A qualitative analysis of savings and internal lending communities in Haiti – do they make a difference?

ABSTRACT

In 2013, the United States Agency for International Development (USAID)/Haiti commissioned an external qualitative assessment of a four-year Savings and Internal Lending Community (SILC) activity funded by USAID that had recently ended. USAID wanted to know whether SILC met members’ needs; had an impact on individual, child and household well-being; whether SILC groups continued after donor support ended, and why or why not. Using a purposive sample of high- and low-performing groups, in-depth interviews were conducted with SILC members and programme implementers including community mobilizers (CMs) and programme staff. Focus group discussions were conducted with SILC members. Interviews and discussions were conducted in Creole, translated verbatim into English and analysed in Atlas-TI. Respondents reported that SILC group activities met members’ needs; members used loans and share-out funds for business investments, school fees, health-related expenses, household consumption of purchasing land or livestock; and that SILC groups differed from other financial institutions due to lower interest rates, sense of ownership of funds, and community solidarity and collective action. Many groups remained active and new groups were created after donor support ended. Other groups disbanded when stipends for CMs ceased, due to lack of engagement from CMs, reduced group member motivation and/or the economic situation. The key factor ensuring success and sustainability seemed to be groups’ sense of ownership of their capital and of the SILC methodology. This assessment produced important findings that can be used to adapt current savings and lending group programming and inform future programming within Haiti and beyond.

KEYWORDS:

• Household economic strengthening
• savings and lending groups
• Haiti
• evaluation
• qualitative
• sustainability

Introduction

Haiti is the poorest country in the Americas: a quarter of the population lives in extreme poverty (World Bank, 2016). Reducing the economic vulnerability of low-income families and enabling them to meet the essential needs of their children is critical (The U.S. President’s Emergency Plan for AIDS Relief, 2012).

Savings and Internal Lending Communities (SILC), a household economic strengthening programme to create user-owned, self-managed savings and credit groups, have been implemented in many developing countries for low-income families that do not have access to formal-sector financial services such as banks (Ferguson, 2012). All SILC groups include the following elements:

• To join a SILC group, members must contribute a minimum amount at each meeting.

• Members decide how often to meet, the minimum contribution amount and how long to function. Cycles typically range from 9 to 12 months.

• Pooled contributions create a loan fund for members to be repaid with interest and a social fund to help members with emergency situations.

• By the end of the cycle, all loans are repaid. Accumulated savings and interest earnings are paid out in proportion to members’ contributions.

• After pay-out, the group may disband or decide to continue for another cycle and may invite new members to join.

From 2009 to 2013, the United States Agency for International Development (USAID)/Haiti funded the Community Health and AIDS Mitigation Project (CHAMP), led by Family Health International in partnership with Catholic Relief Services (CRS) and International Child Care. The overarching goal was to improve the health and quality of life of orphans and vulnerable children (OVC) and their families and people living with HIV (PLHIV) through community-based health service delivery and social assistance (Family Health International, 2009). As part of CHAMP, CRS Haiti developed SILC groups (Mutuelle de Solidarités-MUSOs in French) in 5 of Haiti’s 10 departments: Artibonite, Grand’Anse, Nippes, Nord-Ouest and Sud (Catholic Relief Services, 2013).

Midway through the project, the implementation strategy changed from a facility-centred approach serving PLHIV and directly supported by CRS Haiti to a community-centred approach open to all members of the community and supported by community mobilizers (CM) recruited locally and hired by local community-based organizations (CBOs). The original approach created fewer groups than planned, and there were concerns that PLHIV-only SILC groups could inadvertently stigmatize their members.

An internal baseline survey estimated that 60% of the SILC members in the CHAMP project were living on less than US$1 a day. More than half (59%) of the households surveyed had never taken out a loan prior to their participation in the SILC group, with one in three reporting having taken out a loan from a money lender. Importantly, over two-thirds of participants were women (Catholic Relief Services, 2012).

Shortly after CHAMP closed, USAID/Haiti commissioned the present study to assess the SILC activities and to help inform follow-on programming. USAID/Haiti wanted to know the impact of programme participation on members’ well-being as well as that of their households, children and communities. The study was conducted by a global HIV-focused research consortium.

Methods

This study was a qualitative retrospective examination of savings group members’ needs and the impact of their participation in the programme from the perspectives of SILC group members themselves and project implementers. The retrospective nature of this study also provided the opportunity to assess SILC sustainability. The study had three specific research questions: (1) Did respondents feel that the SILC group activities met members’ needs? (2) Did respondents feel that participation in SILC groups had an impact on individual, child and household well-being? and (3) Did SILC groups continue after donor support ended, and why or why not?

Study sample

We used a purposive sample to capture the greatest range of experiences. Project and CBO staff classified SILC groups in each of the five departments where the project operated as high- or low-performance based on criteria developed jointly between the researchers, CRS staff and the CMs. Classification criteria included number of meetings held, solidarity among members, loans made, loans repaid on time and cooperative investments among members. The research team randomly selected seven high- and seven low-performing groups (for a total of 14 groups out of the 159 groups created) and further refined the sample to ensure inclusion of groups that had or had not completed the first cycle, were on their first or second cycle and were still active or had disbanded. Nine of the 14 groups were still active. The study coordinator and CMs worked together to ensure a balance of SILC members in terms of gender, age and roles. The CM for each group selected was also asked to be interviewed.

Data collection and analysis

The research team contracted the Interuniversity Institute for Research and Development (INURED) collect data and a local study coordinator to oversee the process. Data collectors completed a one-week training by the research team and conducted a pre-test in Port-au-Prince. INURED conducted in-depth interviews (IDIs) with nine CRS Haiti staff, 14 CBO staff and 14 CMs, as well as 56 IDIs and 14 Focus Group Discussions (FGDs) with SILC group members. Four to eight members of each SILC group participated in the FGDs and another four members not participating in the FGDs participated in the IDIs. Gender and age were not collected for all respondents, as in some cases it was not culturally appropriate and in other cases to protect respondents’ identities. Interviews and focus groups were conducted from May to July 2014. All were conducted, recorded and transcribed in Creole and translated verbatim into English. The research team analysed the transcripts following five steps: reading, coding, displaying, reducing and interpreting (Miles & Huberman, 1994; Ulin, Robinson, & Tolley, 2005). Translated transcripts were analysed in Atlas-TI version 7.5.2. To check reliability, 5% of the transcripts were double-coded.

The study was approved by the INURED ethical review board.

Results

Research question 1: Did respondents feel that the SILC group activities met members’ needs?

Overall, SILC group members reported that they joined the SILC groups to help save money and access low interest loans with flexible conditions. They also were attracted to the community solidarity aspect of SILC groups that was unique in comparison to other alternative financial institutions.

Overwhelmingly, respondents thought SILC groups successfully met members’ needs and agreed that access to flexible savings and loans that were adapted to community needs filled an important service gap:

If I save it [money] at home, any little problem that I would have, I would want to take it, but if I put it in the group, in the treasury, the group helps me save, I have some hope. (SILC member)

Respondents also indicated that the SILC groups differed from other local more formal financial institutions in terms of lower interest rates and providing a sense of ownership of the SILC funds, and that SILC groups encouraged community solidarity and collective action such as street cleaning or cooperative farming:

The reason why it [MUSO] helped us with the small means that we deposit, we deposit together, it’s small but we create somehow, it helps us, without other treasuries lending to us. It’s us with our own savings that we help ourselves. (SILC member)

One SILC group member used the Haitian proverb The pig fat cooks the pig to explain how they increase their own capital. Another described the difference in interest rates between the SILC groups and other local financial institutions:

The advantages that I like in the group is when you borrow from the bank, they say the interest is 7%, and I come here and borrow and the interest is 1%, the other 6% is taking so much out of my pocket. When your interest is high, you come from high, you fall high, but when the interest is low you can rest easy, you make progress. (SILC member)

While a few SILC group members did not think the social fund was necessary and re-named the social fund the ‘lost funds’, most felt that the social fund was an important advantage over other financial institutions in that they could access funds in an emergency and that they were contributing to a fund that could help others in need.

Some SILC group members felt that the amount of funds available for loans was not sufficient to meet the demand of all group members, that the size of the loans was not adequate for larger businesses and prevented them from fully growing their businesses. In addition, SILC group members reported that members who were unable to contribute or re-pay loans created difficulties for the entire group. The majority of CMs mentioned that late repayment of loans was their biggest challenge.

While the SILC groups were open to all members of the community, the most economically vulnerable were not able to participate because they were unable to meet the minimum contribution requirements:

Well, the reason for it is that they may not have the means [to join the SILC group] … it’s because they have no business, which is why they can’t join, but it’s not that s/he’s not interested. There are those who give their names and don’t come, they’re embarrassed and they can’t find the money. (Community mobilizer)

Research question 2: Did respondents feel that participation in SILC groups had an impact on individual, child and household well-being?

SILC group members reported using their savings and loans to pay for school fees, clothing, food, health-care costs, land and livestock purchases, or to pay back another loan or lend to a friend or family member. Some described how their participation in SILC allowed them to surmount difficult economic conditions:

It gave me a lot of advantages because it allows me to send my children to school. I did not see how I was going to send them to school, because certain years things are not good, the harvest is not profitable. For now MUSO helps me because it helps me get money to send my kids to school. (SILC member)

SILC group members reported using the social fund for reasons such as death of a family member, child’s illness or other health-related expenses such as paying for treatment during the cholera epidemic.

Some SILC group members reported investing loans in their small businesses and generating additional income. Others reported that the financial management training provided by the CMs and access to SILC loans helped them create new businesses:

Due to the MUSO’s help I bought a piece of land within a year, with the MUSO again I rented land. Now if someone is surprised and saying he is this or that, I am not going to brag, but for me it is the MUSO that changed things for me. (SILC member)

What I liked the most about the MUSO, it’s that in that time, you see the number of people whose lives have changed. Some didn’t have any means to sell anything, the next day, when the person is running a small business, s/he can address some needs. (CBO staff)

Research question 3: Did SILC groups continue after donor support ended, and why or why not?

Of the 159 SILC groups formed during CHAMP, we were able to locate 83 (52%) groups at the time of the study: 69 groups were still active, either as SILC (67) or as a local cooperative (2). Thus, at a minimum, 43% (69/159) of the original cohort continued after donor support ended. Members of groups that continued felt the SILC groups met their financial needs and they were confident in their understanding and implementation of the savings group methodology and trusted that they would benefit from following it.

The MUSO is active because for us, it does a lot for us, we would take loans out, at the end of the year we always get our money that we saved in it, then we saw that the MUSO helped us save the money too. Then we saw that it was interesting for us, we won’t let a group like this fall apart. (SILC member)

CRS Haiti staff thought the communities had adopted the methodology and that this motivated them to continue participating in the groups.

Likewise, CRS Haiti staff, CBO staff, CMs and SILC group members felt that the CMs gave needed support for the groups to mature and thrive:

I think the mobilizer has supported us so much, it made us stronger; it helped us to work well because if there are some areas where we lack strength, he would give us a hand. He is the adult and we are the kids because he received training, he knows about these things. (SILC member)

Some groups did not continue after CHAMP ended. Many respondents felt that without CMs, the SILC groups could not continue. Members of the groups that had disbanded reported lack of engagement of their CMs, and the lack of funds to pay CMs after the end of the project was mentioned by both SILC group members and CMs themselves. Other reasons reported for group dissolution included inability of members to pay their contributions or repay loans, and lack of continued training and support.

Some SILC group members reported that their groups modified the SILC methodology after the end of the project, for example, by increasing group size while other members mentioned that their groups took ‘a pause’ and re-formed or planned to re-form at a later date.

In addition, many CMs reported that new SILC groups were created independently after CHAMP ended.

Discussion

Studies measuring the impact of participating in programmes such as SILC have shown mixed results (Annan, Bundervoet, Seban, & Costigan, 2013; Brunie, Fumagalli, Martin, Field, & Rutherford, 2014; Bureau of Applied Research in Anthropology University of Arizona, & Innovations for Poverty Action, 2013; Gash & Odell, 2013; Thuysbaert, Karlan, & Udry, 2012). While they have been shown to improve asset accumulation, consumption smoothing, investment in income-generating activities, management of finances, savings and solidarity with other members, there have been few documented long-term impacts on household income, poverty reduction, or education or health outcomes for children (Gash, 2013; Gash & Odell, 2013).

The present study provides an in-depth lens into the perceptions of those that participated in the SILC group component of the CHAMP project. Many of our findings echo those of previous studies. SILC group activities met members’ needs; members used loans and share-out funds for business investments, school fees, health-related expenses, household consumption of purchasing land or livestock; and SILC groups differed from other financial institutions due to lower interest rates, sense of ownership of funds, and community solidarity and collective action (Bureau of Applied Research in Anthropology University of Arizona, & Innovations for Poverty Action, 2013; DAI, 2014; Miller & Gash, 2010; Taneja, 2013). Likewise, funds were not always sufficient to meet demand and some SILC members were unable to make regular contributions (DAI, 2014; Emerging Markets Consulting, 2012). Finally, as has been shown in numerous studies of savings and lending group programmes, the most economically vulnerable were unable to participate (Lowicki-Zucca, Walugembe, Ogaba, & Langol, 2014; Miller & Gash, 2010).

The study also provides the opportunity to assess SILC sustainability. At least 4 out of every 10 SILC groups continued after CHAMP ended. Our findings suggest that groups with highly motivated members who trust the SILC methodology are likely to persevere. We also found that groups whose CMs stayed on after CHAMP ended were more likely to continue to function than groups whose mobilizers dropped out. The second finding is more ambiguous: while mobilizers clearly provide needed support early in SILC group formation, if they do not transfer financial management skills to the group’s executive committee, members may not gain the necessary skills and confidence to keep operating on their own. For example, a study in Cambodia compared groups with more or less support from community volunteers and found that groups that continued to be active were less likely to have had an assigned community volunteer attend meetings than dissolved groups; the authors speculated that the CM may have been conducting the majority of the financial management and not effectively transferring those skills to the management committees (DAI, 2014).

Finally, we found that new SILC groups were created independently after CHAMP ended. While it is encouraging that these new groups developed organically, presumably due to the success of the SILC methodology in the community, recent research has shown that using a more structured replication strategy can be even more effective (Bureau of Applied Research in Anthropology University of Arizona, & Innovations for Poverty Action, 2013).

Recommendations

This assessment produced important findings that can be used to adapt and inform current and future savings and lending group programming within Haiti and beyond. We make the following recommendations for programme design and implementation:

• Programmes should more clearly define short- and long-term expectations for CMs and ensure appropriate support as SILC groups develop and mature. This includes remuneration for their work in establishing and nurturing new groups, as well as training to transfer these skills to internal executive committees.

• For programmes implementing social funds, more effort should be made to address participants’ concerns about the utility of these funds from the outset and encourage an ongoing dialogue around the positive impact these funds may have on individual group members as well as the success of the savings group itself.

• Programmes should consider exploring possibilities for increased business-related training and community entrepreneurship opportunities.

• Programmes should consider addressing members’ demand for larger loans by facilitating linkages of mature SILC groups to other local financial institutions, after providing knowledge and skills about how they can effectively use additional financial services and by ensuring that financial institutions do not negatively impact the groups by predatory lending to members.

• Programmes should consider conducting operations research to examine components such as the role and impact of CMs, for example, to compare the effects of different schedules of mobilizer phase-out on group functioning and sustainability.

Acknowledgements

The research team would like to first thank USAID/Haiti for their continued support and guidance of the research team throughout the study. We also thank the HIVCore management team, Ms. Sarah Sandison, Ms. Alison Thurston and Dr. Glenn Post for their commitment and support as well as the USG OVC Technical Working Group for their inputs and insights. Special thanks go to Mr. Jason Wolfe, who provided valuable technical insight throughout the study.

The CRS Haiti team including Mr. Jeffery McIntosh, Mr. William Martin, Ms. Jocelyn Braddock, Mr. Bruce Lee, Mr. Lam Huynh and Mr. Edno Estigene were generous with their time and shared their wealth of experience on household economic strengthening programmes in Haiti with the research team and participated fully in the assessment. Mr. Tom Shaw also provided excellent feedback on the assessment findings.

This study would not have been possible without the dedication and diligence of the staff and data collectors, including Dr. Louis Herns Marcelin, Chancellor, and Mr. Frantzy Lejeune, data collection coordinator, from the Interuniversity Institute for Research and Development (INURED) in Port au Prince, Haiti.

The Population Council HIVCore staff provided tremendous support for this assessment, including Dr. Sam Kalibala, Ms. Nrupa Jani and Ms. Adaku Ejiogu. We would like to particularly thank Ms. Sherry Hutchinson for her helpful guidance on all things related to knowledge management.

We would also like to acknowledge the Palladium team, including Dr. Irit Sinai, Ms. Abir Doumit, Mr. Albert Merkel and Mr. Shahzad Bhatti, and particularly Ms. Kathleen Walsh, for her vital support during the implementation of this assessment.

Finally, we sincerely thank all of the participants in this assessment including the SILC group members, CBO staff, community mobilizers and CRS CHAMP staff for their openness and motivation in helping to produce important evidence on household economic strengthening programmes.

Disclosure statement

Dr. Olbeg Desinor is an employee of USAID/Haiti, which both funded CHAMP and the HIVCore study described in this manuscript. The opinions expressed are of the authors alone and do not reflect the views of USAID or the U.S. Government. The other authors have no competing interests to declare.

Additional information

Funding

This study was made possible through support provided by the President’s Emergency Plan for AIDS Relief and the U.S. Agency for International Development (USAID) via HIVCore, a Task Order funded by USAID under the Project SEARCH indefinite quantity contract [Contract No. AID-OAA-TO-11-00060].