixed SHGs (Rs20,442).This suggests that banks are more willing to disburse loans to all-female SHGs than to other types of SHGs.This may be because of the better repayment rates of allfemale SHGs.

F. Awareness and Benefits of SHGs
The performance of SHGs also depends upon members' awareness of the overall objectives of SHGs, as well as SHGs' capabilities to develop their members' managerial and technical skills.A variety of training and awareness programs conducted by NGOs and other social welfare agencies are available to groups for the latter purpose.Members' awareness of the objectives of SHGs is reported in Table 7. Overall, 42.8 percent of SHGs reported that their members were very familiar with the objectives of the SHG programs.However, a higher percentage of SHGs (52.6 percent) reported only moderate awareness of the objectives of SHGs.Among the different types of SHGs, all-male SHGs had the greatest percentage reporting high awareness.Benefits achieved through formation of SHGs seem to be high in all-female SHGs, where 49.6 percent reported having developed members' technical or professional skills, followed by 49.5 percent of male SHGs and 32.7 percent of mixed SHGs.The survey results on performance indicators of the three types of SHGs suggest that all-female SHGs perform better than other types of SHGs.All-female SHGs reportedly have higher rates of savings per member than other types of SHGs.They also report better performance in terms of loan recovery.This superior performance could be explained by their longer periods of being in operation and being linked to banks as compared to other types of SHGs.Hence, one could draw a tentative conclusion that the better is the performance of SHGs, the higher is the chance of sustainability.The issue of sustainability for the three types of SHGs will be examined in the next section.

IV. SUSTAINABILITY OF SHGS
Considering that the SHG bank linkage program has been operating in India for the last 25 years, the sustainability of SHGs has not been paid much attention in studies so far.Most of the existing studies on the SBLP (Armendariz de Aghion and Morduch 2000, Puhazhendi and Satyasai 2000, Puhazhendi and Badatya 2002, Kropp and Suran 2002, Hannover 2005) have focused only on the performance and impact aspect of the SBLP on SHGs and SHG households.
A sustainable SHG would be one that continues to function well over a long period of time.Moreover, the long-term performance depends on members' overall socioeconomic development, including their empowerment.Hence, the analysis here on sustainability covers the performance of SHGs on various socioeconomic aspects and their dependency on self-help promoting institutions (SHPIs) for various developmental activities.A question was included in the survey about SHGs' economic and social performance and overall success to determine whether they would be dependent on SHPIs for running the group in the future.The preliminary observation from the data suggests that the majority of socially and financially empowered SHGs could sustain their group without help from NGOs and banks.Socially and financially weak SHGs continued to depend on SHPIs.In this paper, we have discussed this issue from a gender perspective.The performance analysis in the previous section suggests that all-female SHGs performed better in economic and social aspects.Further, within a particular SHG, all-female SHGs are not depending on SHPIs and are better off in terms of per capita savings and loan recovery than others.In order to validate our hypothesis, we have applied a binary econometric technique to measure the sustainability of SHGs.The econometric model is discussed in the next section.

A. Model Estimation
Since sustainability, the dependent variable, is a qualitative variable, a logit model is used to estimate the sustainability equation.A logit model could be written as: where L is the logit; P i is the probability of SHGs functioning without the help of SHPIs; and (1-P i ) is the probability of SHGs functioning with the help of SHPIs.As Z varies from - to + (P goes from 0 to 1), the logit L goes from - to +.
Based on the performance analysis in the previous section, we have selected some of the important indicators for explaining the sustainability of SHGs: literacy of SHG members, SHGs' performance in loan recovery and savings, linkage with SHG federations, types of SHG models, and gender categories of SHGs.Literacy helps members to understand the rules of group formation as well as financial transactions with banks and other financial institutions.It also facilitates communication and participation of members in public forums and meetings.Moreover, literate members can easily grasp the information provided in training programs delivered by banks and NGOs.Literacy helps members to depend less on NGOs, which have worked as facilitators for managing the records and books of some SHGs.Loan recovery and savings activities of SHGs are also important for group success in the long term.Better loan recovery suggests better management of SHGs' financial portfolios.This helps SHGs to distribute loans more effectively to meet the financial needs of each member and enlarge the group's funds.The savings per member explains the financial strength and economic condition of SHGs.An increase of savings per member suggests the SHG is functioning well and using its financial resources effectively.Sustainability of SHGs also depends upon the linkage to a federation of SHGs.Usually, SHGs that are linked with a federation receive generous help from other well developed SHGs in such areas as managerial and professional skill development training, capacity building, and participation in local and block-level governance.Moreover, linkage with a federation empowers SHGs to make independent decisions in managing their financial resources and spending them for productive activities.
The following logit model for sustainability of SHGs was estimated 10 : where sustain equals 1 for sustainable SHGs (i.e., not dependent on SHPIs) and sustain equals 0 if the SHG depends on SHPIs."Literate" is the number of literate members in the SHG, "loanrecov" is the percentage of loan recovery from the SHG's own fund (more than 90 percent equals 1, otherwise zero), "savingspc" is per capita savings by the SHG in 2006, "federation" is membership in an SHG federation, "femaleshg" is SHGs with only female members, "mixedshg" is SHGs with both male and female members, and "modeltype1" is SHGs formed and financed by banks. 11The error term u i follows the normal distribution with zero mean and variance equal to 1/[N i P i (1-P i )].
After identifying various factors that contributed to sustainability of SHGs, we have followed a more disaggregated analysis on the sustainability of SHGs from a gender perspective.Various factors have been identified that determine the sustainability of all-male, all-female, and mixed SHGs.A logit model for each type of SHG was estimated separately as follows: 10 The set of variables in equation 2 and in subsequent equations was chosen after estimating different equations with other variables such as SHG's passbook maintained by literate member, amount of loan taken by SHG, SHG model types 2 and 3, and SHG members involved in farm and non-farm activities.
11 The different SHG model types are described in Appendix 1.
where "maleshg," "femaleshg," and "mixedshg" are the sustainability of all-male SHGs, all-female SHGs, and mixed SHGs, respectively (equaling 1 if the SHG is not dependent on SHPIs, and equaling 0 if the SHG depends on SHPIs).Model type 3 is SHGs financed by banks using NGOs and other agencies as financial intermediaries.

B. Analysis of Results
This section covers the analysis of results obtained from the estimation of the equations given in the previous section.The information on basic statistics (mean and standard deviation) for each variable is given in Appendix 2. The logit model results of sustainability of SHGs are given in Table 8a.The results show that all the explanatory variables together have a significant impact on the dependent variable (i.e., sustainability), as the LR statistic is 391.11 and its p-value is 0.00, significant at the 1 percent level.Among the explanatory variables, all except literacy and mixed SHGs are statistically significant.As discussed earlier, loan recovery and savings per capita of SHG members are critical factors that ensure the sustainability of SHGs in the long run.The coefficient of loan recovery is 0.163, which implies that with other variables remaining constant, a one-unit increase of loan recovery leads to an average 0.163-unit increase of sustainability.Similarly, savings per capita of SHG members between bank linkage (2002) and the reference period ( 2006) is positive and has a significant impact on sustainability of SHGs.Other important factors that contribute to the sustainability of SHGs are linkage with an SHG federation and being formed and financed by banks (model type 1).The findings suggest that literacy also has a positive impact on sustainability.
The more meaningful interpretation of a binary regressor will be in terms of taking the antilog of its coefficient.In this case, the coefficient of federation suggests that SHGs that are linked with federations are more than 3 times (antilog of 1.19 = 3.28) more likely to be sustainable than SHGs that are not linked to federations.
In estimating the sustainability of SHGs by gender, we presumed that allfemale SHGs would be more sustainable because their performance in both economic and social terms is better than other types of SHGs.This hypothesis is drawn from the previous section and from case studies reported by NCAER (2008), in which most of the all-male SHGs were defunct and those in operation were unable to repay their loans due to crop loss, natural disasters, or conflicts among the members.All-male SHGs were included in the estimated model to find out the likely impact of all-male membership on sustainability.Table 8b shows that the coefficient for mixed SHGs is insignificant but that for all-female SHGs is significant.The coefficient for all-female SHGs suggests that they are more

C. Caveat
The lacunas of the present study are that we could not cover certain factors of empowerment such as the decision making processes of women and men in children's education, health, family expenditures, and purchase of assets.The information on these variables is collected at the household level but could not be mapped to the group level.We also could not discuss factors such as income and expenditure patterns, which are available at the household level.These issues could be addressed in a separate study using the household-level data.
The sample size in this study for all-male and mixed SHGs was smaller than that for all-female SHGs.This could be a weak point of the study because the comparison of the three models is not exactly representative.However, given that fewer all-male and mixed SHGs are in existence in the selected sample states (and the situation may be similar in other states), under the present condition of data availability, the best available data was that from the NCAER (2008) field survey.

V. CONCLUSION AND POLICY SUGGESTIONS
This paper has examined the performance and sustainability of three different types of SHGs: all-male, all-female, and mixed SHGs.The analysis was based on data from a primary survey from six states in India (NCAER 2008).Overall, the performance analysis revealed female SHGs to be doing better than other types of SHGs.A number of reasons could underlie this outcome.First, female SHGs performed better in terms of loan recovery than other types of SHGs.The per capita savings of female SHGs in 2006 was also much higher than other types of SHGs.The female SHGs also stood out as doing extremely well in financial management practices such as maintaining book accounts and passbooks and updating them regularly.
On the issue of sustainability of SHGs as explained by long-term performance, the econometrics results revealed that only female SHGs were sustainable.The factors that determine sustainability were loan recovery, per capita savings of SHG members, linkage with SHG federations, and formation and financing of SHGs by banks.The all-female SHGs were sustainable because they were more focused and united, adhere to basic objectives of groups, utilize borrowed funds for different productive activities, and are highly concerned about the well-being of their children and family members.Further, female SHG members took membership in the group as a means to educate themselves and confront social, political, and economic problems.On the other hand, members of all-male SHGs have ego problems, work for their own interest, and do not follow the basic objectives and goals of group formation.As a result, they are most irregular in loan repayments and perform badly in economic and managerial activities.
Planners and financial institutions should focus on policy decisions to improve the performance of all-female SHGs so that they can progress to the next level of economic and entrepreneurial activity and improve the living standards of their members.Further, planners need to provide training programs to all-male and mixed SHGs using best practices demonstrated by all-female SHGs.The policy implication is that the credit linking of all-male or mixed SHGs should not be encouraged without first addressing the weaknesses in such groups with respect to group objectives, coordination among group members, social responsibility, financial management, and means of resource use.

APPENDIX 3. CASE STUDY: MALE SHGS' FAILURE TO FLOURISH
This case study was carried out during our field visit to Kalahandi district of Orissa, India, one of the least developed districts of the state.The case study focuses on two all-male self-help groups (SHGs), which were on the verge of becoming defunct.A lack of unity among the group members and infighting for group leadership were the main factors responsible for the failure.Besides that, the lack of foresight in investing the bank loan in the most productive way resulted in financial trouble for the groups.The members also revealed that these kinds of problems do not exist in the case of all-female SHGs, which may explain why they flourished.
An all-male SHG called the Maha Laxmi SHG, located in Kikia village of Kesinga block, was formed by 12 members in 1999.In 2001, the SHG was linked to the Utkal Gramya Bank in Kesinga.Like any other SHG, the desire to improve living standards and conduct income-generating activities motivated the members to form this group.The SHG also wanted to play an active role in the development of the local community.Unfortunately, some unforeseen circumstances and their internal differences have hampered the progress of the Maha Laxmi SHG and it seems to be facing a serious financial crunch.
The Maha Laxmi SHG got a loan of Rs15,000 (US$333.3 at the exchange rate of Rs45 to US$1) from Utkal Gramya Bank in 2001 to start a fertilizer business.Their motive was to sell fertilizer to cotton farmers in the area and make some profits from that.Cotton farming is one of the most profitable businesses in the area due to the suitable local soil and climate.However, due to a natural disaster in the form of a severe drought in 2001, the farmers lost their all crops and could not pay a single rupee to the Maha Laxmi SHG, from whom they had received the fertilizers in advance.As a result, the hopes of the SHG were shattered and they could not repay their loans to the bank.In addition to this, the SHG had several other problems.The members of the group admit that their constant infighting and ego problems are major obstacles to their progress.As a result, plans have not been implemented properly.Monthly collections have stopped and meetings are not held on a regular basis.
Another problem faced by this SHG is alcoholism.As a result of some members' alcoholism, the group was not able to motivate the community to give up drinking.The members are honest about this and understand the need to keep away from liquor.They realize that economic development is not possible if all the money earned is spent on drinking.They also said that drunk members are always fighting, which is not good for the unity of the group.In contrast, most of the all-female SHGs in the same area are up in arms to fight against drinking.
Improving the performance of the SHG is high on the agenda of the members.They have decided to collect small amounts of money regularly to rebuild the SHG fund.This would help them to meet some expenses later on.The paddy business and pisciculture are what they plan to take up next.Making plates from dried leaves is also an option, but with forest areas being wiped out, the members do not know whether they could get enough leaves.The members feel that timely inspection by the field officer would help them escape their problematic situation.In addition, proper training in vocational activities, financial management, and other skills would be helpful in securing the sustainability of the group.
We observed a similar story in another all-male SHG in Kalahandi district of Orissa, India.The Maa Laxmi SHG in Latakabahali village in Kesinga block is facing difficulties due to more or less the same reasons: ego problems, lack of unity, and fighting for group leadership.The group consists of 12 male members, all of whom are poor farmers.All are literate and have studied up to the secondary school level.The group, promoted by a Bhawanipatna-based NGO called Self Employed Women's Association, was formed in 2000 and was linked to Utkal Gramya Bank in 2001.
The group suffered due to the activities of one member who was dissatisfied with the SHG on financial grounds.He is no longer a member of the group, but the incident affected the SHG.Meetings and collections became irregular.Now, efforts are on to restore the activities of the group.Members admitted to the utter lack of unity within the group, which they know is detrimental to productive work.Maa Laxmi members praise all-female SHGs for their unity.They admire the understanding and coordination among female SHG members.Male members of SHGs have problems accepting leaders due to their aggressive attitudes.However, they feel that the male members, being more mobile than their female counterparts, also have an advantage.
The Maa Laxmi SHG took a loan of Rs20,000 (US$444.4) from the Utkal Gramya Bank in 2001 to invest in cotton farming.Each member received Rs1,600 (US$35.6).This group also fell victim to the 2001 drought and was unable to repay the bank loan.The members regret not having taken out any insurance policy.The members have managed to repay only Rs7,000 (US$155.6) of their outstanding debt to the bank.Nevertheless, the Maa Laxmi SHG was successful in some of its social activities.All of the members of the group joined hands to close a liquor den in the area, unlike the other all-male SHG.They also organized a campaign against malaria.As a result, people became aware of health hazards that can be carried by mosquitoes.The group helped to clean drains in the village so that the drains would no longer provide breeding grounds for mosquitoes.People have been encouraged to use mosquito nets.Members cited an interesting case in which they celebrated the holiday of Holi by cleaning the village.Festivals provide an opportunity for the community to interact.The Maa Laxmi SHG has provided a forum for this by organizing various festivals, which has made the group popular in the village and in the wider area.
Since most of the members in the Maa Laxmi SHG are educated, they were concerned about educating their children and others' children as well.Some of the SHG members are on the school committee.A member of the group is on the village education committee.
To sum up, what makes one feel optimistic about both of these all-male SHGs is that in spite of their financial problems, they are hopeful of turning the corner and doing constructive work in the future.Fortunately, the members are able to understand where the problems lie and they have some hope of solving their problems.

Table 4 .
Financial Management Practices NGO = nongovernmental organization; SHG = self-help group.Source: Authors' calculations based on survey data.

Table 5 . Savings Activities of SHGs Members All-Male All-Female Mixed All Types Distribution
of SHGs by frequency of saving services offered in 2006 (%) Rs = rupee; SHG = self-help group; US$ = United States dollar.

Table 7 . Awareness and Benefits of SHGs, 2006 All-Male All-Female Mixed All Types
SHG = self-help group.Source: Authors' calculations based on survey data.