COLLABORATING TO PROVIDE MICROFINANCE TO

COLLABORATING TO PROVIDE MICROFINANCE TO CAREGIVERS OF
ORPHANS AND VULNERABLE CHILDREN IN ETHIOPIA
by
Grace Bae Bahng
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(POLICY, PLANNING, AND DEVELOPMENT)
August 2009
Copyright 2009
Grace Bahng
DEDICATION
This dissertation is dedicated to the millions of children around the globe who have
had a father or mother taken away from them because of war, sickness, poverty, or
other misfortune. We cannot justify your loss but can only work to extend a home of
belonging and well-being to you.
This dissertation is also dedicated to the generation of Koreans who were orphaned
by the war, including my own mother. Your children are indebted to you for your
resilience, sacrifice, and diligence.
ii
ACKNOWLEDGEMENTS
I am not sure what the newly earned three letters after my name will mean for me
from this point forward, but I am extremely grateful for the time that I have had at
USC. My hope is that the time and energy that many have invested in me during my
time here will not simply remain with me but will somehow work its way towards
improving the inclusiveness of well-being in our global community.
To my family, I offer my deepest thank you for your continued support and
inspiration in my life. To my father, thank you for challenging me with your
diligence and generosity. To my mother, thank you for helping me to better
understand the plight of the orphan and for reminding me not to take the fullness of
my life for granted. To Christine and Tracy, thank you for your comforting
sisterhood and for showing me how to give your lives away. To my husband, Phil,
thank you for walking with me with through the highs and lows of this process,
especially towards the end.
To the folks at World Vision International and VisionFund International, thank you
for giving me a window into the good work you do around the globe everyday. To
Lisa Jackinsky, thank you for all the ideas, support, and follow-up emails. Thank you
also to Richard Reynolds, Worku Tsega, and Grace Sebageni for their support in this
iii
research. Finally, thank you to Shimeles Teferra for your invaluable contributions to
the data collection process.
My gratitude goes to my committee members as well. To my advisor, Dr. Shui Yan
Tang, thank you for your patience and wisdom with me during this process. To Dr.
Bryant Myers at Fuller Theological Seminary, thank you for not only providing this
research opportunity and valuable comments, but also for demonstrating a respect for
the poor that I hope I can model one day. To Dr. Terry Cooper, thank you for your
mentorship and encouragement during my time at USC. Finally, to Dr. Donald Miller
from the School of Religion, thank you for your contributions and time to this project
as well.
There are countless others who have helped me to complete this work of research as
well and I ask that the omission of their names be forgiven. I close, however, with
my most humble and genuine offer of thanks. To my God, father of the orphan and
protector of the widow, thank you for sustaining me in every way.
iv
TABLE OF CONTENTS
DEDICATION
ii
ACKNOWLEDGEMENTS
iii
LIST OF TABLES
viii
LIST OF FIGURES
ix
ABBREVIATIONS
x
ABSTRACT
CHAPTER ONE: OVERVIEW OF THE DISSERTATION
1.1 Research Objectives
1.1.1 Including the Most Vulnerable in the Community in Microfinance
1.1.2 Joining Microfinance Institutions (MFIs) with Community
Development Organizations
1.1.3 Strengthening OVC Household Capacity
1.2 World Vision International (WVI)
1.2.1 Organizational Background
1.2.2 World Vision Ethiopia (WVE)
1.2.3 Wisdom Micro Financing Institution S.C. (WISDOM)
1.2.4 Individualized Approach to Development
1.3 Research Questions
1.3.1 Microfinance for Groups Vulnerable to Social Exclusion
1.3.2 Motivating Collaboration with the MFI
1.3.3 Mitigating the Negative Effects of Loss For OVC Households
with Microfinance
1.4 Research Design
1.4.1 The PRISMA Project: Integrating Support and Care with
Microfinance
1.3.3 WVE’s Guraghe Area Development Program (ADP)
1.3.4 WVE’s Wonchi Area Development Program (ADP)
1.5 Sources of Data
1.5.1 Structured Interviews
1.5.2 OVC Household Surveys
1.5.3 Performance Data
1.6 Challenges with Cross-Cultural Research
xii
1
4
7
8
9
10
11
13
15
17
19
20
21
22
24
25
26
28
30
32
34
38
38
v
CHAPTER TWO: INCREASING FLEXIBILITY TO MITIGATE SOCIAL
EXCLUSION IN BORROWING
2.1 An Overview of Social Capital Theory in Development Practice
2.1.1 A Resurging Interest in Social Capital Theory
2.1.2 Effects of Social Capital in Economic Development
2.2 Social Capital in Microfinance Group Lending Practices
2.2.1 Conceptual Overview of Group Lending
2.2.2 Peer Selection and Screening
2.2.3 Peer Monitoring and Enforcement
2.2.4 Microfinance and Groups Vulnerable to Social Exclusion
2.3 Including and Screening Groups Vulnerable to Stigma for Microfinance:
WISDOM Micro Financing Institution
2.3.1 The Guraghe and Wonchi ADPs
2.3.2 WISDOM’s Lending Model: Screening Over Monitoring
2.3.3 Assumptions Concerning OVC Caregivers and Microfinance
2.3.4 Overcoming Assumptions: OVC Caregivers as WISDOM Clients
2.4 Flexibility through Multiple Actors
2.4.1 The Role of WVE Community Workers
2.5 Monitoring through Collaborative Relationships
4.6 Summary and Conclusion
43
45
45
47
50
51
53
54
55
56
57
58
61
66
68
70
75
82
CHAPTER THREE: INCENTIVIZING COLLABORATORS FOR
MICROFINANCE INSTITUTIONS (MFIs)
85
3.1 The “Microfinance Schism”
87
3.2 Wisdom Micro Financing Institution (WISDOM) and World Vision
Ethiopia (WVE)
90
3.2.1 Motivations to Collaborate Among WVE Community Workers
92
3.3 Factors in Motivating Collaboration Among WVE Community Workers 94
3.3.1 WISDOM’s Impact in the Community
95
3.3.2 The Role of the Collaborating Manager
98
3.3.3 The MFI as an Organization in Tension
101
3.4 Conclusion
106
CHAPTER FOUR: MITIGATING THE NEGATIVE EFFECTS OF LOSS FOR
OVCs WITH MICROFINANCE
4.1 Identifying Areas of Vulnerability for the OVC
4.1.1 Negative Effects of Orphanhood on Children
4.2 The Hope of Microfinance: Mitigating the Negative Effects of Loss
for Orphans and Vulnerable Children
3.2.1 Measuring the Impact of Microfinance for OVC Households
3.2.2 Managing Endogeneity
4.3 Discussion and Conclusion
109
110
114
116
119
123
132
vi
4.3.1 Limitations and Other Approaches
4.3.2 Contributions of this Study
133
134
CHAPTER FIVE: CONCLUSION
5.1 Research Questions
5.1.1 Microfinance for Groups Vulnerable to Social Exclusion
5.1.2 Incentives for Collaboration with the MFI
5.1.3 Mitigating the Negative Effects of Loss for OVC Households with
Microfinance
5.2 Significance of These Findings
5.3 Recommendations for Development Work
5.3.1 Integrating Microfinance with Community Development
5.3.2 Understanding the Community as More Than a Label
5.3.3 Providing Incentives to Collaborate
5.4 Concluding Remarks
136
140
140
143
BIBLIOGRAPHY
159
APPENDICES
Appendix A: Interview Protocol for Staff Workers from WISDOM and WVE
Appendix B: Interview Protocol for OVC Caregiver
Appendix C: Household Surveys – OVC Caregiver
Appendix D: Household Surveys – OVC Child
Appendix E: Household Surveys – OVC Adolescent
Appendix F: Household Asset Survey
Appendix G: Calculation of Livestock Assets
Appendix H: Descriptive Statistics – Guraghe ADP
Appendix I: Descriptive Statistics – Wonchi ADP
Appendix J: Additional Regression Results
Appendix K: OVC Psychosocial Indicator Groupings
Appendix L: Village Level Regression Results
172
172
173
174
176
177
179
180
182
184
186
188
189
145
147
149
150
154
156
157
vii
LIST OF TABLES
Table 1: Disproportionate Impact of HIV/AIDS in sub-Saharan Africa (2006) .......... 5
Table 2: Financial Services Provided by WISDOM................................................. 16
Table 3: Kebeles Included in the Sample................................................................. 31
Table 4: Summary of Major Social Capital Theories ............................................... 49
Table 5: Worker Responses on Microfinance for OVC Caregivers .......................... 63
Table 6: Percent of WISDOM Clients Caring for OVC in Guraghe......................... 67
Table 7: OVC Caregivers Participating in WISDOM by Village ............................. 71
Table 8: Comparison of Community Workers Relationship with WISDOM............ 77
Table 9: Impact of Community Worker on Client Performance............................... 80
Table 10: WVE Community Worker (CW) Motivation to Collaborate .................... 97
Table 11: Hansmann's Four-Way Categorization of Nonprofit Firms..................... 102
Table 12: A Comparative Summary of Organizational Types................................ 103
Table 13: Organizational Summary....................................................................... 104
Table 14: Impact of Microfinance on OVC Household Consumption .................... 121
Table 15: Impact of Microfinance on OVC Household Consumption .................... 126
Table 16: Impact of Microfinance on OVC Psychosocial Well-being .................... 130
viii
LIST OF FIGURES
Figure 1: Strains on the OVC Household and Community Safety Nets
10
Figure 2: Individualized Approach to Development
17
Figure 3: PRISMA Implementation Model
26
Figure 4: Edja District Map
27
Figure 5: Wonchi District Map
29
Figure 6: Factors in OVC Well-Being
112
Figure 7 : Individualized Approach to Development
150
Figure 8: Integrated Approach to Development
152
Figure 9: Relationship Between Assistance and Development
153
ix
ABBREVIATIONS
ADP
Area Development Program
AIDS
Acquired Immune Deficiency Syndrome
CBO
Community-Based Organization
CCC
Community Care Coalitions
CIP
Children in Program
CLP
Collaborative Learning Project
CW
Community Workers
FBO
Faith-Based Organization
IGA
Income Generating Activities
HIV
Human Immunodeficiency Virus
KA
Kebele (Village)
MSC
Microcredit Summit Campaign
MDG
Millennium Development Goals
MED
Microenterprise Development
MFI
Microfinance Institution
MSC
Microcredit Summit Campaign
NFPO
Not-for-profit Organization
NPO
Non-profit Organization
NGO
Non-governmental Organization
OVC
Orphans and Vulnerable Children
x
PLWH
People Living with HIV
PLWHA
People Living with HIV/AIDS
PRISMA
Promoting Rural Integration and Security through Microfinance in
Africa
WHO
World Health Organization
WISDOM
Wisdom Micro Financing Institution S. C.
WV
World Vision
WVE
World Vision Ethiopia
WVI
World Vision International
UNAIDS
Joint United Nations Programme on HIV/AIDS
UNICEF
The United Nations Children’s Fund
xi
ABSTRACT
This dissertation studies the collaborative efforts between World Vision Ethiopia
(WVE) and Wisdom Micro financing Institution, S.C. (WISDOM) to provide
caregivers of orphans and vulnerable children (OVCs) in Ethiopia with microfinance.
Because of the millions of new OVCs being created every year, many have thought
microfinance could be an effective tool to mitigate the negative effects of loss for
OVCs. Nevertheless, because OVC households can often be targets of stigmatization
and can have lower levels of social capital, the heavy dependence of microfinance on
mechanisms that use social capital can consequently work to exclude many OVC
households.
Despite the assumptions of the staff working for WVE and WISDOM concerning the
capacity of OVC caregivers to participate in microfinance, it was the collaborative
efforts between WVE community workers with WISDOM staff that resulted in the
inclusion of more OVC caregivers in WISDOM loans. Furthermore, it was the
observations of the positive impact that WISDOM was having on the community that
showed to be the biggest factor in motivating WVE community workers to
collaborate with WISDOM in their lending processes. The result was in a
comparison among newer and older borrowers of WISDOM, in addition to eating
more and attending school more often, OVCs in households who had been
participating in WISDOM for over a year fared better in psychosocial measures as
xii
well. Thus, the potential of microfinance to mitigate the negative effects of loss for
OVCs is substantial.
The irony of this particular case was that both the ability of WVE community workers
to include more OVC caregivers in WISDOM and the success of OVC caregivers in
microfinance occurred without either organization being aware of it. As a result,
collaborative efforts between the two organizations weakened in recent years while
false assumptions concerning the separate needs of different populations among the
poor were never challenged.
However, despite the two organizations’ different missions, their work was not
separate. WVE services may have actually been working to prepare many
households to participate in higher forms of economic development, such as
microfinance, demonstrating the importance of collaboration and integrated models
for development work.
xiii
CHAPTER ONE: OVERVIEW OF THE DISSERTATION
Microfinance refers to the delivery of financial services, typically micro or small
loans, to people living in poverty-stricken areas for the purposes of increasing
financial accessibility. It has become an increasingly popular intervention in reducing
poverty, especially for rural women in the Two-Thirds World. Over the past thirty
years, the microfinance industry has grown to over 10,000 microfinance institutions
(MFIs). In 1999, there were approximately 8 to 10 million households participating
in microfinance programs worldwide (Morduch, 1999, p. 1569). By 2004, there were
approximately 90 million participating, according to the Microcredit Summit
Campaign (MSC) (Daley-Harris, 2005; Carr & Tong, 2002, p. 7). Today,
microfinance continues to increase in recognition with awards, such as the Nobel
Peace Prize to Muhammad Yunus in 20061, and with initiatives, such as the UN
Millennium Development Goals (MDGs).
Still, despite the high hopes people have for microfinance, challenges continue to
exist. Among them lies the question of how to more effectively reach the most
vulnerable groups of the community. Because of its use of social capital
mechanisms, microfinance can unnecessarily shut out groups who are most
vulnerable to social stigma or exclusion. Nevertheless, the argument can be made
1
Yunus is the founder of the Grameen Bank in Bangladesh.
1
that it is the most vulnerable who not only need microfinance the most but who also
can benefit from microfinance the most.
Because pressure continues to mount for MFIs to cut costs and maintain financial
self-sufficiency, collaboration can be a viable option for many MFIs. Not only can
collaboration with other community organizations increase the scope of services
offered by MFIs, it can also increase its ability to reach more, including those who are
vulnerable to social exclusion. Collaboration with other community organizations
can also improve information and monitoring efforts. However, although there are
many potential benefits, collaborating can also add additional complexities and
challenges to an organization’s operations.
This dissertation examines how MFIs can improve their outreach efforts to include
more community members who are vulnerable to social exclusion. It explores how
collaborating with another community organization can help improve both outreach
and monitoring efforts while also posing new challenges in motivating workers at the
individual-level. Finally, it also considers how community members who are
vulnerable to social exclusion actually fare when participating in microfinance.
Using data collected from several sources over a six-week period of field research,
this dissertation studies an initiative implemented by World Vision International
2
(WVI) targetting microfinance services provided by Wisdom Micro Financing
Institution (WISDOM) for households caring for orphans and vulnerable children
(OVCs) in Ethiopia. The spread of AIDS, particularly in sub-Saharan Africa, has
resulted in millions of children who have been left orphaned or vulnerable. Once a
family member becomes chronically ill, the children in the household are also
impacted in several ways, both economically and psychosocially, that threaten their
overall well-being. In addition, OVCs and their familiy members are often targets of
social stigmaization from other community members.
The sheer number of OVCs in a region has spread family networks thin and
weakened household capacities and community safety nets. The success and
popularity of microfinance to alleviate poverty and empower individuals has led to
hopes that it can be used as a tool to mitigate the negative effects of the HIV/AIDS
epidemic and other chronic diseases. In particular, many hope that microfinance can
be used to strengthen household capacities and thereby improve OVC well-being.
Nevertheless, because OVCs are also vulnerable to social stigmatization, reaching
OVC caregivers for microfinance programs can be difficult. This study explores
some of the successes and challenges that WISDOM faced in providing microfinance
to OVC caregivers and considers the validity of the hopes in microfinance to mitigate
the negative effects of death and illness for OVCs in Ethiopia.
3
1.1
Research Objectives
Although different forms of microlending have existed for quite some time, one of the
earliest best-known accounts is how Muhammad Yunus founded the Grameen Bank
in Bangladesh in the 1970s. It was on a visitation to a small village in Bangladesh
that Professor Muhammad Yunus would meet a woman who made her wages making
bamboo stools. She told Yunus of how high interest rates on the money she borrowed
to make her stools would leave her with only a penny profit margin at the end of each
day. The woman’s story would be the catalyst behind a pilot research project Yunus
would conduct in 1976 to give “micro-loans” or small loans to 42 basket-weavers
based on a policy of “joint liability” in the hopes of greater financial empowerment
(Yunus, 2003). In 1983, Yunus founded the Grameen Bank on the same concepts of
his research project. The success of the Grameen Bank in Bangladesh has been noted
worldwide with approximately 1,000 branches and 14,000 staff servicing
approximately 2 million borrowers with a portfolio of $260 million by 1997
(Schreiner, 2003).
MFIs are driven by the notion of granting small loans to poor people who lack
traditional forms of collateral or access to formal financial institutions. They are
aimed at helping people to improve their lives through financial empowerment
resulting in the alleviation of poverty. Microfinance seeks to do this usually by
either: 1) reducing the vulnerabilities resulting from poverty with services, such as
4
savings or insurance or 2) increasing opportunities for enterprise development
through services such as credit or training (Fikkert, 2003, p. 5). Often services are
dispersed through group lending models. However, individual lending models are not
uncommon either.
Table 1: Disproportionate Impact of HIV/AIDS in sub-Saharan Africa (2006)
PLWHA (prevalence)
24.7 million [21.8-27.7 million]
Newly infected in 2006 (incidence)
2.8 million [2.4-3.2 million]
AIDS deaths
2.1 million
Children orphaned due to AIDS
11 million
Source: UNAIDS, 2006, p. 10.
Recent interest in microfinance and its successes has led many to ask the question of
whether it can play a structural role in mitigating the negative effects of HIV/AIDS
on a household (Wright, 2000; Pronyk et al., 2005, p. 28). By 2006, while making up
just over 10% of the world’s total population, sub-Saharan Africa accounted for 63%
(24.7 million) of the world’s total HIV/AIDS population. Similarly, 72% of all AIDS
deaths worldwide occurred in this region at 2.1 million (UNAIDS, 2006, p. 10). See
Table 1 for summary of HIV/AIDS statistics in sub-Saharan Africa. Prevalence rates
do appear to be stabilizing within the region with some countries even experiencing
5
decreasing rates of HIV/AIDS. Nevertheless, large numbers of deaths have resulted
in millions of orphaned children throughout the region.
OVCs are formally defined as: 1) children who have either lost one or both parents
for some reason, 2) children who are living with a PLWHA, 3) children who have
parents that are chronically ill, or 4) children who are identified as vulnerable by
some community standard. Accurate numbers of OVCs are difficult to come by due
to inadequacies in survey language in including children who are living with
relatives, variations in prevalence levels, stigma attached to orphan status, girls who
themselves become parents, and migration activity. In addition, children whose
parents are positively infected or children who are living with foster families often go
unaccounted for in such numbers (Foster & Williamson, 2000, p. S276).
However, despite such difficulties, estimates have been made. In 2001, the USAID in
conjunction with UNICEF and UNAIDS conducted a study estimating numbers of
OVCs. The study estimated that 80% (11 million) of the 14 million children
worldwide who have lost one or both parents to AIDS lived in sub-Saharan Africa
(USAID et al., 2002, 2004; Shetty & Powell, 2003: 25; Levine, Foster, & Williamson,
2005, p. 5). Estimates also expected that the number of children orphaned due to
AIDS in this region would reach 20 million by 2010 and remain exceptionally high
6
until 2030 (USAID, UNICEF, & UNAIDS, 2002, 2004; Foster & Williamson, 2000:
p. S275).
Many fear that the impact of chronic diseases, such as HIV/AIDS, and the high
numbers of OVCs in sub-Saharan Africa threaten to reverse much of the development
gains over the past decades (USAID, 2001, p. 2). Thus, in the face of these great
challenges, the hope of what microfinance can achieve has captured the minds and
imaginations of many development workers working to mitigate the negative effects
death for households in the Two-Thirds World. Microfinance has been shown to
produce economic benefits that can lead to a greater sense of autonomy and resilience
for clients (Cheston & Kuhn, 2002; Pronyk et al., 2005, p. 29). It has also been
linked to greater health benefits (Khandker, 1998; Pronyk et al., 2005, p. 29). Many
have even suggested that its outcomes can address the structural factors, such as
poverty and gender inequalities, that often contribute to the spread and impact of
HIV/AIDS (Parker, Easton, & Klein, 2000; Pronyk et al., 2005, p. 28). Yet, if
microfinance is truly to be used to bring widespread change for vulnerable groups,
such as OVCs, several challenges remain.
1.1.1 Including the Most Vulnerable in the Community in Microfinance
Many scholars have criticized microfinance for unnecessarily excluding community
members from borrowing. Common barriers can be excessive pressure from loan
7
officers or discrimination from other community members. The use of social capital
mechanisms, such as group lending, in microfinance can have unintended effects of
preventing groups vulnerable to social stigma and exclusion. Because stigmatization
can be one of the negative effects for OVCs when they lose a parent or family
member, targetting OVC caregivers for microfinance can be difficult. Nevertheless,
scholars have suggested that increasing flexibility in the processes and products of
MFIs, can work to include more of the vulnerable population in microloans.
1.1.2 Joining Microfinance Institutions (MFIs) with Community
Development Organizations
Increasing pressures to be financially self-sufficient have pushed many MFIs to scale
back the products and services they can offer community members. Social services,
such as health and education programs, have become less common for MFIs because
of the extra costs to the organization. Yet the organized structure of group lending
can also be beneficial in providing other social services to community members.
Therefore, collaborating with other social service and development organizations has
become one way for both organizations to benefit from the other to help community
members. Furthermore, collaborating with community organizations may provide
MFIs with the flexibility they need in their processes to include more community
members in their services, especially those that are hard for MFIs to reach
themselves. Thus, collaboration can provide positive outcomes for both the
8
community members and the organizations themselves. However, working together
can often add a layer of complexity that can become burdensome as well.
1.1.3 Strengthening OVC Household Capacity
Studies in several countries have shown that income levels were approximately 20 to
30% lower in orphan households compared to non-orphan. A lack of pre-planning
and a lack of widows rights can also lead to land-grabbing and increase a household’s
poverty after a family member’s death (Foster & Williamson, 2000, p. S282).
In the midst of all of this, the importance of the extended family network as a safety
net for OVC has become apparent. Aunts, uncles, and grandparents often act as
caregivers for an orphan after a parent has passed away. As a result, interventions
focused on working through households rather than resorting to institutionalization
are often recommended (Foster & Williamson, 2000). A USAID report notes that the
"institutionalization of children often separates them from families and communities
and often delays healthy childhood development" (USAID, 2001, p. 9). Thus, the
importance of "strengthening the safety nets of families to protect and care for
OVC[s]" has been emphasized by policy advisors from the UNAIDS, UNICEF, and
USAID (2004, p. 5).
Nevertheless, the sheer number of OVCs has saturated many family safety nets and
more and more children have begun to fall through the cracks. Figure 1 shows the
9
various strains on an OVC households that can result in children being exposed to
more risks in the community. Indicators of the saturation of family safety nets can
include: high numbers of elderly caregivers, child-headed households, frequent
sibling dispersal, and migration (Foster & Williamson, 2000, p. 7-42). Therefore,
finding mechanisms and interventions that will strengthen household capacities is
vital for the well-being of millions of OVCs.
Figure 1: Strains on the OVC Household and Community Safety Nets
1.2 World Vision International (WVI)
This dissertation focuses on an initiative implemented by WVI to provide
microfinance services to OVC caregivers in Ethiopia. The following section provides
a general overview of the global operations of WVI as well as its national operations
10
in Ethiopia managed by World Vision Ethiopia (WVE) and its microfinance
subsidiary, WISDOM.
1.2.1 Organizational Background
World Vision is a Christian relief and development organization that was started in
1950 by Reverend Bob Pierce. Its first child sponsorship program began in 1953 in
response to the hundreds of thousands of orphans created by the Korean War. Since
that time, WVI has extended its operations to include regions throughout Asia, Latin
America, Africa, the Middle East, and Eastern Europe.
WVI’s primary source of funding is its child sponsorship program where donors are
linked to specific children living in poverty throughout the world and sponsor them
through a monthly donation. Children who are sponsored are called Children in
Program (CIPs) and receive health, education, and food assistance through the
sponsorship program. During the 1970s, WVI incorporated a community
development model into their program activities in the form of Area Deveopment
Programs (ADPs). World Vision ADPs refer to both the community development
programs and the boundaries within which they operate. World Vision ADP staff
work with community members to oversee and implement these development
activities in the community. Typically, in addition to managing the child sponsorship
program, ADPs address community needs caused by poverty and work to improve
“access to clean water, primary health care, nutritious food, basic education, and
11
economic development” over a period of time in areas often covering entire districts
(Tegarden, 2006).
World Vision ADP staff work with community members to oversee and implement
these development activities in the community. The ADP staff is organized
according to a generic structure with the ADP context given consideration. The
generic structure is as follows: 1) ADP Manager, 2) Program Office, 3) Program
Facilitators, and 4) Program Support Staff.
In addition, ADP managers also work with World Vision community workers (CWs)
who work to identify the neediest children in a community and monitor their progress
through the child sponsorship program. Typically, a World Vision community
worker monitors the CIPs in the sub-village or village in which he or she lives. Often
a World Vision community worker generally oversees an area of approximately 500
households and may be assigned a sub-village or an entire village or two depending
on the population density of the area.
WVI also has several other established divisions allocated to different activities,
including emergency relief, HIV/AIDS care and prevention, and microfinance
provision. WVI‘s emergency relief division was created in the 1970s to respond to
various humanitarian crises throughout the globe. Its operations, however, are
12
outside the scope of this dissertation. In the 1990s, WVI began responding to the
HIV/AIDS crisis through various programs designed to aid orphans and their
caregivers. In 2000, WVI formally established the Hope Initiative designed to work
towards prevention, care, and advocacy for HIV/AIDS related issues. Finally, also
during the 1990s, WVI began implementing microfinance programming in its ADPs
and subsequently established a microfinance subsidiary, called VisionFund
International, in 2003 to manage all of its microfinance operations globally.
1.2.2 World Vision Ethiopia (WVE)
World Vision Ethiopia (WVE) has operated as WVI’s national program office for
Ethiopia since 1971. Its national program office is located in Addis Ababa in
addition to several program offices and ADPs throughout the country. WVE’s
mission is stated as, “working with the world’s most vulnerable people…to overcome
poverty and injustice” (World Vision International, n.d).
The organizational structure of WVE is a three-tiered structure with the following
tiers: 1) National Office, 2) Program Offices, and 3) ADPs. The Program Offices and
ADPs implement more of the frontline, on the ground, sort of work while the
National Office works to give strong technical backing to the other two. Currently,
WVE works throughout the country in seven regional states and has over 46 ADPs.
13
WVE’s HIV/AIDS care and prevention program, established by WVI’s Hope
Initiative, has five areas of intervention: 1) prevention, 2) care and support for those
affected by HIV/AIDS, including OVCs, 3) partnership with faith-based
organizations (FBOs), 4) staff capacity building, and 5) advocacy for those affected
and marginalized by HIV/AIDS. Although different strategies are employed, the
program’s main approach involves mobilizing and training Community Care
Coalitions (CCCs).
CCCs were designed so that each village would have one CCC working to identify
the needs of OVCs and mobilize resources for them. Each CCC consists of 10 to 12
community members, including the village chairman. Typically other CCC members
include: members from local churches or other faith-based organizations, government
agencies, businesses, non-governmental organizations or community-based
organizations, or other community stakeholders (WVI, 2006, p. 7). Efforts are aimed
at educating community members on how diseases, such as HIV/AIDS, are
transmitted and at promoting testing for community members and visitors. The WVE
ADP HIV/AIDS staff member meets with each CCC once a month to discuss issues
and conduct training sessions.
14
1.2.3 Wisdom Micro Financing Institution S.C. (WISDOM)
Wisdom Micro Financing Institution S.C. (WISDOM) originated as WVE began
offering micro credit operations along with its relief operations in the 1990s. In 1996,
the Ethiopian government passed Proclamation no. 40, which required all
microfinance operations to be conducted only by registered and regulated institutions
that run as share companies and are wholly owned by Ethiopian nationals (WISDOM,
2006, p. 3). Thus, in 1998, WVE established WISDOM as a separate and
independent for-profit entity where all profits are reinvested in WISDOM’s loan fund
(Tegarden, 2006, p. 5). By 2006, WISDOM was the fifth largest MFI in the country
with a loan portfolio of approximately $3 million and an active loan clientele of
29,668 by December 2005 (WISDOM, 2006, p. 3).
WISDOM’s mission is stated as “improving the economic and social well being of
the productive poor through providing quality financial services that are viable in the
market place” (WISDOM, 2006, p. 4). It works to provide micro loans and savings to
WVE’s ADPs and some other towns2 of Ethiopia primarily in Northern, Southern,
and Central Ethiopia. By 2006, 65% of WISDOM clients were from WV Area
Development Programs (ADPs) with WISDOM operating in 25 of the existing 45
ADPs in Ethiopia (WISDOM, 2006, p. 3-4).
2
Sodo, Arbaminch, Adama, Ambo, Dessie, Fiche, Debreberhan, Hosana, and Durame.
15
WISDOM runs on a “community banking” model involving: “the creation of groups
with a committee, step credit, mandatory savings, monthly interest payments and a
single balloon payment of principal” (WISDOM, 2006, p. 3). Each community bank
is a self-selected group of 20 to 30 women and follows a policy of joint liability.
Solidarity loans have also been introduced as business loans to a small group of
traders (WISDOM, 2006, p. 3).
Table 2 gives an overview of WISDOM financial products and services, which
include: 1) business, agriculture, and consumer loans, 2) savings, and 3) loan
insurance (Tegarden, 2006, p. 5). See Table 2 below. Its portfolio remains fairly
diversified with approximately 12,000 clients in agriculture, 10,00 clients in trade,
and 6,000 clients in service. 48% of its clients are female reflecting the large share of
agricultural loans (WISDOM, 2006, p. 5).
Table 2: Financial Services Provided by WISDOM
Loan Products
Agricultural Loan, Agri-Business, Business
Loan, Enterprise Loan, Consumption Loan,
Individual Loan.
Saving Products
Compulsory Savings, Voluntary Savings.
Insurance Products
Micro-Insurance.
Source: WISDOM, 2006, p. 8.
16
1.2.4 Individualized Approach to Development
Thus, WVE’s national programming strategy involves three separate programs
designed to address specific problems and issues. Figure 2 below shows an overview
of WVE’s national programming strategy with regard to its child sponsorship
program, microfinance lending, and HIV/AIDS care operations.
Figure 2: Individualized Approach to Development
The main program operating in the WVE ADPs is the child sponsorship program,
which began in 1971 and accounts for approximately half of WVE’s program
17
funding. Sponsors donate on a continual basis to provide social services to CIPs,
which include services that benefit the community at large as well. While CIPs
receive direct assistance for food, education, and health, programs working to
improve water, education, and health benefit others in the community as well. WVE
community workers then work to monitor the progress of the CIPs in the community.
The microlending operations of WVE are managed by its affiliated but separate MFI,
WISDOM, which began in 1996. While WISDOM receives grants to fund its loans
from major donors through WVE, it operates with a separate board from WVE’s. In
addition, WISDOM recovers the operational costs of lending through interest
payments collected from its clients. WISDOM staff (including its branch manager,
loan officers, and cashier) work throughout the community mobilizing and
monitoring lending groups.
Finally, WVE’s HIV/AIDS care and prevention program, which began in 2000, is
facilitated predominantly by WVE ADP HIV/AIDS staff person. The WVE ADP
HIV/AIDS staff person works throughout the community to mobilize and monitor
CCCs who in turn work to increase awareness and resources for OVCs and people
living with HIV/AIDS (PLWHA).
18
1.3
Research Questions
This study examines an initiative implemented in WVI in Ethiopia designed to target
microfinance services for households caring for OVCs. The original design of the
initiative included plans to integrate microfinance services provided by VisionFund
International MFIs with HIV/AIDS care provided by WVI ADPs in East Africa. This
study focuses on the implementation of this initiative in Ethiopia. More specifically,
it explores how WISDOM’s collaboration with WVE resulted in greater flexibility
that allowed more OVC caregivers to participate in and succeed with microloans.
The study notes particular areas of challenge and benefit while addressing the
question of how well microfinance can mitigate the negative effects of death for a
child.
Three separate empirical essays address the following main questions:
1. How can MFIs work to exclude or include community members who are
vulnerable to social exclusion in borrowing? How can MFIs increase
flexibility in their processes to include more who are vulnerable in their
services?
19
2. How can MFIs provide incentives for community organization workers to
collaborate with them? What sorts of challenges do MFIs face in motivating
others to collaborate with them?
3. How does participation in microfinance impact the capacity of households
caring for OVCs? How well does it mitigate the negative effects of losing a
parent or having a parent who is chronically ill for the OVC?
The first and second questions focus on the implementation of microfinance and
service delivery. They concentrate on the processes of outreach and collaboration
and assess collaboration outcomes. The third question seeks to address the impact of
microfinance on households who care for OVCs. In particular, it seeks to examine
impact not only by examining caregiver and household outcomes, but also to observe
the impact on the OVC.
1.3.1 Microfinance for Groups Vulnerable to Social Exclusion
How can MFIs work to exclude or include community members who are
vulnerable to social exclusion in borrowing? How can MFIs increase flexibility
in their processes to include more who are vulnerable in their services?
20
The first empirical essay addresses how effective WISDOM was in including OVC
caregivers in their services. WISDOM’s lending model is presented and compared
with others for strengths and weaknesses in screening and monitoring processes. In
addition, data collected during field research in Ethiopia is used to identify what
barriers existed for OVC caregivers in participating in WISDOM loans. The essay
makes the case that collaborating with community workers for community
development organizations can provide a MFI with more flexibility in its loan
processes and improve its ability to reach marginalized social groups. In essence,
collaborating with another community-based organization can substitute a community
worker’s social capital for a borrower’s social capital. Analysis of the impact of
positive collaborative relationships between the two organizations on client outcomes
is considered using household survey data from OVC caregivers. Thus, although
collaboration may present some additional challenges, there is much potential benefit
for the MFI.
1.3.2 Motivating Collaboration with the MFI
How can MFIs provide incentives for community organization workers to
collaborate with them? What sorts of challenges do MFIs face in motivating
others to collaborate with them?
While the first empirical essay is concerned with the outcomes of collaboration for
the MFI and its clients, the second empirical essay considers the process of
21
collaboration for the MFI. More specifically, motivation among community workers
collaborating with MFIs is examined to determine what sorts of factors are
influential. Because community workers from other organizations can play
influential roles in an MFI’s lending processes, it is important for the MFI to reflect
on the various factors that can either increase or decrease this motivation.
A case study focused on the collaborative efforts to organize and monitor WISDOM
community banks in the Guraghe and Wonchi ADPs in Ethiopia is presented.
Qualitative data collected from interviews with program staff from both WISDOM
and WVE are used to assess what sort of interests and challenges existed in
motivating WVE community workers at the individual-level to work with WISDOM
staff. In addition to the impact that WISDOM has on the community, supervising
managers and organizational differences between the two organizations are explored
as possible factors in individual-level motivations among the WVE community
workers.
1.3.3
Mitigating the Negative Effects of Loss For OVC Households with
Microfinance
How does participation in microfinance impact the capacity of households
caring for OVCs? How well does it mitigate the negative effects of losing a
parent or having a parent who is chronically ill for the OVC?
22
The third empirical essay uses baseline data from a quasi-experimental research
design to compare outcomes among four groups of OVC caregivers. Baseline data
includes surveys conducted on caregivers taking care of OVCs, adolescent OVCs
(ages 13 to 19 yrs old), and child OVCs (ages 6 to 12 yrs old).
A cross-sectional analysis of outcomes between OVC caregivers who have
participated in microfinance for over a year with OVC caregivers who have
participated in microfinance for less than a year is conducted. Comparing the impacts
on household capacity addresses the question of whether microfinance has the
intended effect of strengthening household capacity. In addition, comparing the
impacts on OVC well-being will indicate whether or not changes in household
capacity also translate positively to changes in OVC well-being.
Head of the households were surveyed on household capacity using an instrument
again based on an adapted version of the survey instruments developed by
Community REACH3 (Chatterji et al., 2005). Indicators for household capacity
include: 1) household financial security, 2) socioeconomic status, 3) social network,
4) household nutrition and 5) household health status. Indicators of OVC well-being
included both economic and psychological factors from children 6 to 12 years old and
adolescents 13 to 19 years old based on an adapted version of the survey instruments
3
http://www.pactworld.org/reach/OVCResources/
23
developed by Community REACH (USAID, 2001, p. 6; Chatterji et al., 2005).
Children were surveyed on emotional well-being. Adolescents were surveyed on: 1)
material well-being or shelter, 2) school attendence and education, 3) nutrition, and 4)
psychosocial well-being (Chatterji et al., 2005; Sengendo & Nambi, 1997, p. 109).
1.4
Research Design
The study uses survey data that was collected during six-weeks of field research in
Ethiopia in August 2007. The data was collected during the first round of field
research for a quasi-experimental pre-test/post-test research project. The research
design of this original project included one treatment group and two control groups.
The treatment group included villages where WISDOM microfinance services were
coordinated with WVE ADP HIV/AIDS staff. The first control included villages
where WISDOM microfinance services were uncoordinated with WVE ADP
HIV/AIDS staff. Lastly, the second control group included villages where WVE
ADPs existed but no microfinance was available. Again, the first round of data
collection took place in August 2007. The second round of data collection is
scheduled for a subsequent trip in July 2009.
The sample for this study contains households selected from kebeles (villages) in the
Guraghe and Wonchi ADPs in Ethiopia where World Vision’s Promoting Rural
Integration and Security through Microfinance in Africa (PRISMA) project had been
24
implemented. At the time of baseline data collection, the PRISMA project was still in
its beginning stages of implementation in Ethiopia.
1.4.1 The PRISMA Project: Integrating Support and Care with
Microfinance
WVI’s PRISMA initiative began in February 2006 and is scheduled to run until
December 2009. The initiative is currently being implemented in World Vision
ADPs in East Africa, including Ethiopia, Kenya, Rwanda, Tanzania, and Uganda.
The stated purpose of PRISMA is to “improve the lives of very poor women in rural
East Africa by combining economic development assistance with improved access to
HIV/AIDS care, mitigation services, and education” (Tegarden, 2006).
PRISMA was designed to achieve its program objectives by using a coordinated
approach of service delivery between WVI’s ADP staff and VisionFund’s
microfinance staff. The approach involved loan officers from VisionFund MFIs
coordinating closely with WVI ADP HIV/AIDS care staff. Together the two groups
of staff would work to improve access to appropriate financial services for rural
clients, in particular for those who are caring for OVCs. The integrated approach to
OVC care in Ethiopia involved the partnership of WVE and WISDOM, its affiliated
MFI. Figure 3 shows the implementation model below.
25
Figure 3: PRISMA Implementation Model
Source: Tegarden, 2006.
1.3.3 WVE’s Guraghe Area Development Program (ADP)
The Guraghe ADP is located in the Guraghe zone, which has an area of 5932 km2 and
is comprised of 15 woredas and 421 KAs. The area is divided into three zones
according to the terrain of the land. The highland comprises about 45.1% of the land,
the medium altitude comprises about 49.2% of the land, and the low land comprises
about 5.7% of the land. Its total population is approximately 1.8 million people.
26
Figure 4: Edja District Map
Notes: Provided by WVE Guraghe ADP staff.
WVE’s Guraghe ADP began in 2001 and operates in the Guraghe zone in the Edja
and Muhurna Aklil districts covering 56 villages. It is approximately 195 km
southwest of Addis Ababa and has an approximate area of 767 km2 with 40.7% of it
being cultivated land. In addition, it receives an annual rainfall of approximately
1000 to 1250 mm. A map of the Edja woreda is provided in Figure 4. The towns of
Agena, Cheza, and Tekle Haimanot make up the Guraghe ADP. Its total population
is approximately 233,00 people with 42,000 households as of 2007.
27
Because there is much movement to and from Guraghe, the area is more vulnerable to
diseases prevalent in urban areas as well. There are two Ethiopia holidays (Meskali
and Arefa) that bring large numbers of people back to their locality in Guraghe from
urban areas. In addition, many caregivers spend longer periods of time in urban areas
away from their homes for business or employment reasons. WVE HIV/AIDS focal
staff member for the Guraghe ADP estimated a prevalence rate of 4-6% for area. By
August 2007, the Guraghe ADP had identified 9,798 OVCs in the two districts it
oversees.
1.3.4 WVE’s Wonchi Area Development Program (ADP)
The Wonchi ADP is located in the Shoa zone and is about 123 km southwest of Addis
Ababa. It has 26 villages and is part of the Oromia region where the local languages
are Oromo and Amharic. The area is divided into two zones according to the terrain
of the land. The highland comprises about 44% of the ADP and the midland
comprises about 56% of the ADP. It has an altitude of 1800 to 3387 masi and typical
rainfalls occur between the months of June and September for annual levels of 1200
to 1420 mm.
28
Figure 5: Wonchi District Map
Notes: Provided by WVE Guraghe ADP staff.
The Wonchi ADP has an approximate area of 475 km2 with 96.8% of its population
living in rural areas where subsistence agriculture is the predominant source of
livelihood. The rest of the population lives in townships and peri-urban areas. There
are approximately 110,000 people and 22,000 households who are direct beneficiaries
of the ADP. Figure 5 shows a map of the Wonchi ADP.
29
Similar to Guraghe, Wonchi also has large numbers of caregivers who travel back and
forth to urban areas for employment. Often caregivers will have multiple wives or
partners and can transmit diseases to and from Wonchi. By August 2007, there were
nearly 5,000 OVCs identified and registered by WVE CCC’s in the district. The
WVE HIV/AIDS focal staff member for Wonchi was unsure of prevalence rates but
estimated them to be similar to Guraghe.
1.5
Sources of Data
There are three main sources of data that were used for this dissertation. Data was
collected over a six-week period of field research in Ethiopia in August and
September of 2007. The first source of data is structured interviews conducted with
WVE ADP staff, WISDOM staff, WVE community workers, and OVC caregivers.
The second source of data is survey data from OVC caregivers, OVC adolescents,
and OVC children aimed at measuring household outcomes and OVC well-being.
Finally, the third source of data is performance data for relevant WISDOM branches
and clients provided by WISDOM. Additionally, non-participant observation and
organizational literature, such as business plans and program design documents
provided by WVE and WISDOM, were used in this study.
30
Selection of Kebeles (KAs)
The two ADPs included in this study were selected in consultation with WVE and
WISDOM managers based on ADPs that were the furthest along in implementing the
PRISMA project. In these ADPs, the PRISMA project had taken the form of
Credit+Edu programs. Credit+Edu programs combined microfinance services with
HIV/AIDS awareness education for WISDOM community banks. The kebeles for
the treatment group were then selected based on where the Credit+Edu program had
already been implemented in at least one community bank. The kebeles for the first
control group (uncoordinated services) were then selected based on areas that had
both high numbers of OVC caregivers and high numbers of WISDOM clients.
Finally, the kebeles for the second control group (no microfinance) were selected
based on areas that had high numbers of OVC caregivers and were accessible for
travel. Table 3 shows the kebeles that were included in this study in both ADPs.
Table 3: Kebeles Included in the Sample
Treatment Group:
Coordinated
Control Group #1:
Uncoordinated
Control Group #2:
No Microfinance
Sheremo
Shebraden
Agena
Wasamer
Yegobet
Wadiye
Desene
Ambagenet
Guraghe
Wonchi
Belabela
Dimtu
Fite
ADP
31
Selection of OVC Caregivers
Both WISDOM and WVE staff members were unaware of which households were
both caring for OVCs and participating in WISDOM services at the time of data
collection. As a result, coordination between loan officers and HIV/AIDS staff was
needed in order to obtain a main list of possible subjects for the study. First,
WISDOM loan officers provided client lists from the kebeles selected for the study.
Second, WVE community workers provided names of caregivers who were caring for
OVCs for the same kebeles. Next, the two lists were cross-referenced to identify
which caregivers met both conditions of caring for OVCs and participating in
WISDOM services. All households with caregivers that met these two conditions
were included in the sample as client subjects for the treatment group (coordinated)
and the first control group (uncoordinated). Non-client caregivers were then selected
randomly from the lists of OVC caregivers that were provided by WVE community
workers. These households were used as non-client subjects for the treatment group,
first control group, and second control group.
1.5.1 Structured Interviews
Structured interviews with WVE ADP staff, WISDOM staff, and WVE community
workers were based on an adapted version of a protocol used by USC’s Collaborative
Learning Project (CLP). The original protocol was designed to examine collaborative
32
relationships and networks between neighborhood-based groups and local
government agencies.
The protocol was adapted to address the areas of collaboration relevant to
microfinance and support for OVC households. The instrument was contextualized to
learn about the subject’s participation and perceptions of the collaborative efforts
between two organizations, WVE and WISDOM. In particular, the protocol
addressed the following topics: 1) the background and role of the subject within the
organization, 2) the history of collaboration between the two organizations, 3)
information and resources shared between the two organizations, and 4) perceptions
concerning the other organization. See Appendix A for interview protocols.
Among WVE ADP staff, ADP managers and HIV/AIDS staff were interviewed at
both ADPs. Among WISDOM staff, branch managers and loan officers were
interviewed as available. WVE community workers from each kebeles included in
the study were interviewed with the exception of Agena, Desene, and Wadiye.
Additionally, interviews were conducted with OVC caregivers. Both those that were
participating in microfinance as well as those that were not were interviewed.
Interviews were in-depth open-ended questions aimed at learning about the following:
1) the subject’s experiences after the OVC lost one or both parents, 2) the subject’s
33
experiences with WVE CCC’s, and 3) the subject’s experiences with WISDOM
(when applicable). Subjects were selected in consultation with WISDOM staff and
WVE community workers and were based on interest in the case and availability of
the subject.
A total of 26 interviews were conducted. 13 were conducted in Guraghe and 13 were
conducted in Wonchi. 4 ADP staff were interviewed, 5 WISDOM staff were
interviewed, and 7 WVE community workers were interviewed. A total of 10 OVC
caregivers were interviewed. 6 of these 10 were WISDOM clients and 4 were nonclients. 3 of the 26 interviews were not included in the analysis of the case due to
either staff having just transitioned into a position or because clients were not
currently caring for OVCs. I conducted all interviewers in the presence of a
translator(s) provided by World Vision and an assistant who was taking notes.
1.5.2 OVC Household Surveys
OVC household surveys were based on an instrument developed by Community
REACH for a USAID study examining OVC well-being in Rwanda and Zambia
(USAID, 2001, p. 6; Chatterji et al., 2005). Four of the original five parts of the
instrument were used and adapted for the purposes of this study in Ethiopia. The four
parts of the survey included: 1) primary caregiver questionnaire, 2) primary caregiver
34
questionnaire regarding their 6 to 12 year old OVC child, 3) OVC questionnaire for 6
to 12 year olds, and 4) OVC questionnaire for 13 to 19 year olds.
Surveys were adapted to context and included questions divided into the following
sections: 1) household financial security, 2) socioeconomic status, 3) social network,
4) income generating activities (IGA), and 5) food and health support. Children and
adolescents were also asked questions concerning psychosocial well-being around
topics such as worry/stress, overburden/responsibility, locus of control, and selfesteem.
Surveys were translated from English to Amharic in Ethiopia. Two sets of 9
enumerators were hired for data collection and participated in trainings discussing
data instruments and collection procedures. Instruments were also discussed with
enumerators after pilot testing and revised where needed. Enumerators were required
to have basic secondary school education and to have basic reading and writing skills.
Each set of enumerators was supervised as one group.
Each morning, the enumerators were given the names and kebele of the subjects to be
interviewed. Enumerators asked either the kebele chairman or another community
leaders to identify the specific locations of each household. Caregivers of each
household were then asked whether they wanted to participate in the survey. If the
35
caregiver was not home, a second and third attempt was made a few days later. If no
contact could be made with the caregiver after the third attempt, the subject was
removed from the sample. In cases where there was more than one OVC in each age
group, a child OVC and an adolescent OVC were selected at random for the study. In
cases where there were only child OVCs or only adolescent OVCs, only one OVC
was interviewed.
Upon consent, the primary caregiver was interviewed and first asked to provide a
roster of the entire household. All children ages 6 to 12 and 13 to 19 who spent the
previous night at the house were to be included in the roster. In households where
there was an OVC who was between the ages of 6 to 12, the primary caretaker was
then asked additional questions regarding their perceptions of the child’s material
well-being, education, nutrition, health, and psychological well-being.
When present, OVCs from ages 6 to 12 were then asked a series of questions
concerning their emotional well-being. Similarly, when present, OVCs from ages 13
to 19 were asked a series of questions concerning their material well-being,
education, nutrition, health support, and psychological well-being.
In addition, enumerators were also sent out in February 2008 to the households
included in the sample. Data was collected concerning livestock assets in August
36
2007 and in February 2008. In addition, data was also collected concerning the
number of children participating in WVE’s child sponsorship program as CIPs. See
the Appendix C through F for survey protocols.
The original research design called for a total of 350 OVC households. From each of
the two ADPs, 70 OVC households (35 clients and 35 non-clients) were to be
sampled from coordinated villages, 70 OVC households (35 clients and 35 nonclients) were to be sampled from uncoordinated villages, and 35 OVC households (35
non-clients) were to be sampled from villages with no microfinance.
A total of 316 households (316 caregivers, 265 adolescents, and 184 children) were
interviewed. The shortage of households was due to the fact that there were only a
limited number of OVC caregivers participating in WISDOM at the time of data
collection. 119 WISDOM clients were interviewed in addition to 197 non-clients.
169 households were interviewed from the Guraghe ADP (169 caregivers, 141
adolescents, and 98 children). 147 households were interviewed from the Wonchi
ADP (147 caregivers, 124 adolescents, 86 children).
Data was entered in Ethiopia using SPSS statistical software. After data was checked
for no responses and inconsistencies, the resulting sample was 279 OVC households
(279 caregiver surveys, 173 adolescent surveys, and 221 child surveys). 162
37
households were from Guraghe (162 caregiver surveys, 95 adolescent surveys, and
119 child surveys) and 117 households were from Wonchi (117 caregiver surveys, 79
adolescent surveys, and 92 child surveys).
1.5.3 Performance Data
Performance data was obtained from WISDOM records. WISDOM branch managers
were given a list of clients included in the sample. Performance data was gathered
concerning the date and amount of the client’s first loan, the amount of the client’s
current loan, and delinquency status. In addition, general performance measures for
kebeles and ADPs included in the study were collected.
1.6
Challenges with Cross-Cultural Research
Finally, I conclude this chapter with some reflections on the challenges associated
with conducting this sort of research. While being somewhat accustomed to crossing
back and forth between cultures as the daughter of Korean immigrants living in the
United States, I found there to be two major cultural barriers, which created several
challenges, in conducting this research.
First was the barrier of crossing into agricultural life in the Two-Thirds World as a
westerner. In spite of the significant amount of time dedicated to preparing for this
38
field research, many of the decisions concerning the research design still needed to be
made or adapted on the ground where more information was available. The selection
of ADPs, villages, and households for this study all needed to be made during field
research because of the limited amount of information known prior to arriving in
Addis Ababa. While this may not be unusual when doing field research, the
challenges in making these decisions were likely more difficult because of these
barriers. Inconsistencies in what I was told prior to the field research while in the
United States and what I found out once I was on the field in Ethiopia were frequent
as well.
One example of this was concerning the PRISMA project itself. While initial
research plans were designed to examine the PRISMA initiative, the progress and
implementation of the project on the ground varied fairly significantly from what I
was expecting prior to field research. These differences made it necessary to adapt
the research design to include more interviews in the community while focusing less
on the implementation of PRISMA itself as originally designed.
In addition, because travel is more limited in rural areas, especially during rainy
season, it was too costly to attempt to visit several regions for observation prior to
selecting ADPs. Instead, consultations with WVE staff who were familiar with the
regions and the programs were greatly considered in the selection process of the
39
ADPs and villages. Furthermore, because the two organizations themselves were
unaware of which WISDOM clients were also OVC caregivers, a strategy needed to
be devised on the ground concerning how to identify these clients. Additionally,
because rural areas are much less dense and accessible then urban areas, the traveling
time to walk to each household being surveyed provided challenges as well. As the
ADPs were between 500-700 km2 (approximately half the size of the city of Los
Angeles), with limited roads and a great deal of walking, each enumerator was able to
conduct about 3 surveys per day.
What quickly became apparent during field research was the importance of
identifying where the major sources of authority and information were in both the
community and the two organizations, especially because they were often different
from what a westerner might be accustomed. For example, once the households that
needed to be included in this study were identified, the next step was figuring out how
to find out where they lived since there were no addresses written down. Thus, the
decision was made to instruct the enumerators to solicit the aid of the village
chairperson or other community leader to help them find the exact location of each of
the households.
The second barrier was crossing into Ethiopian life as a woman of Korean ancestry.
The major challenge associated with this was the language barrier. Being unable to
40
speak Amharic or Oromo generated few difficulties in the urban areas, such as Addis
Ababa. Nevertheless, as much of this research was conducted in rural Ethiopia where
there are many less English-speakers, dependence on translation was much more
prevalent. Because enumerators were hired to conduct the household surveys, the
language barrier was mostly handled by ensuring the surveys were properly translated
into Amharic. However, interviews with WVE staff and OVC caregivers required the
assistance of a translator. In a few instances, two translators were required because
the English-speaking translator did not speak the local language spoken in the
Guraghe ADP and the OVC caregiver did not speak Amharic.
Finally, perhaps the biggest regret that emerged in my reflections was not spending
more time and energy on field-testing the household surveys. While training for the
enumerators and field-testing of the surveys were conducted, having to throw out
certain survey responses during data cleaning due to inconsistent or misunderstood
responses left a taste of regret in my mouth.
Nonetheless, despite the many difficulties and challenges that emerged during field
research, there was also a great deal of data that I was able to include in this study and
analyze for significant contributions. In addition, having a research coordinator who
was familiar with both the culture and programs being observed greatly facilitated the
data collection process. A research coordinator who was both native to Ethiopia and
41
had previous experience with both WVE and WISDOM provided much of the
translation between cultures and counsel throughout the process. This was
undoubtedly invaluable to the data collecting process.
While there are many challenges related to conducting research among the rural poor
in the Two-Thirds World, it is nonetheless research that can be and needs to be
conducted. While many resources have been devoted to serving and assisting the
poor in the Two-Thirds World, much less attention has been given to actually
examining the consequences of these programs and efforts. And while the former is
assuredly important and valuable, the latter is just as essential and can help ensure
that positive efforts are discovered and continued while harmful efforts are adapted
and improved.
42
CHAPTER TWO: INCREASING FLEXIBILITY TO MITIGATE
SOCIAL EXCLUSION IN BORROWING
Part of the widespread appeal of microfinance is its high view of humanity and poor
people in particular. Microfinance assumes that those living in poverty can improve
their circumstances given adequate access to “reasonably-priced” financial services
(Dean, Schaffner, & Smith, 2005, p. 13). It assumes that poor people can work their
way towards financial improvement if institutional barriers are removed.
However, microfinance can have its own set of barriers that prevent certain groups
from gaining access. While much of the scholarly literature concerning outreach to
certain groups has focused on the “hard core poor” or “poorest of the poor” (Matin,
Hulme, & Rutherford, 1999; Matin & Hulme, 2003), this essay takes a look at
microfinance with regards to groups who are vulnerable to social exclusion.
One of the characteristic practices of microfinance is to rely on group lending to
reduce costs in providing the poor access to financial services. However, the usage of
mechanisms that require social capital can disadvantage groups that are vulnerable to
social exclusion in a community, even though they may have the capacity to borrow.
More specifically, groups that are often targets of social stigmatization can be
disadvantaged when lending groups are organized by the very source of their
discrimination, the community members themselves. Households that are affected by
43
chronic diseases, such as HIV/AIDS, or households that are caring for OVCs are
examples of groups that are vulnerable to stigmatization and negative assumptions
from others.
This essay presents a case study of WISDOM Microfinance Institution, an affiliated
MFI of WVE. The case study illustrates how a variety of actors can work to mitigate
the exclusion of groups vulnerable to marginalization in microfinance. By taking
advantage of its relationship with WVE and collaborating in its lending processes,
WISDOM was able to include more community actors in the lending and make it
more accessible to OVC caregivers. The irony was that the success of this
collaboration was unknown to both organizations. Workers from both organizations
assumed that microfinance was ill-suited to the OVC caregivers needs. Nevertheless,
unbeknownst to the two organizations, collaborating with WVE provided WISDOM
with enough flexibility in its procedures to effectively substitute the WVE community
worker’s social capital for the OVC caregiver’s lack of social capital. This process
helped mitigate socially exclusively lending practices and allowed WISDOM to
include more and monitor more. In particular, WVE community workers were
pivotal in both the screening and monitoring process for WISDOM and played a
unique role that helped bridge the gap for those more marginalized, specifically OVC
caregivers.
44
First, a brief overview of how social capital works to facilitate development practice
is presented followed by an analysis of group lending and the use of social capital in
microfinance. Second, the WISDOM case study is presented demonstrating how it
involved various actors in the screening and monitoring processes for more inclusive
lending. Lastly, the chapter will conclude with some lessons learned from the case
study.
2.1
An Overview of Social Capital Theory in Development Practice
It is not coincidental that a renewed interest in social capital has come alongside the
growing recognition and success of the microfinance movement (van Bastelaer,
2000). Woolcock points out that microfinance programs provide excellent
opportunities to study social capital simultaneously at the community and
institutional-levels (Woolcock, 2001b, p. 194). Although calling for more research
on the topic, Fukuyama adds that microfinance is one of the most successful cases of
creating and using social capital between individuals through institutions (Fukuyama,
2002, p. 33).
2.1.1 A Resurging Interest in Social Capital Theory
Although the idea of social capital has a history that dates back to the early 1900s, it
was not until scholars reintroduced the concept in the 1980s that the term would
45
regain popularity in social science literature4. As the concept continued to evolve in
definition, one of the major topics that social capital literature has focused on has
been its relationship and impact on economic development.
Coleman presents a theory of social capital in response to the lack of theories
explaining social organizations at the time. In his article, “Social Capital and the
Creation of Human Capital”, he defines social capital as consisting of “some aspect of
social structures” and facilitating “certain actions of actors” within the social structure
(Coleman, 1988, p. 98). He describes social capital as productive and explores three
forms: “obligations and expectations, information channels, and social norms”
(Coleman, 1988, p. 95). Obligations and expectations work to reduce transaction
costs. Informational channels work to provide information that could not be obtained
otherwise. Social norms work to encourage individuals to forego self-interest and
engage in collective action (Coleman, p. 1988).
E. Ostrom defines social capital as “the shared knowledge, understandings, norms,
rules, and expectations about patterns of interactions that groups of individuals bring
to a recurrent activity” (Ostrom, 1999, p. 176). She compares social capital to
physical capital and notes several differences. Unlike physical or human capital,
Ostrom observes the following: 1) social capital dissipates with non-use rather than
4
Woolcock notes how Pierre Bourdieu also developed the concept in the 1970s but his work was not
noticed by American scholars until the late 1980s (Woolcock, 2001b, 194)
46
overuse, 2) social capital can be difficult to identify and quantify, and 3) social capital
can be difficult to create through interventions (Ostrom, 1999).
Putnam (2000) further develops the concept of social capital most thoroughly in his
work, Bowling Alone. Putnam defines social capital as, “features of social
organization, such as networks, norms, and trust, that facilitate coordination and
cooperation for mutual benefit” (Putnam, 1993). He explains social capital in terms
of how civic engagement has worked to foster communication, coordination, and
collaboration through values of reciprocity and trust in others. He also notes how
social networks “foster trust, lower transaction costs, and speed [up] information and
innovation” (Putnam, 1993).
2.1.2 Effects of Social Capital in Economic Development
Much interest has been devoted to exploring the relationship of social capital to
economic development. Putnam (2000) argues that social capital can be a public
good that works to facilitate economic development. Fukuyama notes the economic
role that social capital can play to “reduce transaction costs associated with formal
coordination mechanisms like contracts, hierarchies, bureaucratic rules, and the like”
(Fukuyama, 1999, p. 2; 2002).
47
In addition, Fukuyama highlights not only the positive effects but also the negative
effects social capital can have. Social capital may create internal cohesiveness within
a group. However, a consequence of this may be an increased distrust of those
outside the group. He describes positive effects as including honesty, reciprocity, and
dependability while the negative effects might include fragmentation, conflict, and
hostility (Fukuyama, 2002). He suggests that society explore ways of increasing the
radius of trust among individuals between groups by building bridges.
Others have written on the costs of using social capital as well (Burt, 1992, 1997,
1998; Portes and Sensenbrenner, 1993; Portes, 1995; Portes and Landolt, 1996; in
Woolcock and Narayan, 2000). Ostrom points to a “dark side” of social capital,
which can be used to facilitate illegal activity in the same ways that it can be used to
facilitate collective action problems. Social networks can work as the basis of
organizational structures for groups, such as gangs, cartels, and mobs (Ostrom, 1999,
p. 176). Portes and Landolt (1996) observe the “downside” of social capital while
researching immigrant and ethnic communities who were isolated into groups and
networks. They argue that social capital in communities can also exclude some as
much as it includes others.
48
Table 4: Summary of Major Social Capital Theories
COLEMAN
Theory of
Social
Capital
OSTROM
Obligations and
expectations
Shared knowledge
and understanding
Info channels
Shared norms
Social norms
Shared rules and
expectations
Facilitate action
of certain actors
Brings recurrent
activity
PUTNAM
Networks
FUKUYAMA
Informal norms
in relationships
Norms
Trust
Outcome
of Social
Capital
Facilitates
coordination and
collective action
Leads to
cooperation
Can be positive
and negative
Table 4 gives a comparative summary of the social capital theories discussed in this
section. While there are many theories of social capital, this dissertation focuses on
social capital in its particular application to economic development and microfinance.
Therefore, it defines social capital as the norms of behavior and relationships that
transmit trust and information to influence decisions and bring about new recurrent
activities or norms. Furthermore, social capital can be productive but it can also be
destructive. Similarly, while social capital works to includes some, it also works to
exclude others. This exclusionary aspect of social capital is important particularly
when considering groups that are vulnerable to distrust because of stigma or bias
rather than reason or experience. The next section discusses how social capital
49
mechanisms are used in the group lending models of microfinance, noting particular
implications for groups vulnerable to social exclusion.
2.2
Social Capital in Microfinance Group Lending Practices
Microfinance is formed around the mission of providing small loans to people living
in poverty who normally do not have access to formal financial systems. Central to
much of its practice is the policy of group lending, often called joint liability or group
responsibility. The underlying concept of these policies is to utilize the social capital
that the poor often possess to work towards greater financial empowerment for them.
Much attention has been given to how social capital impacts the processes and
outcomes of microfinance. Some studies have focused on how existing forms of
social capital work to improve community and household outcomes. Gomez and
Santor (2001) demonstrate how social capital was a positive determinant in selfemployment earnings for Canada’s largest non-profit MFI. Anderson, Locker, &
Nugent (2002) demonstrates how social capital created from microfinance programs
could be used to solve collective action problems dealing with the management of
common pool resources.
However, the most common topic that researchers have focused on in exploring how
social capital works in microfinance is the group lending process. Group lending
50
involves loans being made to a lending or solidarity group rather than to an
individual. It is the solidarity group that is jointly responsible and together liable for
any defaults or missed payments. The benefits of group lending are experienced both
by lenders and borrowers. For the lender, transaction costs are lowered since loans
are made to the group rather than the individual. For the borrower, increased access
to financial services reduces reliance on moneylenders who can charge exorbitant
interest rates.
Several articles have examined the ways in which groups use existing forms of social
capital to reduce transaction costs for MFIs (Besley & Coate, 1995; Conning, 1998;
Morduch, 1999; Stiglitz, 1990; Stiglitz & Weiss, 1981 in Anderson, Locker, &
Nugent, 2002). Furthermore, some have argued that microfinance programs can
create social capital simply by creating new norms and routines, such as regular
repayments, interactions, and meetings (van Bastelaer, 2000). Chapter Four of this
dissertation explores how social capital was created among OVC caregivers whose
community involvement increased with the longer they participated in microfinance.
2.2.1 Conceptual Overview of Group Lending
Bhatt and Tang use both the information asymmetry and transaction costs
perspectives to better understand group lending practice. They draw on Stigliz’s
(1989) argument that markets themselves have incomplete information. The problem
51
then for a lender in impoverished rural areas is the lack of information regarding the
risk associated with lending to an individual. Only the borrower and other
community members may know this. This leads to inefficiencies in screening clients,
monitoring them, and enforcing contracts with them (Bhatt and Tang, 1998, p. 118).
Therefore, group lending arrangements can work to structure incentives so that other
community members will use the information they have to help the lender select
borrowers. The transaction cost perspective also gives insight into group lending
arrangements because it explains how inefficiencies are reduced. By organizing
borrowers into groups that meet regularly, new norms are created that reduce
transaction costs for both the lender and the borrower (Bhatt and Tang, 1998).
Similarly, Ghatak and Guinnane explore the economics behind institutions that
implement policies of joint liability. Where the following two conditions are met,
they contend that the problems associated with adverse selection (assessing the risk of
a potential borrower), moral hazard (ensuring the borrower uses the loan in intended
ways so repayment can be made), auditing costs (identifying why a borrower was not
able to repay the loan), and enforcement (forcing a reluctant borrower to repay a loan)
can be mitigated through group lending. The first condition is that community
members must know more about each other than outside institutions. The second
condition is that financial sanctions from formal financial institutions for defaulting
52
on a loan are not applicable because most are not fully part of the formal economy yet
(Ghatak and Guinnane, 1999, p. 197).
2.2.2 Peer Selection and Screening
The process of recruiting and screening clients is an important one for a MFI.
Recruitment brings in more borrowers and ultimately more profit for the MFI.
Screening helps to reduce risks and defaults, which can be just as important.
Two main models demonstrate the spectrum of methods used to mobilize and screen
clients in microfinance. The first is what de Agion and Morduch (2005) refer to as
the “Grameen-style” group lending. Solidarity groups organize and screen
themselves into groups for microloans. This has two effects. First, it gives
community members incentive to apply the information they have about a particular
individual’s past history and current work. Second, it reduces the cost of screening
for the lender since the community members are doing it themselves. With regards to
loan disbursement, two members of the group receive the first loan installments. If
the members are able to successfully repay their loans on time, the next two members
of the group are then able to receive their loan installments and so on.
In the second model, FINCA-style MFIs implement group banking at the village
level. In this context, potential borrowers do not necessarily have close ties with
53
other community members. Instead, information is spread throughout the village
about the products and services offered by the MFI. Potential borrowers then
approach the MFI and add their names to a list. Once the list reaches the number of
borrowers needed for a village bank, a solidarity group is created (Karlan, 2003; de
Aghion and Morduch, 2005, p. 93). While the costs of screening is on the lender in
this model, because the borrowers are organized in groups other transactions costs for
the lenders, such as loan disbursement and monitoring clients, are reduced.
2.2.3 Peer Monitoring and Enforcement
In both models, having the loan installments contingent upon the other members of
the solidarity group successfully making their payments is thought to encourage peer
monitoring and peer enforcement. Monitoring occurs in regularly scheduled
meetings with group members. Enforcement involves non-financial or social
sanctions, like peer pressure, from other borrowers who are motivated by the threat of
losing future loans. This helps to enforce repayment while discouraging borrowers
from pocketing their revenues without paying back their loans (de Agion and
Morduch, 2005: 86). Nevertheless, in recent years, questions have surfaced
concerning the effectiveness of social capital and group lending in peer monitoring
and enforcement. In particular, Montgomery (1996) examines a case study on
BRAC’s Rural Development Programme in Bangladesh. He questions the
effectiveness of solidarity groups in enforcing repayments and concludes that heavy
54
pressure from loan officers can often be more of a determinant than the individual
solidarity groups.
2.2.4 Microfinance and Groups Vulnerable to Social Exclusion
In the two models of peer selection discussed above, the FINCA model allows for
risk to be spread throughout a solidarity group by grouping together “safe” borrowers
with “risky” borrowers. This is in contrast to the Grameen model, which arranges
incentives so that safe borrowers group themselves with other safe borrowers (de
Aghion and Morduch, 2005). Thus, there seems to be a tradeoff in these two models
between spreading risk to include more borrowers, both risky and safe, versus
screening more to include only safe borrowers.
Montgomery (1996) argues that more flexible mechanisms in lending are needed to
prevent unnecessary exclusion of capable borrowers. Because the group lending
model of microfinance relies so heavily on social capital mechanisms, attention
should be given particularly to groups who these mechanisms may work against.
Scenarios where social capital can actually function to exclude unnecessarily should
be identified and addressed.
The remainder of this chapter discusses the case of WISDOM Microfinance
Institution and its ability to include groups vulnerable to social exclusion. Although
55
OVC caregivers are often the targets of social stigma and false assumptions,
WISDOM was able to include many into its community banks. This case study
explores how WVE community workers acted as channels that bridged information
and trust between groups to ultimately include more OVC caregivers in microlending.
Thus, it was ultimately WISDOM’s relationship with WVE that would provide it with
the flexibility needed to include more community members who were vulnerable to
social exclusion.
2.3 Including and Screening Groups Vulnerable to Stigma for
Microfinance: WISDOM Micro Financing Institution
WISDOM Micro Financing Institution S.C. (WISDOM) originally began as an effort
to provide microloans to households living in areas where WVE was already
operating. For over three and a half decades, WVE has managed various community
development programs in the form of ADPs throughout Ethiopia. See Chapter One
for more information on WISDOM and WVE ADPs.
From the beginning, WISDOM’s national strategy was to plant sub-branch offices in
the peri-urban areas nearby each WVE ADP. By December 2005, WISDOM had 9
branch offices and 28 sub-branch offices operating throughout Ethiopia. Together,
the WISDOM branch offices and sub-branch offices were servicing 25 of WVE’s 41
ADPs.
56
2.3.1 The Guraghe and Wonchi ADPs
The two WISDOM sub-branch offices that are included in this study are located in
the Guraghe and Wonchi ADPs in the southwest part of Ethiopia. A description of
how these two ADPs were selected and a map of both ADPs are provided in Chapter
One of this dissertation. Nationally, approximately 48% of WISDOM’s clientele are
women. The Wonchi sub-branch office, however, had a much higher proportion of
women clients at 68%.
By August 2007, the Guraghe ADP had identified and registered approximately
10,000 OVCs. Thus, OVCs made up 4.21% of the Guraghe ADP’s total population.
The Wonchi ADP, which is about half the size of the Guraghe ADP, also had
identified and registered close to 5,000 OVCs in its area, constituting about 4.55% of
its total population. 1,800 of these registered OVCs were also participating in WVE’s
sponsorship program as CIPs.
Interviews were conducted with WVE ADP staff, WISDOM staff, and WVE
community workers. Part of these interviews with staff members focused specifically
on households caring for OVCs. Subjects were asked questions concerning typical
needs for OVC households and the appropriateness of microfinance for them. In
addition, subjects were also asked about their own knowledge and experience
57
working with OVC caregivers participating in microfinance. Structured interviews
were also conducted with OVC caregivers, including both those who had borrowed
with WISDOM and those who had not. OVC caregivers were asked about their own
knowledge and experience with microfinance. Furthermore, subjects participating in
WISDOM loans were asked about the process of joining WISDOM and their
subsequent experience in group borrowing. Finally, in addition to structured
interviews, OVC caregivers participating in WISDOM loans were also surveyed
using an adapted version of the Community REACH instrument. See Chapter One
for a general description of the interview protocols used in this study. See the
Appendix B for the interview protocols.
2.3.2 WISDOM’s Lending Model: Screening Over Monitoring
WISDOM’s lending model is somewhere in between the spectrum of the Grameen
and FINCA lending models described above. It involves a variety of actors,
including group members themselves, community leaders, WVE community workers,
and WISDOM loan officers. Each solidarity group is made up of 10 borrowers from
the same village. Once a group is agreed upon by its members and organized, each
member must then approach the village chairman and request a letter of
recommendation for their loan. The village chairman then visits each potential
borrower’s home to assess their needs and capacity and makes a decision about
whether to recommend the community member for a microloan. In addition, each
58
member must have a mandatory savings (usually 15% of the loan) as collateral for the
loan. Once each group member has the savings and letter needed, the group can then
approach WISDOM and apply for their loans by attending a screening interview
together. Following the screening interview, a committee of WISDOM staff then
decides whether or not to grant the loans requested. Once approved, a “community
bank” is formed with the 10 borrowers who are then required to attend four weeks of
training with a WISDOM loan officer. After this is complete, loans are finally
dispersed to borrowers. Interest payments are then made on a monthly basis at group
meetings by each borrower. Each loan lasts for 12 months, at the end of which the
total principal is due. If the entire group is successful in repaying their loans, they are
then all eligible to apply for larger loans.
Thus, WISDOM’s lending model uses a policy of joint liability that is a bit of a
mixture between the Grameen and FINCA models. Similar to the Grameen model, in
the first part of the screening process, community banks are self-selected as members
themselves decide who is welcomed into the group. Nevertheless, the second part of
the screening process is a bit more like the FINCA model in that there is more
involvement from the lender in the form of the WISDOM loan committee. The
village chairman is also a major actor in the screening process because of the letter of
recommendation required for each loan. Thus, WISDOM’s screening process is
comparable to both the Grameen and FINCA models. The first part of the process
59
pushes costs onto community members who utilize their social capital to find a group.
The second part of the process pushes costs onto the lender and other community
leaders.
In theory, such a thorough screening process should result in more qualified and more
capable borrowers for WISDOM, which is important given the fact that loan dispersal
occurs all at once for a group. Unlike both the Grameen and FINCA models
described above, because loans are dispersed at the beginning of the loan cycles to all
borrowers, there is less incentive for the group members to monitor and enforce
payment from others in their group since they have already received their loan. Only
if a borrower wants to continue with a second loan would there be incentive for him
or her to monitor the repayments of others in the group. In this way, the WISDOM
lending model emphasizes screening over monitoring and enforcement.
An attempt to improve the quality of borrowers can make monitoring less important.
One wonders, however, if WISDOM did not sacrifice its breadth of outreach in the
process. Not only do potential borrowers need to find a group of community
members to join, but a small savings deposit and a letter from the village chairman
are needed to qualify for a microloan. As a result, one would think that borrowing for
OVC caregivers would be fairly difficult. Nevertheless, this was not the scenario that
the research described in the Guraghe and Wonchi ADPs. The following sections
60
explain how OVC caregivers, with the help of some WVE community workers,
overcame the assumptions being stacked up against them and participated in
WISDOM microloans to improve their livelihoods.
2.3.3 Assumptions Concerning OVC Caregivers and Microfinance
Because chronic diseases, such as HIV/AIDS, are often not well understood by
community members, OVCs can be vulnerable to stigmatization from other
community members. Irrational fears about disease can work to isolate both those
that are chronically ill and their family members, resulting in diminished social ties
and social capital (Bauman and Germann, 2005, p. 106). As a result, WVE ADP staff
members and CCCs work throughout the villages in rural Ethiopia to raise awareness
concerning HIV/AIDS and OVCs. More on CCCs can be found in Chapter One.
Despite these efforts, groups who are vulnerable to being marginalized by the
community, such as OVCs, are still likely to have less social capital than other
community members. Furthermore, efforts to raise awareness concerning OVC needs
can actually work against OVC caregivers who are interested in microloans.
Messages directed at raising support for OVCs can also indirectly communicate that
OVC caregivers are very needy and make poor candidates for microloans.
61
In the instances of the Guraghe and Wonchi ADPs, much of the doubt concerning the
suitability of microfinance for OVC households came from the staff of both
organizations rather than the community members themselves. Case study interviews
were conducted with eight OVC caregivers. Of the five OVC caregivers interviewed
that had either current or previous experience with microfinance, four responded that
they did not feel discriminated against by other community members. Nevertheless,
the impact of disease was apparent on OVC caregivers interviewed who were not
participating in microloans. Two of the three OVC caregivers who had never
participated in microfinance indicated that fear of dying and passing on debt to their
children was a major factor in why they did not seek microloans for themselves. One
OVC caregiver rhetorically asked, ‘who will pay the debt if [I] pass away?’
Still, interviews showed that much of misgivings concerning whether OVC caregivers
had the capacity to be successful with microfinance was voiced from the staff of both
organizations. Table 5 shows a summary of interview responses concerning the
suitability of microfinance for OVC caregivers and concerning whether or not a staff
member was aware of OVC caregivers participating in WISDOM. As the table
indicates, out of the 16 staff workers interviewed, only the WVE community worker
62
Table 5: Worker Responses on Microfinance for OVC Caregivers
WORKER TYPE
SUITABILITY OF MICROFINANCE
FOR OVC CAREGIVERS
NUMBERS OVC
CAREGIVERS IN
WISDOM
---
“some”
WVE ADP Staff Worker
Guraghe HIV/AIDS
Wonchi HIV/AIDS
OVC HHs are very needy and are not
going to a credit program…they are
afraid of repayment because they do not
have the ability to make money…need a
mechanism to give them access”
“most do not”
Wonchi ADP Mgr
“it is important for WISDOM that OVC
HHs can pay their loans back”
---
Guraghe ADP Mgr*
----
---
WVE Community Worker
Guraghe – Yegobet
Need more direct aid
“no OVC HHs in
WISDOM”
Guraghe – Sheremo
Need direct aid
“none”
Guraghe – Shebraden
Need more strategies to include OVC
HHs…need to revise strategies and need
to adjust interest rates because they are
too high for OVC HHs
“none”
Guraghe – Wasamer
OVC needs are financial support and
material or direct aid…OVC HHs are
afraid of repayment rates and the
difficulties they will have
Wonchi – Fite***
May not be very beneficial
“not aware of any”
Wonchi – Belabela
Important for OVC HHs but the issue of
interest and repayment rates need to be
addressed…people do not want to be
accountable for someone else if they
default on their loan
“some, not many”
Wonchi – Dimtu
Will help OVC HHs but should be
accompanied with training and changed
strategy towards unable poor
“7 out of 20”
“not many…not much”
63
Table 5: Worker Responses on Microfinance for OVC Caregivers, Continued
WISDOM Staff Worker
Wonchi – Branch Mgr
Requirement of additional capital needs
to be addressed for OVC HHs to
participate
Guraghe – Branch Mgr
Methodology may need to be changed
Wonchi – Loan Officer
Good because it enables IGA but there is
an issue of collateral for OVC HHs
“2 or 3…depends on
income”
Guraghe – Loan Officer
Other community members are resistant
to include OVC HHs because of their lack
of capital
---
---
---
Wonchi – Loan Officer**
“No OVC HH”
---
Notes: * Subject was Program Officer standing in as Interim ADP Manager. ** Subject had only been
loan officer for 2 months at time of interview. *** No MFIs were operating in the village at the time
of interviews.
from Dimtu in the Wonchi ADP had an accurate sense of how many OVC caregivers
were participating in WISDOM. Two of the workers interviewed, the Guraghe
Interim ADP Manager and the WISDOM Loan Officer in Wonchi, had limited
experience in their position and thus did not know of any OVC caregivers
participating in WISDOM. Nevertheless, five staff workers responded that they
believed only a few OVC caregivers were participating in WISDOM and five staff
workers responded that they did not believe that any OVC caregivers were
participating at all. The remaining three workers did not indicate whether they were
aware of OVC caregivers in WISDOM or not. Thus, most of the staff interviewed in
64
both organizations believed that few OVC caregivers, if any, were participating in
WISDOM.
Furthermore, both WVE and WISDOM concluded that OVC caregivers in general
were not good candidates for the loans that WISDOM currently offered. Some staff
thought the microfinance in general may not be beneficial. Three staff workers
considered direct aid or support as being more helpful than loans to OVC households.
Others believed WISDOM’s current policies were obstacles that prevented OVC
caregivers from participating. Four staff workers stated that interest or repayments
rates were too high for OVC caregivers to currently participate. Three workers
thought that the initial deposits required for a microloan with WISDOM were too
high for OVC caregivers.
In addition, two workers voiced the opinion that other community members in
general did not want to join a community bank with OVC caregivers and be
accountable for any defaults that might occur. They remarked that the screening
process, in particular, for community banks discouraged OVC caregivers from
participating in microloans. More specifically, they believed that giving group
members decision-making authority concerning who was in their group worked
against OVC caregivers. One worker described community members as being
“resistant” to having OVC caregivers join their group.
65
Finally, two staff workers thought that WISDOM needed to change their strategy in
general towards OVC households. Thus, staff from both organizations generally
concluded that OVC caregivers did not have the financial or social capital to
participate in WISDOM in sizeable numbers. Many believed they were too poor and
that microfinance was ill suited towards their needs.
2.3.4 Overcoming Assumptions: OVC Caregivers as WISDOM Clients
Nevertheless, despite the assumptions of WVE and WISDOM staff, data collected for
this study indicates that OVC caregivers did in fact participate in WISDOM. During
data collection, one of the first tasks that needed to be completed was to identify
WISDOM clients who were also OVC caregivers. This was achieved by crossreferencing WISDOM client lists with WVE OVC household lists for the villages that
had been selected for the study. A description of how these villages were selected is
given in Chapter One of this dissertation. The cross-referencing process had not been
attempted by either organization prior to this study and produced results that were
quite different from the assumptions made by staff members interviewed.
The process resulted in identifying 67 OVC caregivers who were participating in
WISDOM loans in Guraghe and 63 OVC caregivers who were participating in
WISDOM loans in Wonchi. In the Guraghe ADP, 20.8% of the WISDOM clients
who lived in the villages included in this study were caring for OVCs in their
66
households. This is a much higher percentage than was expected especially when
compared to the 4.21% of the Guraghe population who were registered as OVCs with
WVE. See Table 6 below for a breakdown of these percentages by village.
Table 6: Percent of WISDOM Clients Caring for OVC in Guraghe
VILLAGE
% OF WISDOM CLIENTS
CARING FOR OVCS
Agena
Desene
Shebraden
Sheremo
Wadiye
Wasamer
Yegobet
20.4%
36.0%
9.7%
17.0%
15.2%
20.0%
30.1%
TOTAL
20.8%
The variance in percentages may be attributed to different village characteristics. For
example, according to the WVE HIV/AIDS focal staff member for Guraghe, Agena
has higher prevalence rates and higher numbers of OVCs compared to other villages.
This is likely because it is a peri-urban area, which tends to have higher prevalence
rates than rural areas because it has a more transitory population. The WVE ADP
staff member for Guraghe also observed that Agena, Desene, and Yegobet all had
strong CCCs.
67
Thus, although most of the WVE and WISDOM staff members assumed OVC
caregivers were too needy and too isolated to be suitable candidates for WISDOM
loans, the data shows a different scenario. Indeed, many OVC caregivers were
already participating in WISDOM loans and, as Chapter Four argues, many OVC
caregivers were better off for having done so. What then explains this discrepancy?
How was it possible for so many OVC caregivers to join WISDOM even while its
staff believed they would not be suitable borrowers? The next section argues that
WISDOM’s relationship with WVE and its use of multiple community actors helped
it to include more OVC caregivers in WISDOM’s microfinance services.
2.4
Flexibility through Multiple Actors
Ironically, the success of WISDOM’s lending model in including OVC caregivers
was unknown even to itself. When interviewed for knowledge and experience
concerning clients who were also OVC caregivers, very few WVE or WISDOM staff
had knowledge of any such clients prior to data collection for this study. What
contributed to WISDOM’s ability to include groups that are vulnerable to false
assumptions and social stigma?
Over a five-year period prior to data collection, there were a variety of actors that
were involved in WISDOM’s client recruitment processes. This study argues that it
was the involvement of these community actors, and WVE community workers in
68
particular, that provided the flexibility needed to include groups vulnerable to social
stigma, in particular OVC caregivers.
Clients from case study interviews described the various actors who provided the
necessary information they needed to join and enroll with WISDOM. Some clients
had heard about WISDOM through the promotions held in various villages. These
promotions were usually conducted by WISDOM staff in towns or peri-urban areas to
help community members learn about microfinance and how to apply for a
microloan. Others had heard about WISDOM’s products and services through other
community members in their villages. Still in other instances, it was the village
chairman who referred community members to WISDOM for its microloans. In all
three instances, however, the actors involved (with the exception of community
members who may be inviting members to form a community bank together) are
merely providing information to the OVC caregiver.
In other words, learning about microfinance is not enough since the OVC caregiver
still needs to find a community bank to join. Furthermore, if the OVC caregiver is
indeed vulnerable to false assumptions or stigmatization that make joining a group
difficult, merely receiving information about microfinance may not be enough to
include more OVC caregivers in WISDOM. Nevertheless, although information
sharing may not be enough, it is still necessary and important. Thus, having a variety
69
of actors who function to distribute information about WISDOM provides flexibility
and is a vital part in explaining how WISDOM was able to include so many OVC
caregivers in its services.
2.4.1 The Role of WVE Community Workers
When examining the overall success in including OVC caregivers into WISDOM
financial services, observing where and when OVC caregivers joined WISDOM is
telling. Table 7 shows the distribution of OVC caregivers who joined more recently
compared to those who joined earlier. Of the 63 OVC caregivers who joined
WISDOM in Wonchi, only 11 of them had joined WISDOM within three years of
data collection. The majority of the OVC caregivers in Wonchi had joined more than
3 years before data collection. Conversely, of the 67 OVC caregivers identified in
Guraghe as WISDOM clients, 49 of them had joined within three years while only 18
had joined more than 3 years before data collection. Nevertheless, comparing the
totals for each time period for Guraghe and Wonchi ADP numbers can be deceiving.
What is more telling are the average numbers of OVC caregivers that joined each
village. The average number of OVC caregivers in the Wonchi ADP that joined
WISDOM more than three years prior to data collection is 25.5. This is quite a large
number compared to the other averages, which range from 2.6 to 7.0. In a village
where there were approximately 350 OVCs registered with WVE in 280 households,
Belabela had 28 OVC caregivers join more than three years prior to data collection.
70
This means approximately 10% of the OVC caregivers in Belabela joined WISDOM
during this earlier time period.
Table 7: OVC Caregivers Participating in WISDOM by Village
ADP
Guraghe
Village
Joined WISDOM
More Than 3 Yrs
Ago*
Agena
Desene
Shebraden
Sheremo
Wadiye
Wasamer
Yegobet
10
13
5
7
5
0
9
49
7.0
9
2
1
1
0
3
2
18
2.6
Belabela
Dimtu
4
7
11
5.5
28
24
52
25.5
Total
Avg Per Village
Wonchi
Joined WISDOM
Less Than 3 Yrs
Ago*
Total
Avg Per Village
* Prior to data collection in August 2007.
Why were Belabela and Dimtu villages able to recruit, organize, and accept so many
more OVC caregivers for WISDOM loans during this earlier time period? One
possible reason for Belabela and Dimtu having such high numbers of OVC caregivers
who joined during this earlier time period is the role that the WVE community
workers played in the Wonchi ADP. More on the responsibilities of World Vision
community workers can be found in Chapter One. WVE community workers in
Wonchi were highly involved in recruiting and organizing groups for WISDOM in
71
years past. Community workers described during interviews how, in years prior,
they assisted WISDOM with group mobilization, client selection, and group trainings.
Furthermore, it was about 3 years before data collection that WISDOM loan officers
started to take over these responsibilities from WVE community workers in Wonchi.
At the time of the interviews, the Wonchi WVE community workers described
themselves as not currently being very involved in group mobilization and client
selection. Community workers referred community members to WISDOM if they
asked about their services but interaction with WISDOM loan officers was described
as “limited”. Meeting only to consult on issues rather than share in responsibilities,
each worker had “different type(s) of work” to focus on now.
Since the time when WVE community workers have been less involved with
WISDOM, Belabela accounts for only 4 OVC caregivers who joined WISDOM and
Dimtu accounts for only 7 OVC caregivers. Thus, interviews with WVE community
workers from the Wonchi ADP support the argument that one reason for the falloff in
new clients among OVC caregivers in recent years is the decreased involvement of
WVE community workers.
With regards to the Guraghe ADP, all of the WVE community workers interviewed
from Guraghe had less than 3 years of experience. This means that there was no
72
involvement from WVE community workers in recruiting members for WISDOM
community bank groups in Guraghe during the earlier time period. The Guraghe
ADP manager also indicated that there was no collaboration between WISDOM and
WVE beyond financial grants. Thus, although WVE community workers may have
been involved in forming community banks in Guraghe, data collection did not find
the same degree of involvement in Guraghe as there was during the earlier years in
Wonchi. In fact, the more recent WISDOM borrowers in both ADPs were much
more likely to be recruited by WISDOM staff than anyone else.
Thus, data suggests that having WVE community workers involved in organizing
community banks results in a greater number of OVC caregivers joining WISDOM.
Similarly, when WVE community workers were not as heavily involved in
mobilizing community banks, OVC caregivers did not join WISDOM at the same
levels. The earlier period for the Guraghe ADP and the more recent period for both
the Guraghe and Wonchi ADP had similar levels of involvement of WVE community
workers in WISDOM community banks and had similar levels of OVC caregivers
joining WISDOM. It was only during the earlier period in the Wonchi ADP when
WVE community workers were heavily involved that dramatic increases in OVC
caregivers joining WISDOM occurred.
73
Thus, the Wonchi ADP illustrates well the importance of having information channels
and community advocates working for greater inclusion in microfinance services,
especially for those most vulnerable to social isolation such as OVCs. Nevertheless,
there is one question that remains. If WVE community workers in Wonchi played
such an important role in including OVC caregivers in WISDOM services, how do
we explain how so few WVE community workers knew about OVC caregivers who
were participating in WISDOM? In other words, how could they have been such a
vital factor to OVC caregiver inclusion without knowing it?
The answer lies in understanding the role that the WVE community worker plays in
the village. The responsibilities of the WVE community worker necessitate their
frequent interaction with those that are the most vulnerable and marginalized in the
village. WVE community workers frequently build relationships, share information,
and interact with community members that can be excluded by other community
members. See Chapter Three for a description of the WVE community workers role
and responsibilities in the community. Thus, even if they were not intentionally
targeting OVC caregivers for WISDOM services, because of their frequent
interaction and communication with community members, it is not unlikely that they
unknowingly recruited OVC caregivers to join WISDOM community banks. In this
sense, the social capital of the WVE community worker is being given to the OVC
caregivers bringing both information and relationship to the household.
74
2.5
Monitoring through Collaborative Relationships
While the last section focused on the WVE community worker’s relationship with the
village, this section focuses on the community worker’s relationship with WISDOM.
Because incentives for monitoring other borrowers in a community bank are weaker
in WISDOM’s lending model than other models, the monitoring process relationship
to client performance needs to be considered to ensure organization sustainability.
While the two-part screening process for WISDOM involves multiple community
actors, the monitoring process mainly consists of monthly community bank meetings
where each member makes interest payments. WISDOM loan officers also attend
these meetings to collect payments, disperse loans, or conduct any training that
WISDOM might be offering. Thus, formally, the monitoring process involves fewer
actors than the screening process. In addition, as mentioned above, because loans are
dispersed all at once to all members of the community bank, incentives may be
weaker than if dispersals were more staggered across a loan cycle.
Nevertheless, it is again the role of the WVE community worker in a village that
seemed especially important to OVC caregivers’ performance in WISDOM. In this
case, it is the community worker’s relationship with WISDOM that is noteworthy.
More specifically, the quality of relationship between the WVE community worker
for a village and the WISDOM staff appears to have had an impact on client
75
outcomes among OVC caregivers. Although involvement of community workers in
the screening process had decreased in recent years in Wonchi, interviews indicated
that WVE community workers still maintained relationships with WISDOM staff to
exchange information in both the Guraghe and Wonchi ADPs.
Furthermore, how each community worker assessed his or her own relationship with
WISDOM at the point and time of the interviews varied. Some community workers
were enthusiastic about WISDOM and described their relationship very favorably.
Others were more disappointed with their current interaction and described
‘communication gaps’ that existed. Based on the community worker’s own
assessment, each relationship was categorized for this study as either stronger or
weaker. Responses such as ‘good’, ‘positive’, or ‘strong’ were categorized as a
strong relationship with WISDOM. Responses from community workers as either
‘medium’ or having ‘gaps’ were categorized as weaker.
After each relationship was categorized as stronger or weaker, content analysis was
conducted on the WVE community workers interviews identifying three major
characteristics surrounding the collaborative relationships: 1) frequency of
interaction, 2) content of interaction or activities involved, and 3) areas where more
interaction is needed. Table 8 shows a summary of these three characteristics. The
table shows actually a considerable amount of overlap when comparing the stronger
76
and weaker collaborative relationships between WVE community workers and
WISDOM staff. The frequency of interaction was similar between the stronger and
weaker relationships, as both were limited to problems or issues that arose rather than
a more regular interaction. The types of activities the two sides collaborated in were
also similar between the stronger and weaker relationships. In both categories of
Table 8: Comparison of Community Workers Relationship with WISDOM
Comments from CWs with
Stronger Collaboration
Comments from CWs with
Weaker Collaboration
Rating of collaboration
•
•
•
Good relationship
Very positive
Strong relationship
•
•
Medium
Gaps in collaboration
Frequency of interaction
•
As issue comes up
•
Limited
Content of interaction
•
Consults with
WISDOM about how to
care for clients
Promotes WISDOM to
community
Organize community
members and refer
them to WISDOM
Follow up with clients
in WISDOM and
reports on performance
to ADP manager
•
Talk and consult
concerning community
issues
Coordinates with
WISDOM for community
members
Organize community
members and refer them to
WISDOM
Need more awareness
of village
•
•
•
•
•
Areas where more
interaction is needed
•
•
•
Planning and training
WISDOM needs to remain
in contact with us and
other community leaders
more
Note: CW is the abbreviation for Community Worker.
77
relationships, community workers met with WISDOM staff to consult about issues
and clients. WVE community workers still helped organize community members for
WISDOM, though to a lesser degree than in the earlier period in the Wonchi ADP.
Nevertheless, there was one important difference between the two. First, WVE
community workers in stronger collaborative relationships with WISDOM reported
that one of their responsibilities was to also follow up with clients whom they helped
join WISDOM. This served as an additional informal monitoring mechanism for
WISDOM clients. WVE community workers would check up on the progress of
client loans and provide any advice they could offer.
Furthermore, the fact that WVE community workers assessed their relationship with
WISDOM as positive may have had important implications for the quality of
information that was shared with WISDOM staff members. Where WVE community
workers felt more positive about their relationship with WISDOM, he or she may
have felt more comfortable sharing about a client’s progress thereby improving
WISDOM’s ability to monitor its own clients.
Thus, to test the hypothesis that stronger collaborative relationships with WVE
community workers improved monitoring and ultimately resulted in improved client
outcomes, a series of regressions were conducted on data collected from OVC
78
caregivers participating in microloans. Using data from the OVC household surveys,
regressions were run modeling several household outcomes as a function of the
quality of the collaboration between the WVE community worker and WISDOM
staff. Only clients who lived in villages whose WVE community worker had been
interviewed were included in the sample.
Regressions were conducted according to the following equation:
Household = α * Stronger Collaboration + β * Other Controls + ε
Outcomes
Dummy Variable
Based on interviews, a village was categorized as having either stronger or weaker (1
= stronger, 0 = weaker) collaboration between the WVE community worker and
WISDOM staff. The results of the regressions can be found in Table 9. Loan growth
is the percentage of increase from the amount of the first loan borrowed by the client
compared to the amount of the current loan borrowed by the client. Loan growth can
be used as an indicator of success for a client based on the assumption that successful
participation in the first loan cycle often leads to a greater capacity to borrow (de
Aghion & Morduch, 2005). Livestock assets were values based on survey data
indicating the number and type of animal that a household owned. Livestock prices
were based on pricing information from USAID and NGO websites (USAID, 2008).
See the Appendix G for more information on livestock pricing. Average meals is the
79
Table 9: Impact of Community Worker on Client Performance
Loan
Growth
Livestock
Assets
Avg
Meals*
Stronger Collaboration?
(0 = Weaker, 1 = Stronger)
0.249**
0.472***
0.504***
Yrs in Microfinance
Current Loan Amt
Training in IGA
0.552***
0.493***
0.063
-0.027
-0.041
0.320**
0.134
0.005
-0.142
# of M Adults
# of F Adults
# of M Children
# of F Children
0.213*
-0.139
0.018
-0.052
0.201
0.162
-0.188
0.284*
-0.333**
-0.171
-0.047
0.039
Marital Status: Widowed
Marital Status: Never Married
0.232
0.033
0.363*
0.346*
-0.011
0.092
Caregiver Educated?
0.054
0.053
-0.067
Receives Money?
Sell Assets in Past Yr?
Chronic Illness in HH?
Percent of School Aged Children in
CIP
-0.040
-0.024
---------
0.033
----0.200
-----
-------------0.504***
Solidarity Group: Close-knit?
Solidarity Group: Can be trusted?
Solidarity Group: Do not share same
values?
Solidarity Group: Do not generally
get along?
-0.046
-0.015
0.022
-0.066
0.144
-0.165
-0.376***
0.121
0.179
0.006
0.165
0.149
6.681***
0.692
44
2.51**
0.379
44
3.559***
0.488
44
F-stat
Adjusted R2
Observations
Note: Average meals are average number of meals per household member per day. All entries
are standardized coefficients. Sample is taken from clients who were part of villages where
community workers were interviewed and their relationship with WISDOM was assessed as
positive or negative. All VIF scores tested below 2.973.
*** p < 0.01; ** p < 0.05; * p < 0.10;
80
average number of meals consumed per member of the household for the day before
the survey was conducted. See Appendix C to F for survey instruments used to
collect household data.
Overall, the models were able to explain high proportions of variance in the
household outcomes. The adjusted R2 ranged from 0.379 to 0.692. F-stats were also
significant at the 5% level or lower. For client and household outcomes including
loan growth, livestock assets, and average meals per day per household member, the
quality of collaboration was significant at the 5% level or lower. In addition, stronger
collaboration between the WVE community workers and WISDOM staff in the
village resulted in positive impact on outcomes for OVC caregivers participating in
WISDOM loans. Standardized coefficients for the quality of collaboration ranged
between 0.249 and 0.504 for the outcomes tested.
Thus, data analysis supports the hypothesis that collaborating with community
workers can improve monitoring mechanisms for a microfinance agency thereby
improving client outcomes. Furthermore, results also further call into question the
assumptions held by staff members that many OVC caregivers are not suitable
candidates for microfinance. Instead, data supports that with inclusive channels of
information and action, not only can OVC caregivers be included in microfinance but
they can succeed in it as well.
81
2.6
Summary and Conclusion
There are two major mechanisms in microfinance that require social capital from the
borrower. First, the process of being selected into a community bank requires social
capital in the form of both information concerning microfinance and trust from other
community members to join the group. Second, the process of monitoring and
enforcement uses social capital in the form of information and social pressure to
ensure timely repayment of loans.
This study demonstrates how the social capital from a community worker can be used
to substitute for the lack of social capital from community members who are
vulnerable to social exclusion. In this case, it was the WVE community worker that
was able to use his or her social capital on behalf of OVC caregivers, even if it may
have been unknowingly. The involvement of the WVE community workers in
forming community banks for WISDOM served to include more OVC caregivers
than anyone had realized while the collaboration of WISDOM with WVE community
workers in monitoring helped to gain information that would lead to more effective
monitoring. Thus, it was the involvement of multiple community actors that provided
WISDOM with more flexibility in its processes and enable it to include more and
monitor more.
82
Microfinance can be a powerful tool that can lead to financial empowerment for the
poor. Its use of social capital can often serve to increase access to financial services
that would otherwise remain unavailable. Nevertheless, as many scholars have
pointed out, social capital can have both positive and negative effects on communities
and its processes. In particular for microfinance, its use of social capital for client
selection and screening can leave out those who tend to be marginalized or excluded
but are nonetheless capable.
This case study examining WISDOM shows that indeed social exclusion and false
assumptions were at work with regards to OVC caregivers. OVC caregivers were
thought to be ill suited for microfinance by staff workers from both organizations.
Nevertheless, despite these assumptions, research showed a high proportion of OVC
caregivers already participating in WISDOM loans and services.
What contributed to WISDOM’s ability to include this group? The role of the WVE
community worker to serve as a bridge to the community and WISDOM proved
invaluable to the success of OVC caregivers who were participating in microfinance.
Where and when WVE community workers were heavily involved in organizing
community banks, more OVC caregivers were included in the process. In addition,
where there was a stronger collaborative relationship between the WVE community
83
worker and the WISDOM staff, monitoring efforts contributed to better performance
outcomes among OVC caregivers.
Thus, it was ultimately WISDOM’s relationship with WVE that provided the
flexibility it needed to include more and monitor more. WISDOM’s relationship with
WVE allowed flexibility in the sense that it was WVE’s social capital that could then
used to help OVC caregivers succeed in microfinance. It was the unique position of
the WVE community worker that would provide both the relationship, information,
and action that was needed to include more OVC caregivers in borrowing. WVE
community workers are hired to keep close relationships with the community and
with those that tend to be marginalized or excluded specifically. Furthermore,
because of the unique nature of social capital, its use actually strengthens the
organization as opposed to weakening it.
Thus, if microfinance agencies are to be faithful to their mission to empower the poor,
attention must be paid to the groups that are most vulnerable to being left out.
Flexible mechanisms can help agencies to include these groups while still
maintaining the strengths of their models and ability to use social capital to reduce
transaction costs in lending.
84
CHAPTER THREE: INCENTIVIZING COLLABORATORS FOR
MICROFINANCE INSTITUTIONS (MFIs)
With pressure building on MFIs to cut costs and improve financial self-sufficiency,
collaborating with other community organizations has become an increasingly
attractive solution for MFIs. Collaboration can allow MFIs to adjust to their
operational constraints while expanding their catalogue of services. Once offering
health and education services in conjunction with microloans, MFIs now find it
difficult to provide these additional services on their own despite the benefit it may
have for the community. Nevertheless, the unique nature of microfinance with the
manner in which it maintains organized groups serves as an efficient means to deliver
other social services as well. Thus, the approach of working with other agencies that
can provide needed social services to the lending groups can be a win-win situation
for all.
In addition, collaborating with community organizations can improve a MFI’s overall
effectiveness in its service delivery of its microfinance products. Chapter Two
illustrates the importance of collaborating with other community organizations for
MFIs. Working with WVE community workers not only helped include community
members who are vulnerable to social exclusion, but it also helped improve
WISDOM’s information and monitoring efforts. Furthermore, it was WISDOM’s
overall relationship with WVE that provided valuable flexibility in its processes and
85
procedures to include more and monitor more by allowing the transference of social
capital from the WVE community worker to the OVC caregiver.
Nevertheless, while potentially beneficial to those involved, collaborative
relationships can also bring in a whole new set of challenges for an agency. One of
the challenges that exists for the MFI is to consider how to incentivize community
workers and field staff in other community organizations to collaborate with them.
Organizational literature has focused on the importance of motivation at the
individual-level for decades. Lock and Schweiger (1979) identified worker
motivation as one of the top two individual-level determinants of worker
effectiveness while Eckardt and Schuler (1992) identified motivation as one of the
most significant factors for individual professional achievement. More recently,
Thomas (2003) has explored motivations for interagency collaboration and how they
are affected by the role an individual plays in an organization.
While there is much concerning organizational cultures that influences interagency
collaboration, this chapter considers the factors that influence motivation at the
individual-level for community workers to collaborate with MFIs. While the
previous chapter of this dissertation focused on the impact and benefits of
collaboration between WISDOM and WVE, this chapter is more concerned with
factors at the individual-level that influence the processes involved in collaborating.
86
It discusses the collaborative efforts that WISDOM had with the WVE community
workers in the Guraghe and Wonchi ADPs to help screen and monitor borrowers in
the villages.
The next section provides a brief overview of the debate over institutional design that
emerged as the “microfinance schism” and created an expectation for MFIs to work
towards financial self-sufficiency. This is followed by a case study that examines the
various levels of motivation among WVE community workers to collaborate with
WISDOM along with a discussion of the various factors at work influencing these
levels of motivation. Furthermore, the organizational nature of MFIs as a
entrepreneurial/commercial non-profit (Hansmann, 1987) or a social business (Yunus,
2007) brings in an additional element of complexity in collaboration that is
considered as well.
3.1
The “Microfinance Schism”
The “microfinance schism” was the name given to the debate involving
institutionalists on one end and welfarists on the other (Morduch, 2000).
Institutionalists argued that MFIs needed to prioritize financial self-sufficiency and
that design decisions needed to reflect this priority. This approach has been called the
institutionalist, minimalist, or self-sufficiency approach (Morduch, 2000; Schreiner,
2002a; Bhatt & Tang, 2001a). MFIs of this orientation characteristically focused on
87
“creating financial institutions to serve clients who are not served or are underserved
by the formal financial system”, prioritizing financial and operational selfsufficiency, and emphasizing breadth of outreach in its approach to poverty
alleviation (Woller, Wheeler, & Checketts, 1999, p. 2; Morduch, 2000). It was
thought that the major tradeoff for offering a wider range of services was the slower
progress in attaining financial sustainability5 (Morduch, 2000, p. 619). Thus, interest
rates needed to be set accordingly and non-financial services needed to be cut.
Integrated microfinance programs that also offered social services ought to be
converted to stand-alone programs that only offered financial services.
This was the only way, institutionlists argued, to ensure financial self-sufficiency for
the MFI. Thus, the argument was that every dollar needed to be stretched as far as it
could, as the poor are not served as well by organizations whose future is unstable.
Schreiner argued, “the poor are many but the donor dollars are few…selfsustainability produces the strongest incentives to improve social benefits through
time” (Schreiner, 2002a, p. 22). Furthermore, donors had a right to know that their
dollars were being used effectively and not wasted. In addition, borrowers would be
less likely pay back loans if they do believed the MFI would not last (Von Pischke,
2002, p. 75).
5
The Microcredit Summit Campaign attacks this as second of three of the most damaging myths of microfinance
practice (Microcredit Summit Campaign, 2005, p. 11).
88
On the other side of the institutionalist was the welfarist or poverty approach
(Morduch, 2000; Woller, Wheeler, & Checketts, 1999). This side typically argued
for a more integrated package of social services. It was typically described as
focusing on the immediate improvement in well-being of the participants, using
subsidies for operational activities, and emphasizing depth of outreach in its approach
to poverty alleviation (Woller, Wheeler, & Checketts, 1999, p. 3). It argued that
social service programs should not be cut because they contribute to the overall
mission of the MFI. Furthermore, solidarity or lending groups provide excellent
vehicles to provide other social services, such as business and professional training,
political participation training, and health education and services programs.
Finally, welfarists also criticized the premise that MFIs needed to be uniquely
financially self-sufficient. In the article in which he coined the term “microfinance
schism”, Morduch questions the institutionalists and writes:
…it is not clear why the starting point for so many is the belief
that, as a matter of course, funding will be pulled away from
programs, even those able to demonstrate sustained social
effectiveness. Moreover, there has never been a general
presumption that the most effective poverty alleviation programs
can be – or should be – self-financing (Morduch, 2000, p. 619).
89
3.2
Wisdom Micro Financing Institution (WISDOM) and World
Vision Ethiopia (WVE)
It was the institutionalists who largely won this debate and the result has been an
industry filled with MFIs facing the expectation that they will one day become and
remain financially self-sufficient. Nevertheless, in the midst of these pressures, some
MFIs have turned to innovative partnerships with other community groups and
organizations that provide the social services it had once provided itself. In these
sorts of relationships, the MFI and the social service organization operate separately
but partner to coordinate the delivery of services to their shared clients.
Being affiliated to WVE, WISDOM has the opportunity to collaborate with a
community development organization that often does not exist for other MFIs.
WISDOM originally began as part of WVE development program. In the 1980s,
WVE began to offer microloans alongside its development operations. Since then the
Ethiopian government has implemented policy that requires microfinance operations
to be conducted by share companies that are owned wholly by Ethiopian nationals. In
1998, WISDOM was established as a separate organizational entity from WVE and
was issued a business license by the National Bank of Ethiopia. Since its birth,
WISDOM has had a relationship with WVE. Nevertheless, this relationship has
evolved and changed over its history both because of Ethiopian policy and because of
90
WVI’s management strategies. See Chapter One for more information on WISDOM
and WVE ADPs.
This study explores the motivations at work in the collaborative efforts that
WISDOM had with WVE in the Guraghe and Wonchi ADPs to screen and monitor
WISDOM clients. Interviews with WISDOM staff, WVE ADP staff, and WVE
community workers were conducted during field research. Subjects were questioned
around topics concerning their role and responsibilities, their experiences
collaborating with the other organization, information and resources that were shared
between the two organizations, and perceptions concerning the other organization. In
addition, staff members were asked which workers of the other organization they had
recently talked to and how frequently this occurred. See Chapter One for a general
description of the interview protocols used in this study. See the Appendix A for the
adapted interview protocol for staff members of both organizations.
Attempting to observe and measure an individual’s motivation levels can be difficult
largely because of its intangible nature and the inability to isolate motivation from
skill or ability. Studies focusing on the affective models of participation have,
however, suggested that higher levels of worker satisfaction are related to increased
motivation for workers (French et al., 1960; Miller and Monge, 1986). The rationale
is that if a worker perceives a greater reward of satisfaction for increased effort
91
towards a goal, he or she will be motivated to pursue such purposes. Consequently,
following this line of reasoning, worker effort or participation is related to motivation
as well. If a worker is motivated and able, he or she will participate more. Thus, for
the purposes of this study, individual-level motivation for collaborating with
WISDOM is examined in terms of a worker’s satisfaction with his or her
collaboration with WISDOM and in terms of a worker’s level of involvement with
WISDOM’s operations.
3.2.1 Motivations to Collaborate Among WVE Community Workers
The collaborative effort examined in this study occurred between the WISDOM staff
and the WVE community workers. The effort mainly focused on organizing and
monitoring WISDOM community banks and consulting with one another concerning
mutual clients. As described in Chapter Two, WVE community workers were
involved in mobilizing community banks for WISDOM for several years. The extent
and degree to which this occurred, however, varied between ADPs and villages.
Satisfaction concerning a WVE community worker’s own collaborative relationship
with WISDOM varied between “very positive” to “positive” to “needs improvement”.
Two WVE community workers had responses that were categorized as “needs
improvement” because they described the relationship as either having
communication gaps with infrequent interaction or as lacking joint planning with
92
limited interaction. The other four WVE community workers that described their
relationship with WISDOM as either “very positive” or “positive” expressed little
discontentment with the level of interaction between the two organizations and used
positive adjectives to describe their relationship with WISDOM, such as “strong”,
“good”, or “very positive”.
Concerning the varying levels of involvement with WISDOM’s operations, there
were generally four types of activities in which WVE community workers were
involved. They include the following, which are listed in progressively higher levels
of involvement with WISDOM: 1) consulting together on either community issues or
mutual clients, 2) referring community members to WISDOM services, 3) organizing
groups for WISDOM, and 4) following-up with WISDOM clients. Consulting
together involves the lowest level of involvement for WVE community workers since
this generally occurs in reaction to a problem that has developed either with a group
or client. Thus, no initiative is required on the part of the WVE community worker.
Referring community members to WISDOM services involves a slightly higher level
of involvement since the WVE community worker is either doing so in reaction to a
question to a community member or out of his or her own initiative. Furthermore,
because the WVE community worker is also providing counsel to community
members while promoting WISDOM, he or she is also vouching for WISDOM.
Organizing community banks for WISDOM demands more effort from the WVE
93
community worker since it requires coordinating multiple community members and
facilitating their initial group activities. Finally, following-up with WISDOM clients
involves the highest level of involvement or commitment in that it requires high
levels of initiative from the WVE community worker since WISDOM staff
themselves often do not have the time to monitor their clients.
3.3
Factors in Motivating Collaboration Among WVE Community
Workers
The research that Thomas (2003) conducts on individual-level motivations in
interagency collaboration provides an appropriate starting place when considering the
various factors in motivating collaboration among WVE community workers.
Although Thomas’s research is focused on the work of environmental conservation in
the U.S., his conclusion that the role a worker plays in an organization dictates much
of the incentives he or she has to collaborate with others is important for this study. In
Thomas’s research, managers were more concerned with agency stability and,
therefore, were predominantly motivated to enhance their autonomy. On the other
hand professional staff were more interested in solving specific problems and using
the best management practices of their field. Finally, field staff were more interested
in locally based interests since they live and work in the local communities (Thomas,
2003). He also suggests that field staff are more likely to support cooperation when it
enhanced the socioeconomic condition of their local communities (Thomas, 2003).
94
3.3.1 WISDOM’s Impact in the Community
The main responsibilities of the WVE community workers consisted of identifying
children for sponsorship of aid and monitoring the well-being of the children in the
program and their families. See Chapter One for a description of the general role of
the World Vision community worker. Like the field staff in Thomas’s study, WVE
community workers also worked closely with local community members and actually
lived in the villages they serviced. Furthermore, WVE community workers also
tended to interpret a situation in relation to how it affected the community and its
members. While all six of the WVE community workers interviewed made some sort
of observation during interviews about how WISDOM was impacting the community,
four of the six subjects did so in response to a question asking them to describe their
own relationship with WISDOM. Thus, the WVE community worker’s own
evaluation of their relationship with WISDOM tended to be influenced significantly
by their impression of how WISDOM was helping or hurting the community
members.
Two WVE community workers had only positive remarks concerning WISDOM’s
impact on community members while one WVE community worker only had
negative remarks. In addition, three WVE community workers commented on both
the positive and negative effects of WISDOM on community members. Positive
95
effects that were observed among WVE community members included how
WISDOM helped those in the community who wanted to work and strengthen their
household security as well as specific anecdotal examples of clients who were slowly
improving in their circumstances after borrowing from WISDOM. Negative impacts
in the community included long application processes that would often discourage
community members and community members having to leave their villages because
of having defaulted on a loan.
Consequently, in instances where the WVE community worker only commented on
the positive effects of WISDOM, the WVE worker also assessed his relationship with
WISDOM as either positive or very positive. Where the WVE community worker
only had negative remarks concerning WISDOM’s impact in the community, the
relationship assessed more negatively. See Table 10 for a summary of factors and
indicators concerning motivation to collaborate among WVE community workers.
Furthermore, where the WVE community worker only discussed the positive effects
of WISDOM, involvement levels were high as well, either organizing groups or
following-up with clients. Conversely, in the case where only negative effects were
mentioned, the WVE community worker involvement level was lowest at referring
clients to WISDOM.
96
Table 10: WVE Community Worker (CW) Motivation to Collaborate
Shebraden
WVE CW
Sheremo
WVE CW
Yegobet
WVE CW
Wasamer
WVE CW
Dimtu
WVE CW
Belabela
WVE CW
Impact on
Community
Sees both
positive
and
negative
effects
Sees both
positive
and
negative
effects
Sees
positive
effects
Sees both
positive
and
negative
effects
Sees
positive
effects
Sees
negative
effects
Organizational
Identity
Separate
Identified
together
Strong
relationship
Identified
together
Identified
together
Target
Population
Mutual
clients
Mutual
clients
Different
Mutual
clients
Mutual
clients
Different
Necessary to
Each Other
Can help
WISDOM
Can help
WISDOM
Improves
my perf
Improves
my perf
Can help
WISDOM
Worker
Satisfaction
Needs
improve
Positive
Positive
Very
positive
Very
positive
Needs
improve
Worker
Involvement
Organizes
Follow-up
Organizes
Organizes
Follow-up
Refer
Nevertheless, where the WVE community worker observed both positive and
negative effects of WISDOM in the community, other factors seemed to influence the
levels of satisfaction that the WVE community worker had concerning his or her
relationship with WISDOM. Namely, how the WVE community worker understood
the relationship between the two organizations seemed to be significant as well.
There was a range of responses among WVE community workers on this matter.
Three of the WVE community workers commented that WISDOM and WVE were
97
often identified together, one WVE community worker responded that the two had a
strong relationship to each other, and another WVE community worker indicated that
the two were separate emphasizing he did not work for WISDOM but for the sake of
the community. Furthermore, it was the latter of the WVE community workers who
emphasized the separateness of the two organizations who also evaluated his own
relationship with WISDOM as needing improvement. Nevertheless, he was still
involved in organizing groups for WISDOM.
3.3.2 The Role of the Collaborating Manager
It is in establishing an understanding of the relationship between the two
organizations where the role of the WVE ADP manager proved significant. Only two
WVE ADP managers were interviewed since Guraghe and Wonchi were the only
ADPs included in the study. The two WVE ADP managers had fairly different
approaches to collaborating with WISDOM. The WVE ADP manager in Wonchi
saw WISDOM’s activities very much in line with what WVE was doing. When
asked to comment on his relationship with WISDOM, he responded, “The ADP has
the goal to provide household security. One of the goals is carried out through work
with MFIs such as WISDOM…WISDOM is so important to us. They are a Christian
organization and we have common goals”. In addition, the WVE community workers
in Wonchi both commented that the ADP manager frequently vocalized how
WISDOM was important to WVE during staff meetings.
98
Conversely, the Guraghe ADP manager had a different approach towards
collaborating with WISDOM. The subject was actually the ADP Program Officer
who was acting as the interim ADP manager. His response when asked to comment
on his relationship with WISDOM was, “The ADP staff is very tight and busy. We
don’t force them to cooperate with other organizations such as WISDOM”. He did
add that the WVE community workers are told to link community members to
WISDOM. Consequently, perceptions among WVE community workers in Guraghe
were varied concerning the relationship between WISDOM and WVE. One WVE
community worker saw the two organizations as separate, another saw them as
having a strong relationship, another saw them as being identified together, and still
another had no comment concerning the relationship between the two at all.
Thus, the role of the manager ought to be included when considering how to motivate
community workers on the individual-level to collaborate with an MFI. In the case of
the Guraghe and Wonchi ADPs, whether explicitly verbalized or not, WVE
community workers took the cue from their manager with regard to how they ought to
think about WISDOM. In this sense, the ADP manager can act to “frame” the
collaboration by creating meaning for the WVE community workers in how they
should understand and prioritize the relationship between the two organizations. This
act of framing is one example of the sorts of managerial behaviors researchers have
99
argued are needed for managing collaboration between organizations (Agranoff and
McGuire, 2001; McGuire, 2002).
In encouraging the WVE community workers in Wonchi to see WISDOM as a part of
WVE, the ADP manager is working to build trust between the two organizations. In
addition to creating meaning and working to build trust for his workers, the Wonchi
ADP manager demonstrated other skills found to be important for collaborative
management. In particular, the Wonchi ADP manager also suggested that WISDOM
needed to make sure that clients were able to repay their loans. He responded,
“WISDOM should be sure that money that is dispersed comes back so that they can
survive as an organization…For WISDOM it is important that OVC households can
pay money back”. This stands in contrast to the Guraghe ADP manager who
recommended that WISDOM use their resources to better serve vulnerable
households and find mechanisms other than WVE to connect to the community.
Thus, the Wonchi ADP manager appeared to have greater empathy for the financial
expectations placed on WISDOM than the Guraghe ADP manager. Finding
collaborative managers who are able to empathize with the financial constraints of the
MFI may be an important piece to successful collaboration for the MFI, especially
given the dual sets of expectations it faces as an organization.
100
3.3.3 The MFI as an Organization in Tension
The microfinance industry finds itself in a unique category of organizations where
there are expectations of both financial and social performance. Popularized by a
Nobel Peace Prize and promoted by the UN in its Millennium Development Goals
(MDG), much of its support is based in its social mission – to empower the poor by
providing access to financial services. Its primary goal is not to increase the values of
its shares but to instead ensure the sustainability of its services to its clients.
Although WISDOM was established as a share company, it also created bylaws based
on the principle of non-distribution where shareholders are prohibited from receiving
dividends or transferring shares. Thus, it is its social mission and non-distribution
policy among shareholders that puts MFIs in a separate category from traditional
banks and into the category of “social businesses”, a category coined by Yunus
(2007).
Furthermore, it is the unique mixture of grants from donors and sales for its services
as funding sources that also puts MFIs in a separate category from traditional nonprofit organizations. As described in Figure 2 in Chapter One, WISDOM’s funding
streams include one-time grants provided by WVE in addition to interest payments
made by clients for the cost of borrowing. Using Hansmann (1987, p. 28)’s four-way
categorization of nonprofit firms, MFIs generally fall somewhere in between
“donative” and “commercial” types of non-profits, since substantial portions of their
101
income come from both types of sources. See Table 11 below for Hansmann’s
categorizations.
Table 11: Hansmann's Four-Way Categorization of Nonprofit Firms
Mutual
Entrepreneurial
Donative
Common Cause
National Audubon Society
Political Clubs
CARE
March of Dimes
Art Museums
Commercial
American Automobile Assoc.
Consumers Union
Country clubs
National Geographic Society
Educational Testing Service
Hospitals
Nursing homes
Source: Hansmann, 1987, p. 28.
Charity organizations, such as WVE, fall into the category of donative/entrepreneurial
nonprofit firms from Table 11 and are primarily responsible for demonstrating
stewardship to donors. Nevertheless, for organizations that fall somewhere between
the spectrum of donative and commercial from Table 11, such as WISDOM, there is
an additional level of accountability. The MFI is also expected to demonstrate its
ability to recover the cost of lending to its clients through interest payments collected
from its clients. Most especially after the microfinance schism, MFIs have
increasingly faced the expectation that they are able to make enough profit to at least
finance their operational costs. Ethiopian policy limiting microfinance operations to
share companies is evidence of such profit-making standards as well. Table 12
provides a comparative summary of the different organizational types. Organizations
102
like WVE fall into the category of nonprofit charities while organizations like
WISDOM are described by the nonprofit business category.
Table 12: A Comparative Summary of Organizational Types
For-Profit
Corporations
Nonprofit
Businesses
Nonprofit
Charities
Primary Goal
Maximize profits and
shareholders value.
Provide ongoing social
benefit to target
population.
Provide ongoing social
benefit to target
population.
Resource Provision
Profits from an
exchange of goods or
services for money.
Mixture of profits,
grants, and donations.
Mixture of grants, and
donations.
Standards
Shareholders:
Profitability
Donors:
Ability to recover
operational costs.
Donors:
Stewardship of
donated resources for
social mission.
Customers:
Satisfaction
Customers:
Satisfaction
Consequently, as MFIs are pressured into designing its products and services to
ensure enough profit, it also needs to be wary of making too much profit lest it be
perceived as taking advantage of the poor. Furthermore, other organizations that do
not face the same sort of financial standards may interpret the MFI’s intention of
making profit as making it less committed to its social mission, creating room for
distrust between the two organizations. Thus, potential difficulties may arise when
different types of nonprofits with different accountability principles collaborate
together.
103
Table 13: Organizational Summary
MISSION STATEMENTS
•
World Vision
Ethiopia
(WVE)
•
“A Christian relief, development
and advocacy organisation
dedicated to working with
children, families and
communities to overcome
poverty and injustice.”
“…dedicated to working with
the world’s most vulnerable
people”
PROGRAMS
•
•
•
•
•
•
WISDOM
Microfinance
Institution
“…to improve the economic,
social and moral well being of
the productive poor through
quality financial and non
financial services.”
•
•
Child-sponsorship
management
Development programs,
such as agricultural, IGA,
and microfinance
Health services, such as
HIV/AIDS programs
Infrastructure programs,
such as building wells
Works in partnerships with
other organizations
Promotions to help
community learn about
microfinance
Credit and saving services
to the community
Note: WVI’s mission statement was taken from the organization’s website (WVI, n.d.).
WISDOM’s mission statement was taken from the MIX Market website (Mix Market, n.d.).
In the case of WISDOM and WVE, indicators of potential distrust between the two
organizations came in the form of the recommendations that WVE community
workers had for WISDOM concerning their target population. Of the 10 WVE
workers interviewed, including both managers and community workers, half of them
explicitly commented on the differences in target populations or activities between
the two organizations. WVE workers mentioned that WISDOM was for the “able
poor” while WVE was more for the “poorest of the poor”. The differences are visible
in the mission statements of both organizations as well, as shown in Table 13. WVE
104
primarily aims to help the vulnerable in the community and WISDOM aims to help
those that are productive. With regards to different activities, one WVE ADP
HIV/AIDS staff commented that WISDOM was in the “commercial area” making it
different from the work that he did. Furthermore, after commenting on such
differences, three WVE workers suggested that WISDOM ought to do more to help
the poorest of the poor in the community while only one worker, the Wonchi ADP
manager, expressed sympathy for the financial pressures that WISDOM faced.
Thus, whether it is called a “social business” or a “commercial non-profit”, the MFI
exists as a bit of a hybrid organization facing both financial and social performance
standards. Furthermore, it is a category where few of its collaborators will be able to
empathize completely with its dual objectives of financial sales and social benefit.
Nevertheless, where the MFI is able to find empathetic collaborators who can
understand the pressures placed on the MFI without allowing them to create potential
distrust between the two, it will improve the MFI’s chances of successful
collaboration with other community organizations.
Further research beyond this study is needed to determine whether MFIs are
susceptible to greater distrust from other community organizations because of its
profit-making agendas. However, there is evidence that confusion exists even within
the management of MFIs concerning their relationship to other nonprofits. During
105
interviews, one WISDOM branch manager speculated that collaboration between the
two organizations might be weaker because of WISDOM’s own fear of being so
closely associated with a charity organization, such as WVE. Do the financial
pressures on the MFI make it more wary of collaborating with other community
organizations? Do other nonprofits find it more difficult to trust and have empathy
for the constraints that the MFI faces? The answers to these questions have important
implications for the managerial strategies MFIs will have as they seek to collaborate
with others.
3.4
Conclusion
Collaboration has the potential to produce outcomes that positively impact both the
processes and products of the organizations involved. Outcomes of collaboration, to
name a few, include increased levels of knowledge and social capital, shared mental
models and innovative ideas, joint action in program implementation, and ultimately
the evolution of new institutions and rules (Innes and Booher, 1999). Nevertheless,
the process of getting to these outcomes is never a given. Consideration ought to be
given to the factors that will facilitate and obfuscate successful collaboration for the
parties involved.
While the organizational literature focusing on collaboration is vast, this chapter
specifically explores some of the major factors influencing individual-level
106
motivation in interagency collaboration. In the case of WISDOM and WVE, the
responses workers gave in interviews were very much in line with Thomas’s
observations. How each subject perceived and evaluated their relationship with the
other organization was very much dictated by the role that the subject played in the
organization.
Further building on Thomas (2003)’s research, this study also considers the role that
managers can play to help create shared meaning, build trust, and practice empathy in
collaborative relationships. By comparing the impact of the Guraghe and Wonchi
ADP managers on WVE community workers’ perceptions concerning the
organizational identities of WISDOM and WVE, this research suggests that the
abilities of a manager can be an important factor in motivating community workers to
collaborate with MFIs.
While the assessment of the WVE community worker concerning WISDOM’s impact
on the community still appeared to be the primary factor in motivation for
collaborating with WISDOM, where assessments appear mixed, the WVE manager’s
ability to create a shared sense of identity with WISDOM seemed significant.
Furthermore, with the dual expectations for financial and social performance placed
on the MFI, the room for confusion and distrust between the two organizations
seemed greater than in other circumstances. As a result, the role of the manager to
107
create an environment of empathy may go a long way in building trust between the
two collaborating organizations.
Thus, as MFIs consider how to increase the motivation among community workers
from organizations to collaborate with them, demonstrating the positive impact of its
services to the community workers is extremely important. Doing so will create
increased satisfaction and motivation among community workers to collaborate with
MFIs since their primary interests and goals concern the well-being and
socioeconomic conditions of their community members. Nevertheless, finding
effective managers who can advocate for collaboration with the MFI by creating a
shared sense of identity and demonstrating feelings of empathy will also be important
factors for the MFI to consider. Furthermore, it is perhaps only with these managerial
skills that the MFI will be able to combat the potential for confusion and distrust that
its unique category of an organization brings.
108
CHAPTER FOUR: MITIGATING THE NEGATIVE EFFECTS OF
LOSS FOR OVCs WITH MICROFINANCE
By the end of 2003, it was estimated that there were 143 million orphans6 in the
regions of sub-Saharan Africa, Asia, Latin America, and the Caribbean (USAID,
2004, p. 7). With the developmental gains of recent decades, one would have
expected the percentage of children who are orphans to decline. Nevertheless, with
the emergence of the HIV/AIDS pandemic, the number continues to rise globally. In
2003, agencies estimated that more than 16 million children were orphaned with 5.2
million coming from sub-Saharan Africa alone. Furthermore, with AIDS thought to
be one of the leading causes of double orphans7, sub-Saharan Africa has nearly the
same number of double orphans as the entire continent of Asia even though it has
about a quarter of the number of children.
Experts estimate that more than 90 percent of all orphans are cared for by extended
family members, often living in female-headed or grandparent households (USAID,
2004). This undoubtedly causes additional burdens as studies from several countries
have shown income levels to be 20 to 30 percent lower in orphan households than
non-orphan households (Foster & Williamson, 2000, p. S282). Such trends are
alarming; as it is believed that up to 20 percent of all households in some African
countries are caring for one or more orphans (USAID, 2004).
6
7
Ages 0 to 18 years of age.
Children who have lost both the mother and father.
109
This chapter examines the specific areas of vulnerability that orphans and vulnerable
children (OVCs) can experience after the death of a parent and considers how
effective microfinance can be in reducing these areas of vulnerability over the longterm. First, the chapter presents an overview of the literature involving areas of
vulnerability for OVCs. Then, using data from case study interviews conducted with
OVC caregivers in the Guraghe and Wonchi ADPs, two major areas of vulnerability
are identified for households in this region shortly after the loss of one or both parents
of the OVC. Second, this chapter examines how these areas of vulnerability can be
reduced through microfinance programs for the OVC caregiver through several
different analytical models. Both household surveys and performance data are used
to analyze the progress that OVC caregivers made for their household during their
participation in microfinance programs with Wisdom Micro Financing Institution S.
C. (WISDOM).
4.1
Identifying Areas of Vulnerability for the OVC
Living with a person with chronic illness, such as HIV/AIDS, can be difficult for
children especially. Because children are the most dependent members of a
household they are also the most vulnerable to adverse changes in the household.
Preliminary research has shown the OVCs are particularly susceptible when a family
member is chronically ill because of the changes in the distribution of resources and
responsibilities that generally occur (Foster & Williamson, 2000, p. S275). Food,
110
funds, and care are now needed for the urgent needs of the sick family member who
now is too ill to work or perform other household duties.
While it has been recognized that there needs to be a greater understanding of this
phenomenon (Foster & Williamson, 2000, p. S275), initial research has pointed to
negative economic and psychosocial effects for children due to resources being
shifted away from the child and to person who is ill in the household. In addition,
responsibilities tend to be shifted towards the child and away from the PLWHA
(Foster & Williamson, 2000; Shetty & Powell, 2005). New responsibilities for the
child can include: “cooking, cleaning, carrying water and laundry, care giving
activities such as feeding, bathing, toileting, giving medication and accompanying
relatives for treatment, agricultural or income generating activities and childcare
duties” (Foster & Williamson, 2000, p. S278).
Furthermore, psychosocial effects have also been observed on children with a parent
who is positively infected or chronically ill. A study in Uganda showed that children
were more likely to feel hopeless or angry while their parents were sick (Sengendo &
Nambi, 1997; Foster & Williamson, 2000, p. S282). In Zambia, one study showed
descriptions of children being “worried, sad, tried to help in the home and stopped
playing to stay nearby”. In addition, children were described to be “more likely to
111
become solitary, appear to be miserable or distressed and be fearful of new situations”
(Foster & Williamson, 2000, p. S276).
Once a child has had one or both parents die from AIDS, the long-term effects of a
loss of home and income need to be considered. The loss of a parent, the changes in
caretakers and family composition, and the stigma of having orphan status can all
negatively affect the child’s psychosocial well-being (Richter, 2006, p. 9-12).
Researchers have identified that “stigmatization, dropping out of school, changed
friends, increased workload, discrimination, and social isolation of orphans all
increase the stress and trauma of parental death” (Foster & Williamson, 2000, p. 282).
See Figure 6 below for the impact of HIV/AIDS on OVC well-being.
Overall, being orphaned leaves a child more susceptible to malnutrition, poor health,
and early death than non-orphans (Shetty & Powell, 2003, p. 7, 25; UNICEF, 2007, p.
2). Studies have pointed to the reality that children are more likely to dropout of
school because of a lack of resources for school (such as books, uniforms, and
textbooks) or due to the need for another income in the household. This is
particularly true for the girl orphan (Foster & Williamson, 2000; Shetty & Powell,
2003). Additionally, orphans also tend to experience a lowering of expectations for
their future, such as future family and future job, as their self-esteem tends to suffer.
Sociability, however, may not be as affected (Foster & Williamson, 2000, p. S282).
112
113
(Adapted from Lippincott Williams & Wilkins, 2000.)
Figure 6: Factors in OVC Well-Being
4.1.1 Negative Effects of Orphanhood on Children
A total of eight OVC caregivers were interviewed in the Guraghe and Wonchi ADPs
for this study. See Chapter One for a description of the two ADPs. The OVC
caregivers ranged from widowed parents to grandparents to siblings. Case study
interviews focused on major changes that occurred in the household following the
death of one or both parents of the OVC. See Appendix B for interview protocols.
Content analysis of the interviews revealed several common occurrences among
households shortly after the loss of a parent. In particular, two areas of vulnerability
were apparent after interviews. First, shortly after the loss of one or both parents, all
OVC caregivers described the additional responsibilities as very burdensome,
particularly financially. For OVC caregivers who were widowed, husbands felt
burdened by the additional household and childcare responsibilities while wives
experienced the loss of both income and income generating knowledge. As a result
of these new burdens, half of the OVC households interviewed withdrew the eldest
child from school since they were not able to pay for educational materials. Often the
eldest child would also help with tasks on the farm or at home. In one particular case,
the OVC caregiver was 17 years old when she and her husband took in her husband’s
three siblings after the death of both their parents. As a result she was unable to
finish her school as well.
114
Second, circumstances seemed the most difficult for women who had been widowed.
Two factors contributed to their unique difficulties. First, the loss of income for a
widowed wife is generally larger than the loss of income for a widowed husband. In
particular for those with livestock, often the loss of the husband also meant a loss of
knowledge of how to care for the livestock in the household. One widowed wife
described how she lost their three milking cows to disease after their husband passed
away. Second, it is often harder for a woman to remarry after having been widowed
with children than for a man to remarry in the same circumstance. Both of the
widowed husbands interviewed were able to remarry while none of the widowed
wives interviewed were able to remarry. Furthermore, both widowed husbands
described how circumstances improved after they remarried, as their new wives were
able to take care of the children and other household responsibilities. Conversely, for
the widowed wives, losing their husbands required them to find new sources of
income. One widowed wife even sent her children to their paternal grandparents
while she went to live with her own parents after her husband’s death because the
financial burden was too great for her.
The importance of strengthening household capacities is made evident by the
vulnerabilities that households were exposed to after the loss of one or both of the
OVC’s parents. In the Guraghe and Wonchi ADPs, households often adjusted to the
additional financial burden of their loss by withdrawing the eldest children from
115
school. Women who had been widowed in this region were faced with the biggest
losses in their households and seemed to have the hardest time adjusting financially.
In addition to these patterns, subjects also recounted how they often sold livestock
assets to pay for medical care for sick household members or educational materials
for OVCs. One grandfather described how he sold both of his oxen to provide for
medical care for his daughter after her husband had died. After the death of his
daughter, he ended up caring for her three children with their grandmother. Patterns
such as these can have long-term consequences that have multiplicative effects for
both families and communities. Thus, finding interventions that can strengthen
households and help mitigate these new vulnerabilities can prevent a burdensome
inheritance for the next generation.
4.2
The Hope of Microfinance: Mitigating the Negative Effects of
Loss for Orphans and Vulnerable Children
As aunts, uncles, grandparents, and even siblings often act as caregivers for orphans
after a parent has passed away, the extended family plays an important role as a safety
net for OVCs in the community. Despite the difficulties that OVC households can
experience as they cope with loss and addition burdens, the extended family can often
offer the sort of familiarity and stability that the OVC needs as he or she adjusts to the
changes and loss in his or her young life.
116
Therefore, many policy advisors have suggested interventions that strengthen these
“safety nets of families” at the household level (USAID, 2004, p. 5). The success and
popularity of microfinance to alleviate poverty and empower individuals has led to
hopes that it might also be used as a tool to mitigate the negative effects of loss for
OVCs. While producing positive economic benefits, microfinance can also lead
positive psychological benefits for the borrower. One study argued that participation
and success in microfinance led to a greater sense of autonomy and resilience for
borrowers (Cheston & Kuhn, 2002; Pronk et al., 2005, p. 29). Furthermore, if
microfinance can produce positive psychological impacts in the borrower, might it
also have a positive effect on the OVC for whom the borrower is caring?
The following section explores these questions and issues by studying the impact that
participating in WISDOM had for OVC caregivers and their households. Household
surveys were based on the Community REACH instrument that was constructed and
used for a USAID research project on the well-being of children affected by
HIV/AIDS in Zambia and Rwanda. Four of the original five parts of the survey
instrument were used during data collection. The four parts included: 1) primary
caregiver questionnaire, 2) primary caregiver questionnaire concerning the OVC of 6
to 12 years old, 3) OVC questionnaire for children 6 to 12 year olds, and 4) OVC
questionnaire for adolescents 13 to 19 year olds. See Chapter One for more
117
information on data collection methods. See Appendix C to F for the household
survey protocols.
Household surveys from 162 OVC households in the Guraghe ADP and 116 OVC
households in the Wonchi ADP were included in the overall sample of data. Of the
OVC caregivers that provided their age, 20.3% of the sample was 15 to 24 years old,
55.3% were 25 to 49 years old, and 24.4% were 50 years old and above.
Furthermore, 39.1% of the OVC caregivers were married, 1.4% were separated or
divorced, 54.3% were widowed, and 5.1% were never married. The top two
occupations were farmer at 68.3% and professional or civil servant at 10.0%. Other
occupations included: petty trader, general trader, artisan, retired, and other. 8.9% of
the OVC caregivers surveyed were not currently working and 24.5% had attended
school. The average OVC household was 5.38 household members. OVC
households in the Wonchi ADP were slightly larger than in the Guraghe ADP at 5.78
and 5.08 members respectively. Both ADPs had slightly higher average male
children than female children and slightly higher female adults than male adults.
Surveys with 95 OVC adolescents and 119 OVC children were also included in the
overall sample of data. Paternal orphans were the highest proportion of orphans at
66.7%. There were higher percentages of maternal orphans (25.9%) and lower
percentages of paternal orphans (60.3%) in the Wonchi ADP than in the Guraghe
118
ADP. In the Guraghe ADP, 11.9% of the OVCs were maternal orphans and 69.4%
were paternal orphans. In addition, the Guraghe ADP had higher percentages of
OVCs that had lost both parents at 18.7%. See Appendix H and I for more descriptive
statistics of the two ADPs.
3.2.1 Measuring the Impact of Microfinance for OVC Households
This section provides a description of the first approach used to measure the impact of
microfinance in OVC households. A cross-sectional analysis of household outcomes
between OVC household who were participating in WISDOM loans and OVC
households who were not was conducted. Only OVC caregivers who lived in villages
where WISDOM was available were included the sample for this approach.
A simple model of the impact of microfinance on household expenditures was
constructed with the following regression equation:
Yij = Xijα + Hijβ + Vjγ + Mijδ + ε ij
Yij is the household-level outcome for the household i and the village j. Variables Xij
include dummy variables for OVC caregiver characteristics, such as OVC caregiver
occupation, OVC caregiver marital status, and OVC caregiver’s participation in
income generation activities (IGA). Variables Hij include household characteristics,
119
which include the number of male and female adults and the number of male and
female adolescents and minors in the household. Other dummy variables are
included as well, such as the percentage of children in the household who participate
in WVE’s Children in Program, whether or not a household receives or sends money
from other households, and whether there is a family member who is chronically ill in
the household. Vj captures village level characteristics, which is a dummy variable
that indicate the ADP of the household, and εij represents the error term in the
regression. Mij is the most important variable and indicated whether the OVC
caregiver of the household is participating in WISDOM microloans. Table 14 shows
the results of the regression using ordinary least squares for household livestock
assets and average meals per day per member of the household.
Average meals were calculated by totaling the number of meals consumed by the
household for the day before the survey was conducted and dividing this total by the
number of members in the household. Livestock asset scores were calculated by
surveying OVC caregivers on the number of several different types of livestock, such
as cows and goats, which they owned. Prices from the USAID Pastoralist
120
Table 14: Impact of Microfinance on OVC Household Consumption
Avg
Meals
Livestock
Assets
Household
Assets FS
Percent
In School
0.215**
0.004
-0.299**
-0.165*
-0.318***
-0.528***
-0.050
-0.309***
# of M Adults
# of F Adults
# of M Children
# of F Children
0.082
-0.159*
0.028
0.068
0.017
0.109
-0.122
0.188**
-0.115
-0.125
-0.254
-0.225
0.126
0.0145*
0.129
0.139*
Occupation (0 = Not Currently Working)
Farmer
Petty Trading
Trader
Artisan
Professional/Civil Servant
Retiree
Other
0.128
-0.043
0.067
0.140*
-0.021
0.087
0.057
0.051
-0.082
0.297***
0.100
-0.209**
0.159
-0.066
-0.049
---0.317*
--0.072
--0.378*
0.181
-.001
-0.001
0.052
0.084
0.084
0.070
-0.159**
0.022
-0.016
0.134
-0.124
-0.124
0.028
-0.120
-0.016
-0.015
-0.392**
0.179
-0.093
0.241**
0.075
0.118
Percent of School Aged Children in CIP
Receives Money?
Send Money?
Chronic Illness in HH?
-0.255***
0.048
0.008
0.013
0.140*
-0.50
0.084
-0.020
-0.165
-------
0.161*
0.009
0.006
0.024
F-stat
Adjusted R2
Observations
2.541***
0.164
166
2.983***
0.195
173
1.919
0.169
69
1.956**
0.114
157
WISDOM Client (0 = No, 1 = Yes)
ADP (0 = Guraghe, 1 = Wonchi)
IGA Training
Marital Status (0 = Married)
Widowed
Never Married
Separated
Note: All entries are standardized coefficients. All VIF scores tested below or equal to 7.974.
*** p < 0.01; ** p < 0.05; * p < 0.10;
Livelihoods Initiative Market Monitoring Bulletin were used for some of the
livestock prices (USAID, 2008). NGO website used to approximate the market price
121
of the remaining livestock that were not included in the bulletin. See the Appendix G
for more on how livestock assets were calculated. Household assets scores were
based on a series of questions concerning what assets the OVC caregiver, adolescent,
and child possessed. Data was then tested for internal consistency for a Cronbach
alpha score greater than 0.70 and factor scores were calculated using the principal
components method with a varimax rotation. Percent in school was calculated by
dividing the number of children and adolescents in the household who were currently
attending school by the total number of school-aged children and adolescents in the
household.
Overall, the model was moderately able to explain the variance in the two household
indicators with adjusted R2 values ranging from 0.114 to 0.195. The difficulty
impact studies have had with microfinance is separating out its causal role in its
outcome variables. Simply comparing households who are not participating in
microfinance with those that do tends to overestimate the impact of microfinance
because of endogeneity issues that can exists (McKernan, 2002). This can be an issue
especially because clients self-select themselves into lending groups. Thus, there
may be an “entrepreneurial factor” that is not accounted for in impact studies that
simply compare non-participants to participants.
122
For example, when measuring the impact of microfinance on household income,
microfinance may be one of many factors that contribute to the outcome.
Furthermore, the same variables that contribute to household income, such as
entrepreneurship or determination, may also determine a client’s decision to
participate in microfinance. Thus, microfinance participation is an endogenous
variable that may be correlated to other individual and household characteristics that
cause multicollinearity issues in the model.
3.2.2 Managing Endogeneity
One common approach in dealing with the endogeneity problem is to create an
instrumental variable that models a community member’s likelihood of participating
in microfinance. The difficulty with this, however, is finding variables that influence
the decision to participate in microfinance but do not influence any of the other
variables in the model. Thus, creating an instrumental variable for microfinance
participation that was exogenous to individual and household characteristics was very
difficult.
Therefore, the second approach used in this study was similar to the one used by
Coleman (1999) and (2002) in his research of microfinance programs in Northeast
Thailand. Coleman uses a research design where he analyzes data from both clients
who were ongoing participants of microfinance and clients who had decided to
123
participate in microfinance but had not yet received their loans. Comparing
household outcome data among a sample of community members who had all
decided to participate in microfinance yet who were not all yet participating in
microfinance controlled for selection bias, he argued (de Aghion & Morduch, 2005).
Like Coleman’s model, the second approach I used was to compare “new borrowers”
to “old borrowers”. Including only WISDOM borrowers in the sample controlled for
selection bias that may exist between clients and non-clients. New borrowers
included clients who had participated in microfinance for less than a year while old
borrowers included clients how had participated for more than a year. Although,
unlike Coleman’s model, the clients included in the new borrowers group were
already participating in microfinance, their participation was still minimal at less than
a year. This means that they had not yet fully completed one loan cycle.
Furthermore, the greatest benefits of microfinance are also thought to occur after the
first year or two of participation.
Household Indicators
Thus, an adjusted model for the impact of microfinance on OVC households was
constructed using new and old borrowers with the following regression equation:
Yij = Xijα + Hijβ + Vjγ + Tijδ + ε ij
124
The adjustment made in this model replaces Mij with Tij where Tij is a dummy
variable indicating whether a borrower has participated in WISDOM for one year or
more (0 = less than a year; 1 = greater than or equal to a year). Table 15 shows the
results of the adjusted regression model using ordinary least squares on household
indicators. In addition, because the Wonchi ADP only had one OVC caregiver how
had participate in WISDOM microloans for less than a year, only borrowers from the
Guraghe ADP were included in this analysis.
The second approach comparing newer and older borrowers was able to explain a
higher percentage of variance in the dependent variables than the first approach that
was used. The weakest results were for the household assets score with an adjusted
R2 value of 0.137. This may be attributed to the low number of observations for that
particular indicator as well. Nevertheless, the other three indicators had stronger
adjusted R2 values, ranging from 0.262 to 0.491.
Overall, the data concerning household consumption was a bit mixed with longer
participation in WISDOM having positive impacts on average meals per day per
household member (0.302) and the percentage of children in the household attending
school (0.164). OVC caregivers how had participated in WISDOM for more than a
year averaged 2.44 meals per household member per day while OVC caregivers who
had participated for less than a year had an average of 2.29 meals per household
125
Table 15: Impact of Microfinance on OVC Household Consumption
Avg
Meals
Livestock
Assets
Household
Assets FS
Percent
in School
Client Longer Than 1 Yr
(0 = No, 1 = Yes)
0.302*
-0.292*
-0.561**
0.164
# M Adults in HH
0.404**
0.189
-0.057
0.014
# F Adults in HH
-0.120
0.173
0.336
0.298
# M Children in HH
-0.206
-0.234
-0.487*
0.091
# F Children in HH
0.149
0.217
-0.084
0.199
Occupation: (0 = Not Working)
Artisan
0.137
-0.103
-----
0.052
Farmer
0.418
0.535*
-----
-0.235
Trader
0.381
-0.385
-0.374
-0.042
Professional/Civil Servant
0.101
-0.027
0.251
-0.475*
Other
0.047
0.092
0.462
-0.063
0.216
0.172
-----
0.156
Widowed
0.233
0.219
-0.076
0.451*
Never Married
0.064
0.215
0.382
0.009
Avg % of School-Aged Children CIP
-0.042
-0.124
-0.072
-0.160
Receives Money
-0.025
-0.237
-----
-0.666***
Sends Money
-0.370
0.102
-----
0.119
Chronic Illness in HH
-0.022
0.008
-----
-0.365*
F-stat
1.961
2.507
1.317
3.098
0.054
0.013
0.328
0.009
0.262
0.353
0.137
0.491
47
48
23
38
IGA Training
Marital Status (0 = Married)
Sig
2
Adjusted R
N
Note: Standardized coefficients given. Newer clients participated in microfinance for less
than a year and older clients participated in microfinance for at least one year and up to six
years. All VIF scores tested below or equal to 6.849.
*** p < 0.01; ** p < 0.05; * p < 0.10;
126
member per day. In addition, older borrowers had higher percentages of school-aged
attending school at 92.2% compared to newer borrowers at 83.5%. The number of
male adults in the household seemed to influence the average meals positively at
0.404 and was significant at the 1% level while the number of women adults in the
household influenced the percentage of children in school positively at 0.298.
With regards to both livestock and household assets, however, longer participation in
WISDOM appeared to actually have negative effects with -0.292 and -0.561
respectively. One possible explanation may be that OVC caregivers who had
participated in WISDOM longer than a year seemed to be relying on less assistance
than OVC caregivers who were newer borrowers. Older borrowers had lower
percentages of children in their household who were CIP’s in WVE child sponsorship
program compared to newer borrowers at 36.5% and 43.4% respectively. In addition,
older borrowers were also more likely to have paid their last medical bill themselves
at 83.3% compared to 73.3% among newer borrowers.
Furthermore, although longer participation in WISDOM may have had a negative
effect on the livestock and household assets of OVC caregivers, it also made an OVC
caregiver less to have sold assets to pay for basic needs within the past year.
Regressions were conducted on the impact of WISDOM participation on whether the
OVC caregiver sold their assets in the past year to pay for basic needs, such as food
127
and shelter. With an adjusted R2 of 0.266 for the model, the coefficient for the length
of participation in WISDOM was -0.580 and significant at the 1% level. See
Appendix J for a full table of results.
Tests were also conducted on the impact of microfinance on OVC caregivers’
community involvement using the newer and older borrowers approach. OVC
caregivers were asked if they attended meetings to discuss community concerns, such
as health and education problems. Regressions were conducted on whether the OVC
caregiver had attended a community meeting as the dependent variable. Again,
longer participation in WISDOM had a positive effect on whether an OVC caregiver
had attended a meeting. With an adjusted R2 of 0.198, the variable for length of
participation in WISDOM had a coefficient of 0.355 and was significant at the 1%
level. See Appendix J for a full table of results.
OVC Indicators
In addition to household indicators, the model was also used to estimate the impact of
microfinance on OVC psychosocial well-being. Four outcomes were tested,
including: 1) adolescent low self-esteem, 2) adolescent overburden/responsibility, 3)
child worry/stress, and 4) child overburden/responsibility. The four measures were
based on a series of answers to questions from the interview protocol that were
grouped in the four categories of outcomes and tested for internal consistency using
128
the Cronbach alpha score. The occurrences of when a subject agreed with a question
within each category were totaled and assigned as a score for that category. Lower
scores represent greater psychosocial well-being as higher scores indicate more
agreement with questions concerning distress, burden, worry, etc. A list of questions
included in each of the four categories can be found in the Appendix K. A fifth
category for adolescent locus of control outcomes was also tested but did not pass
internal consistency tests. Table 16 shows the results of the regressions conducted on
OVC psychosocial well-being using ordinary least squares.
Each regression performed was able to explain high percentages of variance (adjusted
R2) in the dependent variable and ranged from 0.374 to 0.467. Among adolescents,
the length of participation in WISDOM showed to be significant at the 5% level and
had the effect of decreasing the low self-esteem measures (-0.403). Other significant
factors on adolescent self-esteem were the presence of female adults in the household
(-0.597) and whether the OVC caregiver was a general trader by occupation (0.537).
Traders tend to spend time away from the household for long periods of time and had
an increasing effect on adolescent low self-esteem.
Concerning adolescent overburden and responsibility, the length of WISDOM
participation had a similar effect and was significant at the 5% level as well. Longer
participation in WISDOM had the impact of lowering adolescent
129
Table 16: Impact of Microfinance on OVC Psychosocial Well-being
Adol
Self-Esteem
Adol
Overburden
Child
Worry/Stress
Child
Overburden
-0.403**
-0.503**
-0.219
-0.206
# M Adults in HH
-0.231
-0.210
0.126
0.303*
# F Adults in HH
Client Longer Than 1 Yr
(0 = No, 1 = Yes)
-0.597***
-0.293
0.047
-0.206
# M Children in HH
0.050
-0.473**
0.087
0.051
# F Children in HH
0.026
0.530***
0.117
-0.050
Occupation:
(0 = Not Working)
Artisan
-----
-----
0.040
0.211
Farmer
-----
-----
0.089
0.483
Trader
0.537**
0.386*
0.265
0.759***
Professional/Civil Servant
0.388*
0.043
-0.310
-0.112
Other
-0.242
-0.268
-0.109
0.149
Widowed
-0.114
0.235
0.238
0.125
Never Married
-0.436*
-0.402*
-0.126
-0.236
% of Children in CIP
-0.177
-0.050
-0.268*
-0.100
Chronic Illness in HH
0.189
0.075
-0.020
0.053
Has Someone for Advice
-0.088
0.364*
-----
-----
Has Attended School
0.132
0.036
-----
-----
Regular Place to Sleep
-----
------
0.653***
0.491***
Total Meals a Day
-----
------
-0.282
-0.470**
2.663**
2.815**
2.456**
3.045***
0.445
0.467
0.374
0.456
30
30
40
40
Marital Status (0 = Married)
F-stat
2
Adjusted R
N
Notes: Standardized coefficients given. Newer clients participated in microfinance for less
than a year and older clients participated in microfinance for at least one year and up to six
years. All VIF scores tested below or equal to 6.667.
*** p < 0.01; ** p < 0.05; * p < 0.10;
130
overburden/responsibility measures (-0.503). The numbers of children in the
household were also significant factors with increasing numbers of male children
lowering overburden/responsibility measures while increasing numbers of female
children tended to increase the outcome.
Although the results for child worry/stress and child overburden/responsibility did not
find the length of participation in WISDOM to be significant, the coefficients for
WISDOM participation did have similar effects on the outcome measures as in
adolescents. The model did have a significant F statistic and relatively high adjusted
R2 scores at 0.374 and 0.4506 respectively.
Longer participation in WISDOM did tend to decrease the two measures although by
a lesser amount than in adolescents. The percentage of children in the household who
were WVE’s CIPs also had a similar impact (-0.268) as WISDOM participation (0.219) on child worry/stress. With regards to child overburden/responsibility,
patterns were similar to adolescent overburden/responsibility as the number of male
adults in the household and the OVC caregiver being a general trader by occupation
were important influences on the measure. In addition, the number of meals that a
child ate also played a significant role in decreasing a child’s sense of overburden and
responsibility. It was difficult to explain the role that having the variable of having a
131
regular place to sleep at night had for children as the variable surprisingly had the
effect of increasing both child worry/stress and child overburden/responsibility.
Overall, the models showed evidence that there is a positive impact on microfinance
participation for OVC psychosocial well-being as it tends to decrease elements of
burden, low self-esteem, and worry.
4.3
Discussion and Conclusion
While the previous chapters were devoted to the processes of working with others to
include more vulnerable groups in microfinance, this chapter shifts the focus to the
actual impact of microfinance on vulnerable household once they are included in the
lending groups. Identifying some of the key areas of vulnerability among OVCs and
their households, this chapter then addressed the question of how well microfinance
can mitigate the negative effects of loss for the OVC. Areas of vulnerability include
loss of income and income generating knowledge and selling assets to pay for basic
needs like food and shelter. In addition to financial hardship at the household level,
OVCs also face psychosocial hardships in their adjustments to life after their loss.
In order to estimate the impact of microfinance on OVC households, two approaches
were used. The first approach simply compared non-borrowers to borrowers to
examine the impact of microfinance on household consumption indicators.
132
Nevertheless, due to selection bias and endogeneity issues, a second approach was
used comparing newer borrowers to older ones.
4.3.1 Limitations and Other Approaches
Some have commented that Coleman’s model of “old borrowers” versus “new
borrowers” does not completely eliminate bias in the model. They note that the
sample of old borrowers still does not include those that have not continued with
microfinance or “dropped out” after their first loan cycle. While these reasons may
include success or failure, they nonetheless argue that there is still an attrition bias
present in the model (de Aghion & Morduch, 2005, p. 209).
While this may be the case, constructing a model measuring the impact of
microfinance without any bias is extremely difficult. Coleman’s model does address
the issue of selection bias concerning attributes that may cause a person to participate
in microfinance. Furthermore, this problem of selection bias for participation in
microfinance is likely to cause a greater distortion on the results than attrition bias
since it is unknown which direction the attrition bias is overestimating.
A third approach of analysis was also conducted by measuring the impact that
WISDOM had on an entire village. This model compared villages where WISDOM
was present to villages where WISDOM was not present. 151 OVC households were
133
included in the study. The results proved similar to the “new borrowers” versus “old
borrowers” approach, however, the impact of WISDOM appeared more diluted in the
outcomes examined. See Appendix L for results from the regression analyzed at the
village level.
4.3.2 Contributions of this Study
Several studies focused on microfinance have studied its impact on the financial and
household outcomes of its subjects. Research has shown indication of higher
employment rates, higher household incomes, and higher enterprise productivity in
association with participation in microfinance (Hulme & Mosely, 1996; Bhatt &
Tang, 2001b, p. 1115; Pitt & Khander, 1998; Schreiner, 2002b, p. 4). Little attention,
however, has focused on how participation in microfinance by the household
caregiver is affecting the well-being of the children. This study uses an approach that
compares newer borrower to older borrowers to measure the impact of microfinance
on general household outcomes, such as household consumption, whether assets
needed to be sold, and caregiver community involvement. However, the study goes a
step further to examine the impact of microfinance on the psychosocial well-being of
the OVCs in the household. While results concerning household consumption were a
bit mixed perhaps due to various assistance that OVC caregivers were receiving,
participating in WISDOM for longer than a year had a positive impact on the number
of meals the household consumed, the percentage of children in school, the OVC
134
caregiver not having to sell assets for basic needs, and the OVC caregiver’s
attendance at community meetings. In addition, longer participation in WISDOM
had the effect of decreasing low self-esteem and feelings of being overburdened
among adolescents and worry/stress and feelings of being overburdened among
children. Thus, when considering how the negative impact of loss might be mitigated
for OVCs, microfinance shows to be an effective intervention that helps address more
than just the financial vulnerabilities of a household.
135
CHAPTER FIVE: CONCLUSION
This dissertation considers some of the unique challenges and contributions of
microfinance in the context of providing microfinance services for households caring
for orphans and vulnerable children (OVCs) in the rural southwest region of Ethiopia.
Using data collected from field research with Wisdom Micro Financing Institution
(WISDOM) and World Vision Ethiopia (WVE) in August 2007 in the Guraghe and
Wonchi ADPs, the question was considered of how to effectively strengthen the
households of those that have taken in the most vulnerable members of their
community. OVCs experience the kind of undeserved loss that has lasting
consequences on their health, education, and future. Helping those that care for them
demands thoughtful consideration from researchers and practitioners alike.
Nevertheless, as the process of delivering microfinance services to OVC caregivers
was examined, two major challenges in particular became apparent. First, how do
you provide microfinance to a demographic that is susceptible to social exclusion in a
community? Because microfinance tends to rely heavily on mechanisms that require
social capital for its processes, there is a risk of unnecessarily excluding groups that
have already been marginalized by the community. The question then became how
MFIs can reach marginalized groups while still taking advantage of the social capital
mechanisms that have made it so successful in the past?
136
Chapter Two set out to address this question by studying the relationship that
WISDOM had with the OVC caregivers in the villages they serviced. What was
quickly noticeable was that there was another set of actors to consider in this
relationship: the WVE community workers. Thus, as the three sets of actors were
researched, some of the common misconceptions that staff from WISDOM and WVE
had in relation to OVC households in the Guraghe and Wonchi ADPs were explored.
The findings were surprising in the sense that although most staff workers from both
organizations believed that few OVC caregivers were participating in WISDOM
microloans because they were ill suited to the needs of the OVC caregiver, there were
already a high percentage of OVC caregivers participating.
After considering the circumstances under which most OVC caregivers were
recruited and first applied for WISDOM loans, it became apparent that it was by
working with WVE community workers that WISDOM was able to unknowingly
recruit high numbers of OVC caregivers. WVE community workers proved further
valuable to WISDOM in its dealings with OVC caregivers by consulting with
WISDOM staff on community and client issues.
Thus, in its efforts to reach more marginalized community members, WISDOM was
well served by its relationship with WVE. However, collaborating with WVE was
not without its difficulties. This was the second major challenge that was observed
137
for the MFI as it seeks to provide microfinance for OVC caregivers. How can MFIs
optimize the collaboration it has with other community organizations? In particular,
as motivation levels were observed, there seemed to be a range in degrees of worker
satisfaction and worker involvement in the collaboration that WVE community
workers had with WISDOM.
Chapter Three of this dissertation builds on the organization literature concentrating
on collaborative management. Similar to Thomas’s (2003) research on interagency
collaboration and individual-level motivations, WVE community workers tended to
be motivated in circumstances that improved the socioeconomic conditions of its
community members. Thus, where a WVE community worker generally saw
WISDOM’s impact in the community as beneficial, they were highly motivated to
collaborate with WISDOM.
However, where perceptions of WISDOM’s impact in the community were mixed,
other factors seemed to be at work as well. Namely, the abilities of the WVE ADP
manager to frame the collaboration between the two organizations by showing
empathy and helping to build trust was important. Furthermore, as MFIs contemplate
the unique challenges they face in the form of financial and social expectations,
finding a manager who can advocate for collaboration among his or her workers in
another community organization can be vital as well.
138
Such observations are important for the MFI to consider as it seeks collaborators to
improve the reach and quality of its services, especially to groups that are vulnerable
to social exclusion, such as OVC households. Moreover, in working with
marginalized community members, the MFI faces the unique challenge of making
sure its social capital mechanisms can still benefit those with low social capital. It
also faces the unique challenge of collaborating with other non-profit organizations as
a profit-making institution with a mission of empowering the poor. Nevertheless, the
unique contributions that MFIs can offer community members makes overcoming
these challenges all the more important.
The vulnerabilities that the OVC faces are not limited to one arena of his or her life.
The illness or loss of a parent can have financial, social, and psychological long-term
consequences for the child. In particular, changes in family composition and
stigmatization or discrimination from having orphan status can work negatively affect
the child’s psychosocial well-being (Richter, 2006, p. 9-12; Foster & Williamson,
2000, p. S282).
While much attention has been given to the financial benefits that microfinance can
have for participants, Chapter Four of this dissertation examines some of the
additional benefits that microfinance can have for both its participants and the
children of the household. After identifying some key areas of vulnerability for OVC
139
households in the Guraghe and Wonchi ADPs, analysis comparing OVC caregivers
participating in WISDOM is conducted to explore how well microfinance can be used
to mitigate the negative effects of loss for OVCs. By comparing outcomes among
newer borrowers of WISDOM with older borrowers, data indicated that longer
participation in microfinance could improve aspects of household consumption as
well as reduce the need for assets to be sold for basic needs. In addition, OVC
caregivers who participated in WISDOM longer tended to be more likely to attend a
community meeting. Perhaps more importantly were results that indicated that longer
participation in WISDOM tended to improve OVC psychosocial well-being by
decreasing feelings of low-self esteem, stress, and overburden.
5.1
Research Questions
Chapter One of this dissertation presented three central research questions that are
addressed in the following sections.
5.1.1 Microfinance for Groups Vulnerable to Social Exclusion
How can MFIs work to exclude or include community members who are
vulnerable to social exclusion in borrowing? How can MFIs increase flexibility
in their processes to include more who are vulnerable in their services?
140
Because microfinance relies on mechanisms that utilize social capital for client
selection and monitoring, it would seem particularly susceptible to excluding groups
that are vulnerable to social exclusion. To do away with these mechanisms seems
pointless since the success of microfinance depends largely on its ability use social
capital to reduce transaction costs associated with lending to large numbers of people.
Nevertheless, the question is an important one: does microfinance tend to exclude
people based on social status regardless of their financial capacity?
The first objective of this chapter was to explore whether or not this theory was true.
While interviews with OVC caregivers participating in WISDOM microloans
indicated that they did not feel discriminated against, this did not necessarily address
the question of whether discrimination prevented OVC caregivers from participating
in WISDOM. Interviews with staff from WISDOM and WVE were also conducted
and revealed the widespread belief that OVC caregivers were too poor for WISDOM
loans. Many staff believed that interest rates and initial deposit requirements were
too high for OVC caregivers and instead thought direct aid or material support would
be more appropriate for them.
Nevertheless, while it did appear that there were some misconceptions that could
contribute to discrimination against OVC caregivers, research also showed that there
were sizeable numbers of OVC caregivers participating in WISDOM loans. What
141
was more surprising was the fact that this occurred without either of the staffs being
aware of it.
Was it the case then that OVC caregivers faced few obstacles regarding their OVC
caregiver status when being recruited for a WISDOM loan? Analysis regarding when
and where OVC caregivers joined WISDOM suggests that this is not the case.
Recruitment of OVC caregivers was much higher during periods when WISDOM
was collaborating with WVE community workers to recruit and organize groups of
borrowers. Numbers of OVC caregivers joining WISDOM actually dwindle
considerably when there is less collaboration between the two. WVE community
workers play a further significant role for OVC caregivers who are WISDOM clients.
Regression analysis indicates that the quality of collaboration between WISDOM
staff and the WVE community worker actually contributed to greater success in loan
growth, livestock assets, and average meals consumed for the OVC caregiver
participating in WISDOM.
Thus, the research does indicate that misconceptions concerning the capacity of OVC
caregivers to participate in WISDOM did exist, among WVE and WISDOM staff at
least. However, the research also suggests that such misconceptions have less
influence with regards to the recruitment of OVC caregivers for WISDOM loans
where there is involvement in the lending process from workers who are well
142
connected to the community. Because WVE community workers have a widespread
network, including those who are the most vulnerable and marginalized in the
community, their involvement in the lending process can include those that often fall
through the cracks and are excluded by the community. Collaborating with these
sorts of workers helps create flexibility in the recruitment and monitoring process for
WISDOM and increases their range of influence.
5.1.2 Incentives for Collaboration with the MFI
What sorts of challenges do MFIs face when collaborating with other community
organizations? How can MFIs provide incentives for community organization
workers to collaborate with them?
Interviews with WVE staff workers in the Guraghe and Wonchi ADPs were used to
identify potential challenges that existed during collaborative efforts between
WISDOM and WVE. Individual-level motivations among WVE community workers
were examined and found to exist at various levels as indicated by worker and
involvement concerning collaboration with WISDOM. Thus, one challenge that
exists for the MFI when collaborating with others is answering the question of how to
provide effective incentives for its collaborators.
Analysis of the interviews conducted with the staff from both organizations indicated
that the primary factor that influenced worker motivation was whether or not a WVE
143
community worker had a positive perception concerning the impact that WISDOM
had in the community. Where the perception was positive, WVE community workers
were highly motivated and showed strong degrees of satisfaction and involvement.
However, where the perceptions of a WVE community worker was more mixed
concerning WISDOM’s impact in the community, motivation seemed to be
influenced also by the WVE ADP managerial skills and behavior. In particular, in the
instance where the WVE ADP manager was able to frame the collaboration with
WISODM as a priority and create a shared sense of identity, motivation to collaborate
generally seemed high even though the WVE community worker’s perception of
WISDOM’s impact in the community may have been mixed. Thus, the role of the
ADP manager was also an important factor in the motivations of the WVE
community workers.
In addition, it is possible that the profit-making abilities of the MFI created some
questions among WVE community workers concerning whether WISDOM was doing
enough for the community. Particularly because the mission of WVE is to serve the
poorest and most vulnerable in the community, WISDOM’s strategy to focus instead
on the productive poor in the community may have worked to undermine the sense of
commitment that WVE community workers believed WISDOM had to the
community. If this is the case, it may also be true that the profit-making nature of the
MFI creates room for potential distrust or confusion among its collaborators.
144
Furthermore, there was also evidence that the dual obligations of the MFI to financial
and social performance standards created some questions or confusion within
WISDOM’s own management. Thus, further research should be conducted
concerning how the dual obligations of the MFI can affect the trust of its
collaborators.
5.1.3 Mitigating the Negative Effects of Loss for OVC Households with
Microfinance
How does participation in microfinance impact the capacity of households caring
for OVCs? How well does it mitigate the negative effects of losing a parent or
having a parent who is chronically ill for the OVC?
Household survey data interviewing both OVC caregivers and OVCs was used to
consider the effectiveness of microfinance in mitigating effects of loss for OVC
households. As previously discussed, many assume that OVC households are too
poor to participate and succeed in microfinance. Nevertheless, of the 125 OVC
caregivers that were identified and included in this study as WISDOM borrowers in
the two ADPs, only two were delinquent for a repayment rate of 98.86% at the time
of data collection.
With regards to the impact of WISDOM participation on outcome variables, two
main approaches were considered. The first approach compared outcomes among
145
clients and non-clients. However, due to selection bias and endogeneity issues, a
second approach was used. The second approach compared clients who had
participated in WISDOM for less than a year with clients who had participated in
WISDOM for more than a year. Using this approach of newer clients versus older
clients helps reduce bias since all OVC caregivers in the analysis have self-selected
themselves into the program. Although it does not completely rid itself of attrition
bias, it is likely to be a lesser factor than selection bias.
The results of the regression analysis showed somewhat mixed results concerning
household consumption variables. While OVC caregivers participating longer in
WISDOM were likely to have households that consumed more meals and had a
higher percentage of children in school, asset data was less favorable. One possible
explanation for this may be that the group of OVC caregivers who had participated in
WISDOM for more than a year had lower proportions that were receiving aid from
WVE. In addition, longer participation in WISDOM also contributed towards being
less likely to have sold assets in the past year to pay for basic needs. In addition, one
of the most significant findings of this analysis was the impact that longer
participation in WISDOM had on the OVC. Analysis found that longer participation
in WISDOM contributed to increased psychosocial well-being among OVC
adolescents and children.
146
Thus, excluding asset data, research suggests that participation in microfinance can
help mitigate the negative effects of loss for the OVC that go beyond the financial.
OVC households participating in WISDOM for longer periods of time were more
independent, more involved in the community, and had fewer feelings of being
overburdened and stressed among OVCs.
5.2
Significance of These Findings
One of the most important findings of this study was the revelation that there was a
great deal of crossover between what many staff believed were separate target
populations. Almost all of the managers in both WISDOM and WVE assumed that
microfinance benefited the productive poor while direct aid and HIV/AIDS programs
benefited OVC caregivers and their households. The two populations were separate
and had different needs in the minds of many of the staff. Even most of the WVE
community workers, who were closest to the frontlines of the community, had little
knowledge of the crossover between the two groups.
The reality in the community, however, told a different story. There was in fact a
good measure of crossover between the two populations as many OVC caregivers
were indeed already participating in WISDOM services. Approximately 20% of the
WISDOM clients among the villages studied in the Guraghe ADP were in fact
caregivers of at least one OVC. Furthermore, all of this was occurring without either
147
organization being fully conscious of it. To add to this irony is the evidence that it
was likely the involvement of the WVE community workers that helped reach many
of the OVC caregivers that were currently participating in WISDOM. In other words,
the WVE community workers were helping to unknowingly include OVC caregivers
in WISDOM services.
Being unaware of how this collaboration between WVE community workers and
WISDOM staff contributed to successfully including more OVC caregivers in
microfinance, however, had two important consequences for both organizations. The
first consequence was that the involvement of WVE community workers in
WISDOM lending groups was minimized in recent years. Had either organization
been aware of the successes that were occurring from collaborating, it is very possible
that collaboration levels would have continued instead of decreasing. Furthermore,
this minimization had the effect of decreasing the numbers of OVC caregivers who
were recruited for WISDOM loans in recent years. Because microfinance not only
has the potential to positively benefit the financial well-being of households but, as
Chapter Four of this dissertation indicated, also has the potential to positively benefit
the psychosocial well-being of OVCs in the household, decreasing the involvement of
OVC caregivers in microfinance is all that more significant and detrimental.
148
The second consequence of being unaware of the crossover between the two target
populations had on the organizations was that the assumptions that there were indeed
separate target populations with different needs and different problems was never
challenged. Many of the staff in both WISDOM and WVE continued to assume that
microfinance was for the productive poor while direct aid and social services were for
the poorest of the poor, which included OVC caregivers. The problem with these
assumptions was that they were not true and in the end actually worked to discourage
collaboration between many of the staff between WISDOM and WVE. Because
WISDOM staff believed their services were designed to benefit the productive poor
and WVE staff believed their services were designed to benefit the most vulnerable of
the poor, many found the other organization irrelevant to the work that they were
doing. Conversely, where there was some awareness of how the other organization
was contributing to the overall goals and well-being of the target population of their
own organization, WVE community workers especially were not only willing but also
wanted to collaborate with WISDOM.
5.3
Recommendations for Development Work
These findings have important implications for the organizations as they seek to
overcome the challenges of collaborating with others to better serve more people in
the community.
149
5.3.1 Integrating Microfinance with Community Development
To understand why staff in both organizations may have had blinders on concerning
the crossover of target populations occurring in the community, the history of WVE
needs to be considered. Figure 3 from Chapter One is shown again below as Figure
7.
Figure 7 : Individualized Approach to Development
150
WVE began with a community-based development model that aimed to address the
needs of the poorest of the poor in the community through child sponsorship
programs. More than two decades later, WVE began implement microfinance
services to the community that were designed to address the needs of the productive
poor in the community who would become WISDOM clients. Almost a decade and a
half after this, the needs of those affected by HIV/AIDS, including OVCs, were
designed to be addressed through mobilizing aid through the WVE CCCs.
Thus, as the services of WVE evolved, each model was added one by one. However,
the three models remained separate while the models concerning the relationships
between the three efforts remain undeveloped. The importance of these models is
demonstrated by the impact they had on the motivations and perceptions of the staff
in both organizations, most especially the managers of the two organizations.
Nevertheless, the unintended consequence of such models was the inability of many
of the staff to perceive the crossover between target populations that was occurring in
the community. Because organizational mental models were stagnant, much of the
staff in WISDOM and WVE were unable to perceive what was actually occurring
within the community. The reality was that there was no set boundary among the
poorest of the poor and the productive poor. Instead, these groups were overlapping
and mixed. Figure 8 below depicts this phenomenon occurring in the community.
151
Figure 8: Integrated Approach to Development
Development organizations and microfinance institutions need to understand and
model for their workers how their various programs are related so that a greater
integration and collaboration of services is possible. Failing to model such
relationship misses out on the opportunity for synergy that occurs when programs
interact together as opposed to remaining independent. In the case of WVE and
WISDOM, the lack of models showing the relationship between the various programs
resulted in mixed perceptions among staff concerning the value that the other
organization to the development of the community.
152
Figure 9: Relationship Between Assistance and Development
Furthermore, the findings of this dissertation suggest that what was actually occurring
was that WVE services were working to prepare many households through assistance
and training to participate in WISDOM services. Figure 9 shows the relationship
between the development assistance provided by WVE ADPs and the microfinance
services provided by WISDOM on an economic development scale. In fact, many
WVE community workers actually acknowledged this phenomenon in their
interviews and were therefore all the more motivated to collaborate and help
153
WISDOM. Nevertheless, these acknowledgements were based more on their own
observations in the community than organizational policies. Therefore, while some
WVE community workers saw this relationship, others simply saw the two programs
as separate and unrelated.
While more research may be needed to explore this relationship, if in fact
development assistance and training does work to prepare caregivers for microfinance
participation, this indeed provides more incentives for MFIs to seek collaborators in
its lending processes. Development organizations can work to organize, train, and
support caregivers as they grow in income generating capabilities and help prepare
households for microfinance participation. By collaborating with other development
organizations that regularly work with the rural poor, this may also enable MFIs to
extend their reach to include more of the rural poor rather than staying limited to the
more urbanized regions. Thus, the relationship between development assistance and
microfinance has significant implications and have the potential of increasing
microfinance access to a whole new group of people.
5.3.2 Understanding the Community as More Than a Label
This leads to the second recommendation coming out of this study. It is important for
the staff in development organizations to understand the community in which they are
serving. The fact of the matter is that labels, such as “the poorest of the poor” or “the
154
productive poor”, are often just labels and may not depict the entire reality of a group
or population.
The community members in the villages examined in this study understood that
reality does not often fit into categories. While WVE works to target the poorest and
WISDOM works to target the productive, community members simply saw a menu of
services from which they saw could improve their circumstances. They did not ask
themselves into which target population might they fit. Likewise, development
organizations need to do the same. Development organizations need to understand
the diversity of the populations they serve while learning also how to benefit from the
use of models and labels. Tools, such as models and labels, can indeed be helpful to
guide or clarify circumstances for staff but they should not be used to replace their
decision-making ability and need to be adapted as the community changes.
Field staff working on the frontlines of the community, such as the WVE community
workers, already know the importance of being in touch with the community. Many
of the WVE community workers suggested that WISDOM needed to be more in
touch with the community. Some WVE community workers even saw how
understaffed WISDOM was and suggested they work together to improve the
information that WISDOM had concerning the community. Thus, MFIs should
consider following the recommendation provided by these workers to collaborate
155
with other community organizations to improve the information and relationships
they have with their communities.
5.3.3 Providing Incentives to Collaborate
Because the work of collaboration often demands valuable resources from an
organization, it is essential for the MFI to be deliberate about how it approaches
collaborating with others. One of the considerations for MFIs then is how to provide
incentives to other organizations to collaborate with them.
The primary factor found to influence the motivations of the WVE community
workers to collaborate with WISDOM was their own assessment of how WISDOM
was benefiting the community. Where their assessment was positive, so was their
disposition towards WISDOM. Thus, MFIs should make an effort to communicate
with other organizations about how their activities are leading to improved conditions
for the community. This can occur through both informal and formal channels but
should be most directed at the field staff or community workers in other
organizations. While the perceptions of managers in other organizations are
important, it is the role and therefore opinions of the community workers that have
the most potential to help MFIs in their lending activities.
Furthermore, once collaboration occurs, the MFI should attempt to identify the
positive outcomes of its collaboration with workers in the community and
156
communicate them. In the case of WISDOM and WVE, WVE community workers
were able to make a significant difference in the reach of the organization. However,
because these results were unknown, involvement of the WVE community workers in
mobilizing groups decreased, as did the numbers of OVC caregivers participating in
WISDOM loans. Thus, awareness of the positive outcomes of collaboration can help
sustain the positive outcomes of collaboration by both increasing motivation among
community workers and preventing the abandonment of effective processes.
5.4
Concluding Remarks
The increased inclusion of OVC caregivers in WISDOM services led to positive
effects on OVC household consumption and OVC psychosocial well-being. With
chronic illnesses, such as HIV/AIDS, creating millions of new orphans every year,
long-term solutions need to be considered to address the increasing vulnerabilities of
the next generation. Few interventions have the potential to change lives as much as
microfinance. While direct aid can bring short-term relief, when successful,
microfinance can help create a livelihood for families leading to improvements in
health, education, and psychosocial well-being. While collaboration may be
necessary for a MFI like WISDOM to increase the breadth of its reach in a
community, working with others does not come without challenges. Nevertheless, by
being deliberate in its selection of potential partners and working to communicate
157
with the workers in the community, MFIs like WISDOM can work with others to
provide important and needed services to a broader population of the poor.
158
BIBLIOGRAPHY
Agranoff, R., & McGuire, M. (2001). After the Network is Formed: Process, Power,
and Performance. In M. P. Mandell (Ed.), Getting Results through Collaboration:
Networks and Network Structures for Public Policy and Management (pp. 11-29).
Westport: Quorum Books.
Anderson, C. L., Locker, L., & Nugent, R. (2002). Microcredit, Social Capital, and
Common Pool Resources. World Development, 30(1), 95-105.
Anderson, C. L., Gugerty, M. K., Levine, R. O., & Weaver, M. (2002). Microfinance
and HIV/AIDS: Five Key Questions on Program Impact: University of Washington.
Andrews, G., Skinner, D., & Zuma, K. (2006). Epidemiology of health and
vulnerability among children orphaned and made vulnerable by HIV/AIDS in subSaharan Africa. AIDS Care, 18(3), 269-276.
Balkenhol, B. (Ed.). (2007). Microfinance and Public Policy. Geneva: International
Labor Office.
Barnes, C., Keogh, E., & Nemarundwe, N. (2001). Microfinance Program Clients
and Impact: An Assessment of Zambuko Trust, Zimbabwe. Washington, D.C.:
USAID.
Barnett, T., & Blaikie, P. (1992). AIDS in Africa: Its Present and Future Impact.
London: Belhaven Press.
Bauman, L., & Germann, S. (2005). Psychosocial Impact of the HIV/AIDS Epidemic
on Children and Youth. In G. Foster, C. Levine & J. Williamson (Eds.), A Generation
at Risk: The Global Impact of HIV/AIDS on Orphans and Vulnerable Chidlren (pp.
93-133). Cambridge: Cambridge University Press.
Besley, T., & Coate, S. (1995). Group Lending, Repayment Incentives and Social
Collateral. Journal of Development Economics, 46, 1-18.
Bhatt, N., & Tang, S.-Y. (1998). Group-based Microfinance and Economic
Development. In T. Liou (Ed.), Handbook of Economic Development. New York:
Marcel Dekker.
Bhatt, N., & Tang, S.-Y. (2001a). Delivering microfinance in developing countries:
Controversies and policy perspectives. Policy Studies Journal, 29, 319-333.
159
Bhatt, N., & Tang, S.-Y. (2001b). Designing group-based microfinance programs:
Some theoretical and policy considerations. International Journal of Public
Administration, 24, 1103-1125.
Bhatt, N., Painter, G., & Tang, S.-Y. (2002). Ch7: The Challenges of Outreach and
Sustainability for U.S. Microcredit Programs. In Replicating Microfinance in the
United States. Washington DC: Woodrow Wilson Center Press.
Bhatt, N., & Tang, S.-Y. (2002). Determinants of Repayment in Microcredit:
Evidence from Programs in the United States. International Journal of Urban and
Regional Research, 26, 360-376.
Borzel, T. A., & Risse, T. (2003). Public-Private Partnerships: Effective and
Legitimate Tools of International Governance. In E. Grande & L. W. Pauly (Eds.),
Complex Sovereignty: On the Reconstruction of Political Authority in the 21st
Century. Baltimore: Johns Hopkins University Press.
Bozeman, B. (1987). All Organizations are Public. San Francisco: Jossey-Bass
Publishers.
Brau, J. C., & Woller, G. (2004). Microfinance: A Comprehensive Review of the
Existing Literature. Journal of Entrepreneurial Finance and Business Ventures, 9(1),
1-26.
Burt, R. (1992). Structural Holes: The Social Structure of Competition. Cambridge:
Harvard University Press.
Burt, R. (1998). The Gender of Social Capital. Rationality and Society, 10(1), 5-46.
Burt, R. (1999). The Contingent Value of Social Capital. Administrative Science
Quarterly, 42, 339-365.
Buse, K. (2004). Governing Public- Private Infectious Disease Partnerships. Brown
Journal of World Affairs, 10(2), 225-242.
Campbell, D. T., & Stanley, J. C. Experimental and Quasi-Experimental Designs for
Research. In D. T. Campbell & J. C. Stanley (Eds.), Quasi-Experimental Designs for
Research. Chicago: Rand McNally.
Carr, J. H., & Tong, Z. T. (2002). Introduction: Replicating Microfinance in the
United States - An Overview. In Replicating Microfinance in the United States.
Washington DC: Woodrow Wilson Center Press.
160
Castells, M. (2000). The Rise of the Network Society. Oxford: Blackwell.
Central.Statistical.Authority.[Ethiopia], & ORC.Macro. (2001). Ethiopia
Demographic and Health Survey 2000. Addis Ababa, Ethiopia and Calverton,
Maryland, USA: Central Statistical Authority and ORC Macro.
Chatterji, M., Dougherty, L., Ventimiglia, T., Mulenga, Y., Jones, A., Mukaneza, A.,
et al. (2005). The Well-Being of Children Affected by HIV/AIDS in Lusaka, Zambia,
and Gitarama Province, Rwanda: Findings from a Study. Washington DC:
Community REACH Program, Pact.
Cheston, S., & Kuhn, L. (2002). Empowering Women Through Microfinance. In S.
D. Harris (Ed.), Pathways Out of Poverty: Innovations in Microfinance for the
Poorest Families. Bloomfield: Kumarian Press.
Coady, D. P. (2004). Designing and Evaluating Social Safety Nets: Theory, Evidence,
and Policy Conclusions: International Food Policy Research Institute.
Coleman, J. (1988). Social Capital in the Creation of Human Capital. American
Journal of Sociology, 94(Supplement), S95-S120.
Coleman, B. (1999). The Impact of Group Lending in Northeast Thailand. Journal of
Development Economics, 60, 105-142.
Coleman, B. (2002). Working Paper No. 9 - Microfinance in Northeast Thailand:
Who Benefits and How Much? : Asian Development Bank.
Conning, J. (1998). Outreach, Sustainability and Leverage in Microfinance Lending:
A Contract Design Approach. Paper presented at the NEUDC Conference, Economic
Growth Center.
Cooper, T. L. (1985). The Public-Private Continuum. Public Budgeting and Finance,
5(3), 99-115.
Daley-Harris, S. (2005). State of the Microcredit Summit Campaign Report 2005.
Washington, DC: Microcredit Summit Campaign.
de Aghion, B. A., & Morduch, J. (2005). The Economics of Microfinance.
Cambridge: MIT Press.
161
Dean, J. M., Schaffner, J., & Smith, S. L. S. (2005). Christian Collaboration in the
Attack on Poverty: Christian Practitioners and Academics in Collaboration. In J. M.
Dean, J. Schaffner & S. L. S. Smith (Eds.), Attacking Poverty in the Developing
World (pp. 3-23). Federal Way: Authentic Media.
Donahue, J. (2005). Forward-Looking Review: World Vision's Approaches to
Integrating Microenterprise Development and HIV/AIDS Response: World Vision.
Donahue, J. (2005). Strengthening Households and Communities: The Key to
Reducing the Economic Impacts of HIV/AIDS on Children and Families. In G.
Foster, C. Levine & J. Williamson (Eds.), A Generation At Risk: The Global Impact
of HIV/AIDS on Orphans and Vulnerable Children (pp. 37-65). Cambridge:
Cambridge University Press.
Donahue, J., Wamimbi, R., & Kiyaga, E. (2005). End of Pilot Program Evaluation
Microenterprise Development and HIV/AIDS (MEDHA) - May 23rd to June 2nd,
2005: A Collaborative Initiative of Microenterprise Development Network and World
Vision Uganda): World Vision.
Doocy, S., Norell, D., Teffera, S., & Burnham, G. (2006?). Outcomes of an Ethiopian
Microfinance Program and Management Actions to Improve Services. Journal of
Microfinance, 7(1), 79-95.
Dowla, A. (2006). In credit we trust: Building social capital by Grameen Bank in
Bangladesh. The Journal of Socio-Economics, 35(2006), 102-122.
Drew, R. S., Makufa, C., & Foster, G. (1998). Strategies for Providing Care and
Support to Children Orphaned by AIDS. AIDS Care, 10(Suppl), S9-S15.
Eckardt, H. H., & Schuler, H. (1992). Berufseignungsdiagnostik. In R. S. Jager & F.
Petermann (Eds.), Psychologische Diagnostik. Weinheim: Psychologie Verlags
Union.
Eikenberry, A. M., & Kluver, J. D. (2004). The Marketization of the Nonprofit
Sector: Civil Society at Risk? Public Administration Review, 64(2), 132-140.
Fikkert, B. (2003, 2003/1/5-6/). Christian Microfinance: Which Way Now? Paper
presented at the Association of Christian Economists 20th Anniversary Conference:
"Economists, Practitioners, and the Attack on Poverty: Toward Christian
Collaboration".
162
Flynn, P. (2007). Microfinance: The Newest Financial Technology of the Washington
Consensus. Challenge: The Magazine of Economic Affairs, 50(2), 1-9.
Foster, G., Makufa, C., & Drew, R. S. (1996). Supporting Children in Need Through
a Community-Based Orphan Visiting Programme. AIDS Care, 8, 389-403.
Foster, G., & Williamson, J. (2000). A Review of Current Literature of the Impact of
HIV/AIDS on Children in Sub-Saharan Africa. AIDS, 14, 275-284.
Frederickson, H. G. (2004). Whatever Happened to Public Administration?:
Governance, Governance Everywhere. Belfast: Institute of Governance Public Policy
and Social Research.
French, J. R. P., Israel, J., & As, D. (1960). An Experiment in a Norwegian Factory:
Interpersonal Dimensions in Decision-Making. Human Relations, 13, 3-19.
Fukuyama, F. (1999, 1999/10/1/). Social Capital and Civil Society. Paper presented at
the IMF Conference on Second Generation Reforms.
Fukuyama, F. (2002). Social Capital and Development: The Coming Agenda. SAIS
Review, 22(1).
Ghatak, M., & Guinnane, T. W. (1999). The Economics of Lending with Joint
Liability: Theory and Practice. Journal of Development Economics, 60, 195-228.
Gomez, R., & Santor, E. (2001). Membership has its privileges: the effect of social
capital and neighborhood characteristics on the earning of microfinance borrowers.
Canadian Journal of Economics, 34(4), 943-966.
Hansmann, H. (1987). Economic Theories of Nonprofit Organization. In W. W.
Powell (Ed.), The Nonprofit Sector: A Research Handbook. New Haven: Yale
University Press.
Haque, M. S. (2001). The Diminishing Publicness of Public Service under the Current
Mode of Governance. Public Administration Review, 61(1), 65-83.
Helms, B. (2006). Access for All: Building Inclusive Financial Systems. Washington,
DC: The International Bank for Reconstruction and Development
The World Bank.
Hermes, N., & Lensink, R. (2007). The Empirics of Microfinance: What Do We
Know? The Economics Journal, 117(February), F1-F10.
163
Hodge, G. A. (2000). Reviewing the Effectiveness of Privatization as Contracting
Out. In Privatization: An International Review of Performance (pp. 97-156):
Westview Press.
Hulme, D., & Mosley, P. (1996). Finance Against Poverty: Routledge.
Hulme, D. (1997). Impact Assessment Methodologies for Microfinance: A Review.
Paper presented at the AIMS Project.
Hulme, D. (2000). Impact Assessment Methodologies for Microfinance: Theory,
Experience and Better Practice. World Development, 28(1), 79-98.
Hunter, S. (1990). Orphans as a Widow on the AIDS Epidemic in sub-Saharan
Africa: Initial Results and Implications of a Study in Uganda. Social Science
Medicine, 31, 681-690.
Innes, J. E., & Booher, D. E. (1999). Consensus Building and Complex Adaptive
Systems: A Framework for Evaluating Collaborative Planning. Journal of the
American Planning Association, 65(4).
Karlan, D. (2003). Social Capital and Group Lending.
Kettl, D. (1995). Privatization as a Tool of Reform. THE LaFOLLETTE POLICY
REPORT.
Kettl, D. (1997). The Global Revolution in Public Management: Driving Themes,
Missing Links. Journal of Policy Analysis and Management, 16(3), 446-462.
Khandker, S., R. (1998). Fighting Poverty with Microcredit: Experience in
Bangladesh. New York: Oxford University Press.
Khandker, S., R. (2003). Microfinance and Poverty: Evidence Using Panel Data from
Bangladesh. Wold Bank Economic Review, 19(2).
Kim, J.-O., & Mueller, C. W. (1978). Factor Analysis: Statistical Methods and
Practical Issues. Beverly Hills: Sage Publications.
Knott, J. (1993). Comparing Public and Private Management: Basic Management
Functions and External Constraints. Journal of Public Administration Research and
Theory, 3, 93-119.
164
Levine, C., Foster, G., & Williamson, J. (2005). Introduction: HIV/AIDS and Its
Long-Term Impact on Children. In G. Foster, C. Levine & J. Williamson (Eds.), A
Generation At Risk: The Global Impact of HIV/AIDS on Orphans and Vulnerable
Children. Cambridge: Cambridge University Press.
Lewis, R. J. (2000). Power Analysis and Sample Size Determination: Concepts and
Software Tools. Paper presented at the 2000 Annual Meeting of the Society for
Academic Emergency Medience (SAEM), San Francisco, CA.
Locke, E. A., & Schweiger, D. M. (1979). Participation in Decision-Making: One
More Look. Research in Organizational Behavior, 1, 265-339.
Luippold, B., & Roden, P. (2006). The Hope Initiative: World Vision's Global
Response to HIV and AIDS, FY06 Annual Report. Federal Way, WA: World Vision.
Matin, I., Hulme, D., & Rutherford, S. (1999). Paper No. 9 - Financial Services For
The Poor and Poorest: Deepening Understanding to Improve Provision. Manchester:
University of Manchester.
Matin, I., & Hulme, D. (2003). Programs for the Poorest: Learning from the IGVGD
Program in Bangladesh. World Development, 31(3), 647-665.
Mayoux, L. Whom Do We Talk to? Issues in Sampling.
McGuire, M. (2002). Managing Networks: Propositions on What Managers Do and
Why They Do It. Public Administration Review, 62(5), 599-609.
McKernan, S.-M. (2002). The Impact of Microcredit Programs on Self-Employment
Profits: Do Noncredit Program Aspects Matter? Review of Economcis and Statistics,
84(1), 93-115.
Miller, K., & Monge, P. (1986). Participation, Satisfaction, and Productivity: A MetaAnalytic Review. Academy of Management Journal, 29(4), 727-753.
Milward, H. B., & Provan, K. G. (2000). Governing the Hollow State. Journal of
Public Administration and Theory, 10(2), 359-379.
Mix Market.Profile for Wisdom (WISDOM). Retrieved June 29, 2009, from
http://www.mixmarket.org/en/demand/demand.show.profile.asp?ett=1203
165
MkNelly, B., & Dunford, C. (1998). Impact of Credit with Education on Mothers and
Their Young Children's Nutrition: Lower Pra Rural Bank Credit With Education
Program in Ghana. Davis: University of California, Davis - Program in International
Nutrition.
Moe, T. (1984). The New Economics of Organization. American Journal of Political
Science, 28, 739-777.
Montgomery, R. (1996). Disciplining or Protecting the Poor? Avoiding the Social
Costs of Peer Pressures in Micro-Credit Schemes. Journal of International
Development, 8(2), 289-305.
Morduch, J. (1999). The Microfinance Promise. Journal of Economic Literature, 37,
1569-1614.
Morduch, J. (2000). The Microfinance Schism. World Development, 28(4), 617-629.
Mosley, P. The Use of Control Groups in Impact Assessments for Microfinance.
Geneva: International Labor Office - Enterprise and Cooperative Development
Department.
Mosley, P. (2001). Microfinance and Poverty in Bolivia. The Journal of Development
Studies, 37, 101-132.
Mwaipopo, R. (2005). Evaluation of Tahea Supported "Mama Mkubwa" Initiative in
Makete District, Iringa Region. Dar es Salaam, Tanzania: UNICEF.
Ostrom, V., & Ostrom, E. (1991). Public Goods and Public Choices: The Emergence
of Public Economies and Industry Structures. In The Meaning of American
Federalism (pp. 163-197). San Francisco: ICS Press.
Ostrom, E. (1999). Social Capital: A Fad or Fundamental Concept? In Social Capital:
A Multifaceted Perspective (pp. 172-214). Washington DC: The World Bank.
Parker, R., Easton, G. D., & Klein, C. H. (2000). Structural Barriers and Facilitators
in HIV Prevention: A Review of International Research. AIDS, 14(Suppl 1), 22-32.
Pattberg, P. (2005). The Institutionalization of Private Governance: How Business
and Non-profits Agree on Transnational Rules. Governance: An International
Journal of Policy, Administration and Institutions, 18(4), 589-610.
166
Perry, J. L., & Rainey, H. G. (1988). The Public-Private Distinction in Organization
Theory: A Critique and Research Strategy. The Academy of Management Review,
13(2), 182-201.
Peters, B. G., & Pierre, J. Governance without Government? Rethinking Public
Administration. Journal of Public Administration Research and Theory: J-PART,
8(2), 223-243.
Phiri, S. N., & Tolfree, D. (2005). Family- and Community-Based Care for Children
Affected by HIV/AIDS: Strengthening the Front Line Response. In G. Foster, C.
Levine & J. Williamson (Eds.), A Generation At Risk: The Global Impact of
HIV/AIDS on Orphans and Vulnerable Children (pp. 11-36). Cambridge: Cambridge
University Press.
Pitt, M. M., & Khandker, S., R. (1998). The Impact of Group-based Credit Programs
on Poor Households in Bangladesh: Does the Gender of the Participants Matter?
Journal of Political Economy, 106(5), 958-996.
Portes, A., & Sensenbrenner, J. (1993). Embeddedness and Immigration: Notes on the
Social Determinants of Economic Action. American Journal of Sociology, 98(6),
1320-1350.
Portes, A. (Ed.). (1995). The Economic Sociology of Immigration: essays on
Networks, Ethnicity, and Entrepreneurship. New York: Russell Sage Foundation.
Portes, & Landolt. (1996). The Downside of Social Capital. American Prospect,
26(May-June), 18-21.
Pronyk, P. M., Kim, J. C., Hargreaves, J. R., Makhubele, M. B., Morison, L. A.,
Watts, C., et al. (2005). Microfinance and HIV Prevention - Emerging Lessons from
Rural South Africa. Small Enterprise Development, 16(3), 26-38.
Putnam, R. D. (1993). The Prosperous Community: Social Capital and Public Life.
The American Prospect, 13, 35-42.
Putnam, R. D. (2001). Bowling Alone: The Collapse and Revival of American
Community. New York: Simon & Schuster.
Richter, L. (2006). Ch2: The Impact of HIV/AIDS on the Development of Children.
In J. Williamson, C. Levine & G. Foster (Eds.), A Generation at Risk? HIV/AIDS,
Vulnerable Children and Security in Southern Africa. Cambridge: Cambridge
University Press.
167
Salamon, L. M. (1999). The Nonprofit Sector at a Crossroads: The Case of America.
International Journal of Voluntary and Nonprofit Organiztions, 10(1), 5-23.
Schreiner, M. (2002a). Aspects of Outreach: A Framework for the Discussion of the
Social Benefits of Microfinance.
Schreiner, M., & Morduch, J. (2002b). Ch1: Opportunities and Challenges for
Microfinance in the United States. In Replicating Microfinance in the United States.
Washington DC: Woodrow Wilson Center Press.
Schreiner, M. (2003). A Cost-Effectiveness Analysis of the Grameen Bank of
Bangladesh. Development Policy Review, 21(3), 357-382.
Sengendo, J., & Nambi, J. (1997). The Psychological Effect of Orphanhood: A Study
of Orphans in Rakai District. Health Transition Review, 7(S), 105-124.
Shetty, A. K., & Powell, G. (2003). Children Orphaned by AIDS: A Global
Perspective. Seminars in Pediatric Diseases, 14(1), 25-31.
Sriram, M. S. Microfinance and the State: Exploring Areas and Structures of
Collaboration. Vastrapur: Sir Ratan Tata Trust Fund for Research Collaborations in
MicroFinance, Indian Institute of Managment.
Stanczak, G. C., Miller, D. E., & Dyrness, G. R. (2006). Spiritually Engaged
Development: Faith and Community in Transforming Tanzania. Los Angeles: USC
Center for Religion and Civic Culture.
Stiglitz, J., & Weiss, A. (1981). Credit Rationing in Markets with Imperfect
Information. American Economic Review, 71(3), 393-410.
Stiglitz, J. (1989). Financial Markets and Development. Oxford Review of Economic
Policy, 5(4), 55.
Stiglitz, J. (1990). Peer Monitoring and Credit Markets. World Bank Economic
Review, 4(3), 351-366.
Tang, S.-Y. (2005). Individual-Level Motivations for Interagency Cooperation.
Public Administrative Review, 65(3), 377-378.
Tegarden, K. (2006). Integrating Microfinance with HIV/AIDS Care: Improving
Income for East African Women in High-Prevalence Areas (A Proposal). Los
Angeles: World Vision.
168
Thomas, C. W. (2003). Bureaucratic Landscaptes: Interagency Cooperation and the
Preservation of Biodiversity. Cambridge: Massachusetts Institute of Technology.
UNAIDS. (2004). 2004 Report on the global AIDS epidemic. Geneva: UNAIDS.
UNAIDS. (2006). 2006 AIDS Epidemic Update. Geneva.
UNICEF. (2007). Children and AIDS - A stocktaking report: Actions and progress
during the first year of Unite for Children, Unite against AIDS. Geneva.
USAID. (2001). Care for Orphans, Children Affected by HIV/AIDS and Other
Vulnerable Children: A Strategic Framework. Arlington: Family Health International.
USAID, UNICEF, & UNAIDS. (2002). Children on the Brink: A Joint Report on
Orphan Estimates and Program Strategies. New York: UNICEF.
USAID, UNICEF, & UNAIDS. (2004). Children on the Brink: A Joint Report on
New Orphan Estimates and a Framework for Action. New York: UNICEF.
USAID. (2008). Pastoralist Livelihoods Initiative Livestock Market Monitoring
Bulletin. In ACDI/VOCA (Ed.) (Vol. 4). Arlington: USAID.
van Bastelaer, T. (2000). Imperfect Information, Social Capital and the Poor's Access
to Credit. Institutional Reform and the Informal Sector (IRIS) working paper(234).
Von Pischke, J. D. (2002). Ch2: Microfinance in Developing Countries. In
Replicating Microfinance in the United States. Washington DC: Woodrow Wilson
Center Press.
Webb, D. (2005). Interventions to Support Children Affected by HIV/AIDS: Priority
Areas for Future Research. In G. Foster, C. Levine & J. Williamson (Eds.), A
Generation At Risk: The Global Impact of HIV/AIDS on Orphans and Vulnerable
Children (pp. 232-253). Cambridge: Cambridge University Press.
Weiss, J., Montgomery, H., & Kurmanalieva, E. (2003). Micro Finance and Poverty
Reduction in Asia: What is the Evidence? Paper presented at the ADB Institute
Research Paper Series No. 53.
WHO. (2005). HIV/AIDS Epidemiological Surveillance Report for the WHO African
Region 2005 Update. Geneva: WHO.
169
Williamson, O. E. (1981). The Economics of Organization: The Transaction Cost
Approach. American Journal of Sociology, 87(November), 548-577.
Williamson, O. E. (1988). The Logic of the Economic Organization. Journal of Law,
Economics, & Organization, 4, 65-93.
Williamson, O. (1999). Public and Private Bureaucracies: A Transaction Cost
Economics Perspectives. Journal of Law, Economics, and Organization, 15, 306-342.
Williamson, O. E. (2002). The Theory of the Firm as Governance Structure: From
Choice to Contract. Journal of Economic Perspective, 16(3), 171-195.
Williamson, J. (2005). Finding the Way Forward: Reducing the Impacts of
HIV/AIDS on Vulnerable Children and Families. In G. Foster, C. Levine & J.
Williamson (Eds.), A Generation At Risk: The Global Impact of HIV/AIDS on
Orphans and Vulnerable Children (pp. 232-253). Cambridge: Cambridge University
Press.
WISDOM. (2006). WISDOM Business Plan for FY'2006-FY'2010. Addis Ababa:
WISDOM.
Wolch, J. (1999). Decentering America's Nonprofit Sector: Reflections on Salamon's
Crises Analysis. International Journal of Voluntary and Nonprofit Organiztions,
10(1), 25-35.
Woller, G., Wheeler, G., & Checketts, N. (1999). Evaluation practices in microcredit
institutions. Journal of Developmental Entrepreneurship, 4, 59-80.
Woller, G., Dunford, C., & Woodworth, W. (1999a). Where to Microfinance?
International Journal of Economic Development, 1(1), 29-64.
Woolcock, M., & Narayan, D. (2000). Social Capital: Implications for Development
Theory, Research, and Policy. World Bank Research Observer, 15(2).
Woolcock, M. (2001a). The Place of Social Capital in Understanding Social and
Economic Outcomes. The Canadian Journal of Research Policy, 2(1), 65-88.
Woolcock, M. (2001b). Microenterprise and Social Capital: A Framework for
Theory, Research, and Policy. Journal of Socio-Economics, 30, 193-198.
170
World.Vision. (2005). HIV/AIDS Hope Initiative: Guide to Mobilising and
Strengthening Community-Led Care for Orphans and Vulnerable Children: World
Vision International.
World.Vision.Internation."Who We Are". Retrieved June 29, 2009, from
http://wvi.org/wvi/wviweb.nsf/maindocs/3F50B250D66B76298825736400663F21?o
pendocument
World.Vision.International. (2006). World Vision Core HIV and AIDS Response
Monitoring System (CHARMS): Important Information and Practical Tips: World
Vision.
Wright, G. A. N. (2000). Microfinance Systems: Designing Quality Finanical
Services for the Poor. London: Zed Books.
Yunus, M. (2003). Banking to the Poor: Micro-lending and the Battle Against World
Poverty: Public Affairs.
Yunus, M. (2007). Creating a World Without Poverty: Social Businesses and the
Future of Capitalism. New York: Public Affairs.
171
APPENDICES
Appendix A: Interview Protocol for Staff Workers from WISDOM
and WVE
•
•
•
•
How long have you worked with [WISDOM/WVE]?
Describe your role in [WISDOM/WVE].
Describe any experiences of which you are aware in which [WISDOM/WVE] has
worked with [WVE/WISDOM].
Describe any experiences you have personally had working with
[WVE/WISDOM].
o Would you describe these experiences as very positive, positive, negative,
or very negative?
•
What do you consider the role of [WVE/WISDOM] to be in the community?
•
•
Whom would you identify as the leaders in [WISDOM/WVE]?
What kind of interactions with [WVE/WISDOM] do the leaders in
[WISDOM/WVE] encourage, if any?
•
•
What resources do [WISDOM/WVE] have that [WVE/WISDOM] needs?
What resources do [WVE/WISDOM] have that [WISDOM/WVE] needs?
•
How often do you trust [WVE/WISDOM] to incorporate [WISDOM/WVE]
suggestions in their decision-making and activities?
•
Please write the names of all the [WVE/WISDOM] staff that fit the criteria
below:
o I communicated with this individual in the last week
o This individual is important for getting my job done
o I seek advice from this individual
o I consider this individual to be a friend
172
Appendix B: Interview Protocol for OVC Caregiver
•
•
Please list the names and ages of your household members.
Please describe your relationship to [NAME OF OVC].
o When and how did she/he come to live in this household?
o What sort of changes did your household experience when [NAME OF
OVC] came to live with you?
o How did [NAME OF OVC] appear to adjust to this change?
o What has changed in your household since then?
o What has changed for [NAME OF OVC] since then?
•
Please describe your relationship to WVE’s Community Care Coalitions.
•
Have you ever participated in WISDOM’s services?
o If yes, please describe your relationship to WISDOM community banks.
How did you get connected to them? When did you begin to receive
services?
 How were the loans used?
 What has been the result from your participation in these services?
 Describe your experiences with the WISDOM community bank.
Would you describe these experiences as very positive, positive,
negative, or very negative?
 What ways could WISDOM improve their services?
o If no, is there a reason why you haven’t participated in WISDOM’s
services?
173
Appendix C: Household Surveys – OVC Caregiver
Household Roster
• Please give me the name of all persons who currently live in this household.
o Is [NAME] male or female?
o What is [NAME]’s relationship to you?
o How old is [NAME]?
o How many meals did [NAME] have yesterday?
o Is [NAME] currently attending school?
o Are you [NAME]’s primary caregiver?
o Has [NAME] lived with you their entire life?
 If no, how long as [NAME] lived with you?
o Is [NAME]’s natural mother alive?
 If alive, has [NAME]’s mother been sick for at least 3 months
during the past year?
 If not, in what year did [NAME]’s mother die?
o Is [NAME]’s natural father alive?
 If alive, has [NAME]’s father been sick for at least 3 months
during the past year?
 If not, in what year did [NAME]’s father die?
Household Assets
• Does your household have:
o Electricity
o Radio
o Television
o Refrigerator
o Telephone
• Does any member of your household own:
o Bicycle
o Motorcycle
o Car
Socioeconomic Status:
• In what month and year were you born?
• Have you ever attended school?
• What religion are you?
o Orthodox
o Catholic
o Protestant
o Muslim
174
•
•
•
•
o Traditional
o Other
Are you currently married, separated, divorced, or widowed?
What type of work do you do most of the time?
o Not working now
o Farmer
o Sell small goods
o Trader
o Artisan
o Professional/Civil Servant
o Pensioner/Retiree
o Carpentry
o Other
o No response
What was your household’s total income last month?
How long have you lived in this community?
o Less than one year
o 1 to 5 years
o 6 years or more
o All of my life
o I don’t know
Social Network
• Does your household regularly send any money or goods to relatives or other
households to assist them?
• Does your household regularly receive money or food from any other relative or
household?
•
Do you agree or disagree with the following statements?
o This is a close-knit community
o People in this community can be trusted
o People in the community do not share the same values
o People in this community generally don’t get along with each other
•
If your household needed help, like with money or food, are there people beyond
this household, besides relatives, from whom you could get assistance?
Do you ever go to meetings with your neighbors to discuss community concerns,
such as schools or health centers?
In the past year, did your household have to sell any land, livestock, or equipment
in order to have money to buy food, clothing, or pay for heath care or schooling?
In the past 12 months, has anyone who has lived in this household, been very sick
(i.e. had an illness for at least 3 months)?
•
•
•
175
Appendix D: Household Surveys – OVC Child
Caregiver: Child Assets
• I would like to first beging with a few questions regards [NAME]:
o Does [NAME] have
 A pair of shoes
 Spare set of clothes
 School uniform
 Notebook
 Pens, pencils
 Blanket
OVC Child: Psychosocial Well-being
• I am going to ask you some questions about your feelings. I want you to tell me
how often you have felt this way in the last 2 weeks. Please tell me if you have
felt that way often, sometimes, or never.
o Felt worried
o Felt happy
o Refused to eat at mealtimes
o Felt frustrated
o Felt like running away from home
o Had free time to play outdoors
o Have nightmares
•
I am going to ask you some questions about some things that may or may not
bother you. I want you to tell me for each of the following things that I read if it
bothers you a lot, a little, or if doesn’t bother you at all
o Having too many responsibilities
o Taking care of family members
o Not having enough money for things, such as clothing or food
o Worrying about the health of a family member
o Having problems with friends and neighbors
o Having problems getting along with your family
o Feeling unsafe in your neighborhood
o Feeling unsafe in your home
176
Appendix E: Household Surveys – OVC Adolescent
Adolescent Assets
• Do you have
o A pair of shoes
o Spare set of clothes
o School uniform
o Notebook
o Pens, Pencils
o Blanket
Psychosocial Well-being
• Please tell me how much you agree or disagree with the following statements
about yourself (Agree, Disagree, Don’t know, No response)
o You have little control over the things that happen to you
o There is really no way you can solve some of the of the problems you
have
o You feel you have much to be proud of
o You often feel helpless in dealing with problems
o When you make a mistake you take responsibility for it
o You are a good person
o There is little you can do to change many of the important things in your
life
o At times you think you are no good at all
o You often find yourself angry that you get into a fight
o It is important for you to help other people

I am going to ask you some questions about yourself. Please tell me how
often you feel this way or how often these things occur. Please tell me if it is
often, sometimes, or never.
o You are a good person
o You feel happy
o You feel worried
o You refuse to eat at mealtimes
o You feel frustrated when something does not go your way
o You feel like running away from home
o How often would you say you spend time playing, either at home or
somewhere else

I am going to read you some statements about yourself. Please tell me
whether these things bothers you often, sometimes, or never
177
o
o
o
o
o
o
o
o
Having too many responsibilities
Taking care of family members
Not having enough money for things such as clothing and food
Concerns about the health of a family member
Problems with friends and neighbors
Problems getting along with your family
Feeling unsafe in your neighborhood
Feeling unsafe in your home
178
Appendix F: Household Asset Survey
•
How many of the following did your household have in August 2007?
o Ox
o Horse
o Donkey
o Mule
o Sheep
o Goat
o Poultry
o Other
•
How many of the following does your household currently have?
o Ox
o Horse
o Donkey
o Mule
o Sheep
o Goat
o Poultry
o Other
179
Appendix G: Calculation of Livestock Assets
Valuation of livestock assets was calculated from an approximation of data from the
USAID and World Vision US. The foreign exchange rate used to convert the birr to
dollars was 0.11222 dollars per birr or 8.911 birr per dollar8.
Prices for oxen, sheep, and goats were taken from an average of prices that each
respective livestock was sold at during the month of March 2008 in the Oromia
region. Information provided by USAID’s Pastoralist Livelihoods Initiative
Livestock Market Monitoring Bulletin (USAID, 2008). See calculations for ox,
sheep, and goals below.
LIVESTOCK
Ox
Subtotal
Sheep
Subtotal
Goats
8
Number
Sold
47
31
16
23
85
76
24
38
17
92
38
20
54
128
39
728
125
92
118
96
431
68
51
59
42
Avg Price
(Birr)
4750
4800
3300
5050
2800
2850
2000
3050
1500
2300
3050
1300
2400
1650
1000
435
380
195
190
430
385
435
195
Total Price
(Birr)
223250
148800
52800
116150
238000
216600
48000
115900
25500
211600
115900
26000
129600
211200
39000
1918300
54375
34960
23010
18240
130585
29240
19635
25665
8190
http://www.oanda.com for August 1, 2007. Accessed on April 1, 2008.
180
61
281
Subtotal
190
11590
94320
Because there is little data on livestock pricing in Ethiopia, the prices for poultry,
donkey, and cows were based on a proportion of costs from the World Vision US
website9. The proportion was determined by the price of an ox provided by the
USAID livestock bulletin divided by the price of an ox provided by the World Vision
US website. See calculations for poultry, donkey, and cows below. In addition,
because no data could be found on mule or horse pricing, the price for the donkey
was also used for the mule and the price for the cow was also used for the horse.
Because there is a great deal of variance in prices, with the most expensive being the
ox at 2635 Birr and the least expensive being the price of poultry at 46 Birr, an
approximation of prices is suitable for valuing total livestock assets for each
household.
WVUS Price
Ox
Sheep
Goat
Poultry
Donkey
Mule
Cow
Horse
% of Ox Price
from WVUS
715
105.00
0.146853147
75.00
0.104895105
12.50
0.017482517
225.00
0.314685315
Assume same as donkey.
500.00
0.699300699
Assume same as cow.
% of Ox Price Based
on USAID (USD)
2635.027473
386.9620764
276.4014831
46.06691386
829.2044494
829.2044494
1842.676554
1842.676554
9
http://www.worldvision.com.au/Smiles/GiftCatalogue/Gift.aspx?GiftId=57. Accessed on April 1,
2008.
181
Appendix H: Descriptive Statistics – Guraghe ADP
Descriptive Statistics of OVC Caregivers in Guraghe ADP
Total
Sample
NonClients w/o
Access
NonClients
w/ Access
Clients
< 1 yr
Clients
> 1yr
24.93
22.62
24.41
26.33
27.14
15 to 24 years (46)
21.7%
14.3%
18.5%
33.3%
33.3%
25 to 49 years
58.7%
71.4%
63.0%
33.3%
44.4%
50+ years
19.6%
14.3%
18.5%
33.3%
22.2%
100%
100%
100%
100%
100%
16.9%
12.1%
16.9%
23.3%
15.6%
1.3%
3.0%
1.5%
-----
-----
75.6%
84.8%
75.4%
66.7%
75.0%
6.3%
-----
6.2%
10.0%
9.4%
100%
100%
100%
100%
100%
13.0%
9.7%
12.3%
14.8%
16.1%
Not Working
15.5%
3.2%
31.7%
-----
9.4%
Farmer
AGE (162)
Total
MARITAL (160)
Married
Separated/Divorced
Widowed
Never Married
Total
SCHOOL (154)
Yes
OCCUPATION (155)
47.1%
51.6%
36.5%
62.1%
50.0%
Petty Trading
2.6%
6.5%
3.2%
-----
-----
Trader
3.9%
9.7%
1.6%
3.4%
3.1%
Artisan
1.9%
-----
1.6%
-----
6.3%
Professional/Civil
17.4%
12.9%
14.3%
27.6%
18.8%
Other
11.6%
16.1%
11.1%
6.9%
12.5%
TOTAL
100%
100%
100%
100%
100%
3.3%
9.7%
1.6%
-----
3.3%
IGA TRAINING (153)
Yes
5.08
5.32
5.39
4.93
4.31
# of M Adults (157)
1.01
0.79
1.10
1.03
1.03
# of F Adults
1.20
1.12
1.24
1.23
1.19
# of M Children
(158)*** (5.913)
# of F Children
1.55
2.15
1.52
1.47
1.06
1.23
1.33
1.33
1.20
0.97
AVG HH SIZE (162)
Note: Total number of subjects who answered the question is given in parentheses next to subheadings.
182
Descriptive Statistics of OVCs in Guraghe ADP
Total
Sample
Non-Clients
w/o Access
Non-Clients
w/ Access
Clients
< 1 yr
Clients
> 1yr
HHs w/Maternal Orphans
11.9%
18.5%
HHs w/Paternal Orphans
69.4%
63.0%
10.9%
7.1%
12.5%
72.7%
67.9%
70.8%
HHs w/Both Orphans
18.7%
18,5%
16.4%
25.0%
16.7%
100%
100%
100%
100%
100%
Mother Chronically Ill
31.2%
8.0%
53.5%
14.3%
30.0%
Father Chronically Ill
0.9%
4.0%
-----
-----
-----
Both Chronically Ill
1.8%
4.0%
2.3%
-----
-----
73.4%
88.2%
60.0%
79.2%
77.8%
93.0%
93.3%
91.4%
90.9%
100.0%
0.65
0.85
0.35
1.23
0.53
Orphan Status (134)
Total
Sickness Status (109)
Adolescents
Have Regular Place to
Sleep at Night (94)
Attending School (86)
Self Esteem Score (162) **
1.42
1.53
1.09
2.10
1.34
HIV edu** (93) (9.240)
88.2%
82.4%
100.0%
75.0%
88.2%
HIV edu in last 3 mo? (78)
59.0%
46.7%
68.8%
47.4%
66.7%
Test availability (68)
80.9%
66.7%
80.0%
93.8%
83.3%
Tested* (84) (6.942)
23.8%
7.1%
18.2%
41.7%
23.1%
51.9%
63.3%
41.8%
65.2%
47.6%
Buren Score* (162)
Children
Have Regular Place to
Sleep at Night (129)
Attending School (112)
99.1%
100.0%
97.9%
100.0%
100.0%
Worry Score (162)
0.49
0.68
0.38
0.70
0.34
Buren Score (162) 1.20
1.20
1.24
1.11
1.53
1.06
Note: Total number of subjects who answered the question is given in parentheses next to subheadings.
183
Appendix I: Descriptive Statistics – Wonchi ADP
Descriptive Statistics of Caregivers in Wonchi ADP
Total
Sample
Non-Clients
w/o Access
Non-Clients
w/ Access
Clients
> 1yr
21.59
24.29
20.40
21.54
15 to 24 years (77)
19.5%
23.5%
16.7%
20.8%
25 to 49 years
53.2%
47.1%
55.6%
54.2%
50+ years
27.3%
29.4%
27.8%
25.0%
100%
100%
100%
100%
69.9%
58.3%
63.5%
84.6%
1.7%
-----
3.8%
-----
25.2%
33.3%
28.8%
15.4%
AGE (116)
Total
MARITAL STATUS (115)
Married
Separated/Divorced
Widowed
3.5%
8.3%
3.8%
-----
100%
100%
100%
100%
39.5%
37.5%
41.5%
37.8%
Farmer
96.5%
91.7%
98.1%
97.4%
Trader
1.7%
8.3%
-----
-----
Retiree
0.9%
-----
1.9%
-----
Other
0.9%
-----
-----
2.6%
TOTAL
100%
100%
100%
100%
11.5%
-----
13.5%
16.2%
Never Married
Total
SCHOOL (114)
Yes
OCCUPATION (115)*
IGA TRAINING (113)
Yes
5.78
5.50
5.98
5.69
# of M Adults (113)
1.09
1.22
1.06
1.05
# of F Adults
1.10
1.17
1.06
1.10
# of M Children
1.90
1.70
2.02
1.87
# of F Children
1.60
1.35
1.67
1.67
AVG HH SIZE (116)
Note: Total number of subjects who answered the question is given in parentheses next to subheadings.
184
OVC Well-Being in Wonchi ADP
Total
Sample
Non-Clients
w/o Access
Non-Clients
w/ Access
Clients
> 1yr
HHs w/Maternal Orphans
25.9%
21.1%
25.0%
33.3%
HHs w/Paternal Orphans
60.3%
57.9%
70.8%
46.7%
HHs w/Both Orphans
13.8%
21.1%
4.2%
20.0%
100%
100%
100%
100%
Mother Chronically Ill
16.2%
5.3%
20.0%
21.1%
Father Chronically Ill
16.2%
15.8%
13.3%
21.1%
74.9%
76.5%
78.9%
66.7%
Orphan Status (58)
Total
Sickness Status (68)
Adolescents
Have Regular Place to
Sleep at Night (79)
Attending School (74)
87.8%
86.7%
86.5%
90.9%
Self Esteem Factor Score (116)
0.40
0.33
0.43
0.38
Burden Factor Score (116)
1.21
1.29
1.40
0.90
HIV edu (73)
89.0%
93.3%
86.1%
90.9%
HIV edu in last 3 mo? (72)
43.1%
46.7%
40.0%
45.5%
Test availability (51)
58.8%
53.8%
47.6%
76.5%
7.6%
6.7%
6.3%
10.5%
71.4%
58.8%
71.1%
79.3%
90.5%
90.9%
97.1%
77.8%
Worry Score (116)
0.52
0.46
0.66
0.36
Buren Score* (116)
1.28
1.00
1.62
1.00
Tested (66)
Children
Have Regular Place to
Sleep at Night (91)
Attending School* (63)
Note: Total number of subjects who answered the question is given in parentheses next to subheadings.
185
Appendix J: Additional Regression Results
Impact of Microfinance on OVC Household Capacity
Attend Community Meetings
Length of Microfinance (0 = Less than 1
yr, 1 = Greater than 1 yr)
0.413***
ADP (0 = Guraghe, 1 = Wonchi)
Occupation (0 = Not Currently Working)
Farmer
Trader
Professional/Civil Servant
Other
Income Generating Activities
Caregiver Educated?
-0.003
# of M Adults
# of F Adults
# of M Children
# of F Children
Orphan Status (0 = Maternal)
Paternal
Both Parents
Lived in Village (0 = Entire Life)
6 or more years?
Assistance Beyond Family?
Receives Money?
Send Money?
Sell Assets in Past Yr?
Chronic Illness in HH?
Percent of School Aged Children in CIP
F-stat
Adjusted R2
Observations
0.035
-0.203
-0.237
-0.210
0.152
-0.010
0.207
0.302**
0.055
-0.068
0.244
-0.248
0.019
------------------------2.671***
0.330
52
Note: All entries are standardized coefficients. Newer clients participated in microfinance for
less than a year. Longer clients participated in microfinance for greater than one year and up
to 6 years. All VIF scores tested below or equal to 8.064.
186
Impact of Microfinance on OVC Household Capacity
Sell Assets for Basic Needs
Client Longer Than 1 Yr
(0 = No, 1 = Yes)
# M Adults in HH
# F Adults in HH
-0.580***
0.341**
-0.266
# M Children in HH
-0.316**
# F Children in HH
0.303*
Occupation: (0 = Not Working)
Artisan
-----
Farmer
-0.165
Trader
0.321
Professional/Civil Servant
-0.230
Other
-0.624**
Marital Status (0 = Married)
Widowed
Never Married
0.095
-0.635**
IGA Training
-0.193
Caregiver Educated
0.447*
Avg % of School-Aged Children CIP
-0.077
Food Assistance
0.091
Chronic Illness in HH
0.033
F-stat
1.975*
Sig
0.058
Adjusted R2
0.266
N
44
Note: All entries are standardized coefficients. Newer clients participated in microfinance for
less than a year. Longer clients participated in microfinance for greater than one year and up
to 6 years. All VIF scores tested below or equal to 7.207.
187
Appendix K: OVC Psychosocial Indicator Groupings
Adolescent Self-Esteem Measures
• You are a good person
• You feel unhappy
• You feel worried
• You refuse to eat at mealtimes
• You feel frustrated when something does not go your way
• You feel like running away
Adolescent Overburden/Responsibility Measures
• Having too many responsibility
• Taking care of family members
• Not having enough money for things such as clothing and food
• Concerns about health of a family member
• Problems with friends and neighbors
• Problems getting along with your family
• Feeling unsafe in your neighborhood
• Feeling unsafe in your home
Child Worry/Stress Measures
• Felt worried
• Felt unhappy
• Refused to eat at mealtimes
• Felt frustrated
• Felt like running away from home
• Have nightmares
Child Overburden/Responsibility
• Having too many responsibilities
• Taking care of family members
• Not having enough money for things, such as clothing or food
• Having problems with friends and neighbors
• Having problems getting along with your family
• Feeling unsafe in your neighborhood
• Feeling unsafe in your home
188
Appendix L: Village Level Regression Results
Livestock
Assets
Village with WISDOM (0 = No, 1 =
Yes)
-0.210***
ADP (0 = Guraghe, 1 = Wonchi)
-0.502***
Occupation (0 = Not Currently Working)
Farmer
Petty Trading
Trader
Artisan
Professional/Civil Servant
Retiree
Other
0.098
-0.122
-0.174*
-0.015
-0.084
0.227***
-0.26
Income Generating Activities
0.078
# of M Adults
# of F Adults
# of M Children
# of F Children
0.016
0.043
0.031
0.246***
Orphan Status (0 = Maternal)
Paternal
Both Parents
-0.286***
-0.150
Assistance Beyond Family?
Receives Money?
Send Money?
Sell Assets in Past Yr?
Chronic Illness in HH?
Percent of School Aged Children in CIP
F-stat
Adjusted R2
Observations
----0.014
---------0.013
-0.006
2.939***
0.197
151
Note: All entries are standardized coefficients. Newer clients participated in microfinance for less than
a year. Longer clients participated in microfinance for greater than one year and up to 6 years. All
VIF scores tested below or equal to 3.751.
189
Adol
Self-Esteem
Adol
Burden
Village with WISDOM (0 = No, 1 =
Yes)
0.171*
0.187*
ADP (0 = Guraghe, 1 = Wonchi)
0.032
0.166
-0.139
0.134
-0.002
-0.490***
-0.161
-0.012
-0.049
-0.125
0.029
-0.461***
-0.077
0.043
Caregiver Educated?
0.087
-0.133
# of M Adults
# of F Adults
# of M Children
# of F Children
-0.014
0.102
-0.003
-0.118
-0.076
-0.008
-0.033
-0.060
Marital Status (0 = Married)
Separated
Widowed
Never Married
-0.051
-0.114
-0.035
0.031
-0.115
0.077
Assistance Beyond Family?
Receives Money?
Send Money?
Sell Assets in Past Yr?
Chronic Illness in HH?
Percent of School Aged Children in CIP
----0.121
-0.338**
-----0.118
0.007
-------------0.146
0.081
0.036
F-stat
Adjusted R2
Observations
2.256***
0.199
102
2.531***
0.222
103
Occupation (0 = Not Currently Working)
Farmer
Petty Trading
Trader
Professional/Civil Servant
Retiree
Other
Note: All entries are standardized coefficients. Newer clients participated in microfinance for less than
a year. Longer clients participated in microfinance for greater than one year and up to 6 years. All
VIF scores tested below or equal to 5.529.
190