Content uploaded by Roy Mersland
Author content
All content in this area was uploaded by Roy Mersland on May 21, 2019
Content may be subject to copyright.
What are savings groups?
A description of savings groups based on
information in the SAVIX database
Authors: Roy Mersland, Bert D’Espallier, Rolando Gonzales & Linda Nakato
Global Indicators of Savings Groups
1
Presentation
For the past decade, the FAHU Foundation has been dedicated to supporting Savings Group projects with a
specific emphasis both on “pilot” projects that test new methods, utilize new technologies, or reach new target
groups and research projects that would lead to advancements in the field. The Foundation finds Savings
Groups to be a proven low cost, effective method of reducing poverty that puts the power for transforming lives
in people’s own hands,
FAHU is proud to have collaborated with Hugh Allen of VSL Associates, Professors Roy Mersland and Bert
D’Espallier of the University of Agder and KU Leuven, respectively, as well as Ph.D. candidates Rolando
Gonzales and Linda Nakato to bring you this analysis of the SAVIX. It is our sincere hope that this report that
utilized the vast dataset of a quarter of a million Savings Groups across nearly 50 countries, provides
practitioners, facilitating agencies, and donors with valuable insights that will help with the design and
implementation of successful Savings Group projects, and ultimately help spread and scale this model to reach
even more of the current one billion people with no access to financial services.
Norway/Denmark 21th of May 2019
Annikka Berridge
President
FAHU Foundation
Global Indicators of Savings Groups
2
The CERSEM team in charge of the research
The Center for Research on Social Enterprises and Microfinance (CERSEM) at the University of Agder in Norway
(https://cersem.uia.no/) is in charge of the research being done on the SAVIX database. CERSEM is one of
Europe’s leading research centers on social enterprises and microfinance including savings groups. Hugh Allen
and other associates at VSL have facilitated access to the SAVIX data and FAHU foundation is sponsoring the
research. The results presented in this report are the responsibility of CERSEM only.
Roy Mersland, Bert D’Espallier, Linda Nakato and Rolando Gonzales are the main persons of the CERSEM
team in charge of analyzing the SAVIX database. Roy Mersland, Director of PhD education at the School of
Business and Law at the University of Agder in Norway and director of CERSEM, is the head of the SAVIX
research project. In addition to his academic career Mersland has long practical experience in designing,
initiating and evaluating savings group projects. For more information about the SAVIX database please contact
one of the team members.
Professor
Roy Mersland
Project manager
roy.mersland@uia.no
Professor
Bert D’Espallier
Deputy leader
bert.despallier@kuleuven.be
Rolando Gonzales
Ph.D. candidate
rolando.gonzales@uia.no
Linda Nakato
Ph.D. candidate
linda.nakato@uia.no
Global Indicators of Savings Groups
3
Contents
The CERSEM team in charge of the research.............................................................................. 2
List of Figures .................................................................................................................................. 4
List of Tables ................................................................................................................................... 4
I. Introduction ......................................................................................................................... 5
II. Savings Groups and the SAVIX database ........................................................................ 6
What is a Savings Group? ............................................................................................................. 6
The SAVIX project and the SAVIX database .............................................................................. 8
III. Overview of the SAVIX database .................................................................................... 10
Distribution of Savings Groups in the SAVIX database ............................................................ 10
Group characteristics .................................................................................................................. 13
Donors and Facilitating Agencies .......................................................................................... 13
Group Formation ...................................................................................................................... 15
Group Status .............................................................................................................................. 16
Group Composition .................................................................................................................. 16
The Integration of Additional Services and Linkage to Formal Financial Institutions .......... 18
Integration of Additional Services .......................................................................................... 18
Linkage to Formal Financial Institutions ................................................................................. 20
IV. Performance of Savings Groups ...................................................................................... 23
Portfolio Indicators........................................................................................................................ 23
Financial performance ratios ..................................................................................................... 26
Measuring Return on Savings and Return on Assets ............................................................ 26
v. Further Research ............................................................................................................... 29
References ................................................................................................................................... 31
Appendix I: Variables in the database ..................................................................................... 32
Appendix II: Detailed tables ...................................................................................................... 34
Global Indicators of Savings Groups
4
List of Figures
Figure 1: Savings Groups by Year and Geographical Location in the SAVIX Database ........ 9
Figure 2: Country Concentration of Savings Groups in the SAVIX Database ......................... 11
Figure 3: Geographical Distribution of SGs in Africa and Asia .................................................. 12
Figure 4: Regional Distribution of SGs ............................................................................................ 13
Figure 5: Donors and Agencies ...................................................................................................... 14
Figure 6: Group Formation .............................................................................................................. 15
Figure 7: Group Status ..................................................................................................................... 16
Figure 8: Provision of Additional (plus) Services ........................................................................... 20
Figure 9: Financial Linkages ............................................................................................................ 22
List of Tables
Table 1: Information about Members in Savings Groups ........................................................... 17
Table 2: Balance Sheet (in US dollars) ........................................................................................... 24
Table 3: Other Financial Measures................................................................................................. 25
Table 4: Performance Measures .................................................................................................... 26
Global Indicators of Savings Groups
5
I. Introduction
For centuries, savings groups have been instrumental in the financial lives of rural and semi-
urban populations. With increasing popularity, these types of groups have attracted the attention
of several development agencies and researchers interested in financial inclusion and
development finance. However, given the informal nature of these groups, information about the
group composition, characteristics and performance has been insufficient.
The Savings Groups Information Exchange (SAVIX) database is the first platform of its kind to
provide extensive information on a large number of savings groups. The SAVIX was developed by
Village Savings & Loan (VSL) Associates with funding from the Bill & Melinda Gates Foundation
and facilitating agencies including CARE, Catholic Relief Services, Oxfam America and Plan
International.
1
This platform reports standardized information on savings groups in an unbalanced panel
structure from across the globe. At the end of 2017, the platform contained information on over
276,000 groups in 44 different countries. With financial support from the FAHU foundation, the
Center for Research on Social Enterprises and Microfinance (CERSEM, University of Agder,
Norway) has entered into an agreement with VSL Associates to conduct academic research on
the “economics” of savings groups reporting to the SAVIX database.
The purpose of this report is to give an overview of the contents of the SAVIX database as of
December 2017. Starting with a brief background of savings groups,
2
the report provides
information on the geographical distribution of the savings groups in the database and group
characteristics, such as group status, composition and group formation. It further gives a
snapshot of interventions into these savings groups – such as linkages to a formal financial
institution and inclusion of other development initiatives, like health care and Income Generating
Activities (IGAs) in the groups. Additionally, several portfolio and performance indicators are
analyzed to give an idea of the sustainability and efficiency of the savings groups. The main idea
of this report is to provide an overview of the outspread of savings groups and the data available
1
http://www.thesavix.org/home/
2
For a more detailed description of savings groups, see Allen, H., & Panetta, D. (2010).
Global Indicators of Savings Groups
6
in the SAVIX database. Hopefully this can motivate rigorous academic research relevant for policy
makers and practitioners of the savings group methodology.
II. Savings Groups and the SAVIX database
What is a Savings Group?
The World Bank global Findex (Demirguc-Kunt, Klapper, Singer, Ansar, & Hess, 2018) reports
that over 1.7 billion people are still unbanked with no access to any form of formal financial
services. Traditional banks and microfinance institutions have failed to meet the needs of such
populations. However, through informal systems based on mutual relationships, people form
groups, save in a common “pool” through these groups, and when the funds accumulate,
members can then take loans from this common “pool” and pay back with an interest after an
agreed time period. Such groups are referred to as “Savings Groups.”
Savings groups are often denominated Village Savings and Loan Associations (VSLAs). They are
grassroot, community-based organizations made up of usually 10 to 30 self-selected members
(Allen & Panetta, 2010).
The pooling of resources and savings among village members is a practice that goes back for
centuries (Bouman, 1995). Thus, modern savings groups like those reporting to SAVIX build on
an ancient, self-managed banking model often referred to as Rotating Savings and Loan
Associations (ROSCAs) or Accumulating Savings and Loan Associations (ASCAs). The popularity
of savings groups is therefore not surprising, as villagers have a long history of participating in
groups like Susus, Tontines, Merry Go Round, etc. For example, Bouman (1995) reports that 50%
of the adult population in Congo and between 50-95% in rural areas of Liberia, Ivory Coast, Togo,
Nigeria and Cameroon participate in some form of savings groups. This popularity further
signifies the importance and persistence of informal finance in emerging economies, as
observed by several scholars (Madestam, 2014; Tsai, 2004; Goedecke et al. 2018).
Over the past few decades, development organizations like CARE International, Plan
International and many others have found savings groups to be an ideal way of assuring
sustainable and low-cost financial inclusion of vulnerable, and often rural, populations. The
Global Indicators of Savings Groups
7
savings group models that these organizations use can vary, but generally, they resemble the
ancient ASCA model; the main difference is that in modern savings groups, the lifecycle of the
group is bounded and agreed upon beforehand. This is commonly referred to as the cycle of a
group. The reason for being time bound is that it enables a kind of “live” auditing of the groups,
since all available resources have to be declared and distributed among the members at the end
of the cycle, which normally lasts one year. Hence, at the beginning of the cycle, groups are
constituted, group leaders are elected and a constitution to govern the group activities is
established. The constitution stipulates the contribution by each member, the meeting
frequency, penalties for member indiscipline, procedure to be followed to request for a loan,
interest rates on borrowing etc. Savings and loan activities then take place throughout the cycle
with the accumulated savings and accrued interest being shared out at the end of the cycle.
These groups are transparent, simple and flexible enough to stand even in the midst of both
macro-economic and political changes. Group activities take place during group meetings in the
presence of the members. Normally every member keeps a basic passbook that details their
saving and loan transactions. Collected savings are kept in a “lock box,” which usually has three
locks to ensure safety of the funds. The keys to the lock box are kept by three different persons,
who are chosen by the group members, and this box is only opened during the group meetings.
Some SGs also keep a social fund alongside the usual funds in the lock box. Equal contributions
are made by each member in a smaller amount than what members contribute for the loan fund,
and this amount is meant to meet members’ needs in case of an emergency, such as sickness
or death of a family member, school fees and loss of crops due to drought.
In addition to the financial inclusion agenda pursued through savings groups, many development
actors also try to pursue other non-financial agendas aimed at improving the quality of life for
the SG members. Such non-financial activities are often referred to as “plus” activities and may
include female empowerment, business education, health programs and many other
development initiatives.
The first donor promoted savings groups – i.e., those providing data to SAVIX – were initiated by
CARE International in the early 1990s in Niger. Owing to the success and sustainability of these
groups, other development agencies, often referred to as facilitating agencies, started promoting
variants of savings groups based on the CARE model. These variants include the Saving for
Global Indicators of Savings Groups
8
Change (SfC) by Oxfam/Freedom From Hunger/Strømme Foundation (which uses memorization
for record keeping), Saving and Internal Lending Community (SILC) by Catholic Relief Services,
(which uses ledgers) and , Plan International, the Aga Khan foundation and World Vision (which
use passbooks). Currently, it is estimated that there are over one million facilitated SGs with a
total of 20 to 30 million members globally (Seel, 2018).
3
The SAVIX project and the SAVIX database
The SAVIX is an ongoing project where groups continue to upload their information at
http://www.thesavix.org/. The database contains quarterly information of savings groups from
the first quarter of 2010 to the last quarter of 2017. The dataset is an unbalanced panel, since
groups do not always upload their information every quarter or may stop reporting after some
time. Hence, the dataset does not contain the same number of groups each year: for the year
2010, the database has information of 46,970 savings groups from Africa and 2,845 groups
from Asia, while in 2017 the database has information of 177,486 groups from Africa, 6,430
from America and 15,948 from Asia. The distribution of savings groups across continents and
years is presented in Figure 1.
This report was elaborated with the SAVIX data at the group level aggregated over time. The data
was cleaned for duplicate group names, and groups with less than 3 members and more than
100 members were excluded from the database, thus leaving 261,633 savings groups for
analysis. The upper limits of the following variables were trimmed based on expert criteria: value
of total assets, cumulative value of loans, number of loans in the cycle, seed capital, cash in
other funds, number of loans outstanding, cash in a box, net value of savings and equity. The
performance indicators return on assets (ROA), returns on savings (ROS), equity per member,
assets per member, average savings per member and the savings-to-loan ratio were trimmed at
the 1 and 99 percentiles. The list of all variables available in the SAVIX dataset as well as the
variables’ definitions can be found in Appendix I at the end of this document.
3
In addition there are millions of informal groups as well as millions of externally promoted groups like
for instance around 10 million Self Help Financial Groups backed by the National Bank for Agriculture
and Rural Development (NABARD)in India.
Global Indicators of Savings Groups
9
Figure 1: Savings Groups by Year and Geographical Location in the SAVIX Database
Regional distribution and total number of
saving groups (panel data)
Africa
Americas
Asia
Europe
2010
46970
0
2845
0
2011
74260
0
7642
0
2012
52834
124
4576
631
2013
64733
3856
6880
631
2014
91736
5943
10608
639
2015
140723
5712
13164
0
2016
166427
6189
16265
0
2017
177486
6430
15948
0
Continent Location of Savings Groups
(collapsed data at group level)
Freq.
Percent
Cum.
Africa
228863
87.48
87.48
Asia
24547
9.38
96.86
Americas
7599
2.9
99.76
Europe
622
0.24
100
Total
261,631
Note: The SAVIX database is a panel dataset with information of savings groups from four
continents (Africa, Americas, Asia and Europe) across seven years (2010 to 2017). Figure 1 (left)
shows that the panel SAVIX contains information mainly from Africa, starting with 46,970 groups
in 2010 and increasing to 177,486 groups in 2017. The panel dataset was aggregated over time
at group level, using the name of group as identifier, duplicates were cleaned and groups with less
than 3 members and more than 100 members were excluded from the database, thus leaving
261,633 savings groups for analysis. Figure 1 (right) shows that the savings groups in the SAVIX
are located mainly in Africa (88%) and in Asia (9.4%).
88%
9%
3%
Africa
Asia
Americas
Europe
Global Indicators of Savings Groups
10
III. Overview of the SAVIX database
Distribution of Savings Groups in the SAVIX database
SGs represented in the dataset are from 4 continents (Africa, Asia, Americas and Europe). As the
SG model was mainly grounded in Africa, the majority of the groups in the dataset are in Africa
(228,862 groups), representing 88% of the groups. This is followed by Asia (24,543), Americas
(7597) and Europe (622 groups).
Figures 2, 3 and 4 provide information about geographical locations of the groups reporting to
SAVIX. As seen in figure 2, Uganda is home to the largest number of groups reporting to SAVIX
with 25,397 groups, closely followed by Mali (24,218), Tanzania (22,887), Burkina Faso
(14,983), Ghana (13,680), Ethiopia (11,396), Mozambique (10,801) and Senegal (10,786).
Together, these 8 countries represent over 50% of the groups in the dataset.
The savings groups from the Americas are mainly from Colombia (7,145 groups, 3% of the
database), and the groups from Asia are mainly from Afghanistan (8,214 groups, 3%), India
(4,419 groups, 1.7%), Cambodia (3,611 groups, 1.4%) and the Philippines (3,398 groups, 1.3%).
See Appendix II at the end of the document for details about the number of savings groups in
each country of the SAVIX database.
Generally, the largest percentage (65%) of groups is located in the rural areas of the different
countries. This is in line with expectations as the savings groups method is mostly presented as
a rural methodology for donor-funded financial inclusion and included development efforts.
Actually, only 2% of the projects reporting to the SAVIX database operate in urban areas only,
while 33% operate in both rural and urban areas.
Global Indicators of Savings Groups
11
Figure 2: Country Concentration of Savings Groups in the SAVIX Database
Note: In the figure above, the size of the box is proportional to the number of savings groups in
each country, based on the information of 261,633 savings groups analyzed in this report. Larger
boxes represent countries with a high number of savings groups in the SAVIX database. Countries
from Africa are colored in dark blue, countries from Americas are colored in orange and countries
from Asia are colored in green. The rest of the countries are colored in purple boxes. The graph
shows the high presence of savings groups from African countries in the SAVIX database,
particularly Uganda (25,397 groups), Mali (24,218 SGs), Tanzania (22,887 SGs), Burkina Faso
(14,983 SGs), Ghana (13,680 SGs), Ethiopia (11,396 SGs), Mozambique (10,801 SGs) and
Senegal (10,786 SGs). The savings groups from the Americas are mainly from Colombia (7,145
groups, 3% of the database), and the groups from Asia are mainly from Afghanistan (8,214 groups,
3%), India (4,419 groups, 1.7%), Cambodia (3,611 groups, 1.4%) and the Philippines (3,398
groups, 1.3%). See Appendix II at the end of the document for details about the number of savings
groups in each country of the SAVIX database.
Global Indicators of Savings Groups
12
Figure 3: Geographical Distribution of SGs in Africa and Asia
Location of Savings Groups in Africa
Freq.
Percent
Cum.
Eastern Africa
114,759
50.14
50.14
Western Africa
102,250
44.68
94.82
Northern Africa
5,983
2.62
97.43
Southern Africa
3,256
1.42
98.86
Central Africa
2,615
1.14
100
Total
228,866
Location of Savings Groups in Asia
Freq.
Percent
Cum.
Southern Asia
13,935
56.77
56.77
Southeast Asia
7,949
32.38
89.15
Central Asia
2,261
9.21
98.36
Eastern Asia
402
1.64
100
Total
24,547
Note: The figures above show the regional location of savings groups from Asia and Africa. In
Africa, 114,759 groups (50%) are located in Eastern Africa and 102,250 groups (45%) in Western
Africa. In the case of Asia, 13,935 groups are from Southern Asia, representing 57% of the Asian
savings groups.
50%
45%
3%
1%
1% Eastern Africa
Western Africa
Northern Africa
Southern Africa
Central Africa
57%
32% 9%
2%
Southern Asia
Southeastern Asia
Central Asia
Eastern Asia
Global Indicators of Savings Groups
13
Figure 4: Regional Distribution of SGs
Regional Location of Savings Groups
Freq.
Percent
Cum.
Rural
133,939
64.6
64.6
Rural+Urban
68,781
33.2
97.8
Urban
4,533
2.2
100
Total
207,253
100
Note: In the database, 54,387 (21%) of the groups in the SAVIX do not have a registered location
in the database. From the remaining 207,253 groups, 65% are located in rural regions, 33% in
urban-rural regions and only 2% (4,533 groups of the database) are located in urban regions.
Group characteristics
Donors and Facilitating Agencies
As seen in Figure 5, the majority of the groups in the dataset are supported by Plan International
(about 28% of the groups).
4
This is followed by CARE (18%), the Aga Khan Foundation (11%),
Catholic Relief Services (10%), Oxfam (10%) and World Vision (9%). Together, these agencies
support about 85% of the groups. Notably, groups with no facilitating agency comprise about 7%
(19,035 groups).
Usually, there are donors that provide funding to back the work that facilitating agencies do with
the SGs. Table 5 shows that majority of the support comes from the Mastercard Foundation
4
The database also contains information at the project level, but these data are not reported here.
65%
33%
2%
Rural
Rural+Urban
Urban
Global Indicators of Savings Groups
14
(27%), closely followed by the Bill & Melinda Gates Foundation (24%). These two organizations
represent about 51% in donor support. Other donor organizations include Barclays Bank (10%),
Canadian International Development Agency-CIDA (8%) and United States Agency for
International Development-USAID (6%).
Figure 5: Donors and Agencies
Donors to Savings groups
Freq.
Percent
Cum.
Master Card Foundation
38972
26.86
26.86
Bill & Melinda Gates Fdt.
35377
24.38
51.25
Barclays Bank
14291
9.85
61.1
CIDA
10913
7.52
68.62
USAID
7786
5.37
73.99
Others
37743
26.01
100
Total
145,082
100
Facilitating agencies
Freq.
Percent
Cum.
Plan International
74092
28.32
28.32
CARE
47657
18.22
46.53
Aga Khan Foundation
27995
10.7
57.23
Catholic Relief Services
25221
9.64
66.87
Oxfam
25097
9.59
76.47
World Vision
23787
9.09
85.56
No Facilitating Agency
19035
7.28
92.83
Other agencies
18,747
7.17
100
Total
261,631
100
Note: Information reported is based on groups that report on that variable. Master Card
Foundation and the Bill & Melinda Gates Foundation are the main donors of savings groups
registered in the database; together, they represent 51% of the registered donors. There are
116,558 savings groups in the database that lack of a registered donor. In terms of facilitating
agencies, Plan International (28%) and CARE (18%) represent 46% of the registered facilitating
agencies of the savings groups in the SAVIX database.
27%
24%
10% 8% 5%
26%
Master Card Foundation
Bill & Melinda Gates
Foundation
Barclays Bank
CIDA
USAID
Others
28%
18%
11%
10%
10% 9% 7%
7%
Plan International
CARE
Aga Khan Foundation
Catholic Relief Services
Oxfam
World Vision
No Facilitating Agency
Other agencies
Global Indicators of Savings Groups
15
Group Formation
Usually, a facilitating agency hire a local NGO to mobilize and train groups. The local NGO in
charge of a SG project use either a field officer or a village agent in the formation and monitoring
of the groups. Field officers are normally hired staff, while village agents are normally not formally
hired and are typically savings group members selected from within the community. While some
village agents may work for free, our experience is that they are typically paid some form of small
incentives, either by the local NGO (funded by the facilitating agency) or, more usually, by the
savings group itself. Moreover, the SG model frequently self-replicates, with some groups
springing up spontaneously in villages as people witness the benefits that come from being a
part of such groups.
In general, as reported in Figure 6, the local NGOs mainly use field officers who represent 52%
with village agents closely following with 40%. Notably, groups that form spontaneously represent
1% (2,774 groups) only. These are groups formed by the members without any external support
whatsoever. It should be noted, however, that the number of spontaneously formed group is in
practice much higher, but such groups will typically not report to the SAVIX database.
Figure 6: Group Formation
Group formation
Freq.
Percent
Cum.
Field officer
136,962
52.4
52.4
Village agent
104,775
40.1
92.6
Other
16,651
6.4
98.9
Spontaneous
2,774
1.1
100
Total
261,162
100
Note: The information reported is based on groups that report on that variable (261,162 groups).
In the SAVIX database, 136,962 (52%) were formed by field officers and 104,775 groups (40%)
by village agents. Also, 2,774 groups (1%) were spontaneously formed according to the records
of the dataset.
Global Indicators of Savings Groups
16
Group Status
In the earlier stages of a savings group (at its formation), the local NGO directly and continuously
monitors the group activities either through the field officer or a village agent. As a group enters
into successive cycles, it continues to be tracked by the facilitating agency until the group
graduates into self-management. In the dataset, about 63% of the groups are self-managed,
which implies that they are still surviving even after the direct support from the local NGO and
facilitating agency phases out. This normally happens after having finished one full cycle. This
speaks to the sustainability of the savings group model. The remaining 37% are supervised
groups.
Figure 7: Group Status
Group status
Freq.
Percent
Cum.
Self-managed
163647
62.5
62.5
Supervised
97993
37.5
100
Total
261,640
100
Note: Self-managed groups represent 62.5% of the savings groups in the database. This
category includes groups that are active and graduated. Supervised groups represent 37.5% of
the database and include tracked groups.
Group Composition
This section presents the composition profile of the savings groups. On average, as we can see
in Table 1, the typical group is composed of 22 members. Savings groups were traditionally
aimed at women, and this is also evidenced in the composition metrics of the SAVIX groups. As
many facilitating agencies have a particular focus on women, this is not a surprising picture.
Taking an example: Oxfam, Freedom from Hunger, and Stromme Foundation, through their
Saving for Change project, mainly aim at catering for women.
5
5
See https://policy-practice.oxfamamerica.org/work/rural-resilience/saving-for-change/
63%
37% Self-managed
Supervised
Global Indicators of Savings Groups
17
Table 1: Information about Members in Savings Groups
Gender composition of savings groups
Number of
members
Male members
Female members
Mean
21
4
17
Median
22
3
17
Standard deviation
7
5
7
Min
3
0
0
Max
100
91
100
Percentile 5
10
0
5
Percentile 10
12
0
7
Percentile 25
16
0
12
Percentile 75
26
8
23
Percentile 90
30
12
26
Percentile 95
30
15
29
Group dynamics
Members at
start of the
cycle
Dropouts
since start of
the cycle
Members
attending
meetings
Mean
21
1.98%
90.94%
Median
20
0%
95.33%
Standard deviation
7
7.39%
11.89%
Min
3
0%
8.57%
Max
100
100%
100%
Percentile 5
10
0%
66.67%
Percentile 10
12
0%
74.51%
Percentile 25
15
0%
86.00%
Percentile 75
25
0%
100%
Percentile 90
30
5%
100%
Percentile 95
30
12.24%
100%
Note: Savings groups in the SAVIX database tend to have 5 to 30 members and are mainly
composed of women. The groups in the database tend to start the cycle also with 21 members on
average, present a low percentage of drop-outs (2%) and show a high-attendance of the members
to the group meetings (91%).
Global Indicators of Savings Groups
18
Even if SGs are presented as a “female activity,” numbers show that men make up a surprisingly
high share of the groups. Truly, women make up the large majority. In fact, in 87% of the groups,
women are the majority. However, on average, 50% of the groups have at least 3 men in their
membership.
Table 1 also indicates stable membership with a low number of members dropping out of the
groups. Specifically, 75% of the groups experience no drop-out during the cycle, and 95% of the
groups report about 12% drop-out or less during the cycle. This is a good indication of the
relevancy of savings groups to the members.
Moreover, with respect to meeting attendance, on average, the attendance rate is 91% and in
25% of the groups, the attendance rate is 100%. Our experience, which has also been confirmed
by Hugh Allen in personal conversations with him, is however that attendance rates are
somewhat lower, maybe 85% on average, since field officers are reluctant to admit that members
don’t show up for meeting. Nevertheless, attendance rates are very high and this points to the
level of discipline exercised by the group members and further evidences the importance of these
groups to the members. It shows the importance in terms of potential empowerment and
community building.
The Integration of Additional Services and Linkage to Formal
Financial Institutions
Integration of Additional Services
Increasingly, owing to their sustainability and ability to reach the most rural dwellers in a cost-
efficient manner, facilitating agencies combine savings and credit activities with other
development initiatives (usually referred to as “plus” services). These include women
empowerment, health (e.g., HIV programs and sanitation programs), agricultural training,
guidance in Income Generating Activities, programs for orphans and vulnerable children, to
mention but a few. CARE, for example, has added agricultural inputs onto SGs in Zimbabwe, and
Catholic Relief Services has added agricultural marketing in Tanzania, Pact combines several
Global Indicators of Savings Groups
19
women’s empowerment programs with SGs in Nepal. In Mali and Senegal, Oxfam and Freedom
from Hunger champion malaria prevention education alongside SGs.
Regardless of the popularity in using SGs as platforms for other development efforts, Figure 8
shows that 65% of the groups in the SAVIX dataset do not report any plus services. Of those,
including plus services in the groups, the majority (30%) include business and/or financial
education, 18% include income generating activities, 9% add women’s empowerment, 7%
include health related services and around 6% include employment skills, while 5 % add climate
change knowledge to the groups’ activities.
Global Indicators of Savings Groups
20
Figure 8: Provision of Additional (plus) Services
Additional (plus) services provided to Savings Groups
Freq.
Percent
Cum.
Without plus services
170,369
65.1
65.1
With plus services
91,271
34.9
100
Total
261,640
100
Freq.
Percent
Cum.
Business-financial education
27,301
29.9
29.9
Others
23,281
25.5
55.4
Income generating activities
16,002
17.5
73.0
Women's empowerment
8,605
9.4
82.4
Health
6,061
6.6
89.0
Employment skills
5,098
5.6
94.6
Climate change
4,923
5.4
100
91,271
100
Note: From the total 261,640 savings groups, 91271 (35%) have a record of an additional (plus)
service provided to the group, related mainly to business and/or financial education (30%) and
income generating activities (18%). The category women’s empowerment (9%) includes services
focused on women, such as gender justice, governance, leadership and literacy, while health
(6.6%) includes the strengthening of health systems and considers also nutrition, sanitation and
HIV/AIDS protection. Employment skills (5.6%) covers the development of entrepreneurship skills
and micro-enterprises, while climate change (5.4%) relates to adaptation activities to climate
change, disaster management and mitigation, food security and the development of low carbon
technologies.
Linkage to Formal Financial Institutions
Most savings groups use only internally generated funds (Allen & Panetta, 2010). However, given
the nature of member contributions, there is sometimes a higher demand for loans than what
members are able to save in the loan fund. Moreover, as reported by Burlando, Canidio, & Selby,
Global Indicators of Savings Groups
21
(2016), many groups decide not to lend out all the savings and instead often keep considerable
amounts of cash in their cash-boxes. This can be when the group considers it too risky to lend
out to members, or when lenders have a low appetite for borrowing based on for example
seasonal fluctuations in demand. Moreover, excessive cash in a group is particularly the case
towards the end of the cycle, when loans are being repaid. Such excessive cash in the group box
poses security risks for the group. For these reasons, SGs are sometimes seeking to link to
financial institutions either in the form of opening a group bank account or by taking a group loan
from a financial institution in order to increase the group’s lending opportunities. Examples of
linkage projects include the Banking on Change project. This project was initiated as a
collaboration between Barclays Bank, CARE and Plan International between 2009 and 2015;
this collaboration aimed to increase access to formal financial services by SGs in Egypt, Ghana,
India, Kenya, Tanzania, Uganda and Zambia. Other interventions include efforts by CARE in
Rwanda (i.e., linkages with Umurenge Savings & Credit Cooperatives and Vision Finance
Company).
Until now, it has been unclear if such linkages are beneficial or detrimental to SGs. Some
scholars (Ghate, 1992; Pagura & Kirsten, 2006; Piprek, 2007) argue that linkage with formal
financial institutions are beneficial for improving access of financial services to the rural areas
as they tap into the benefits accruing to both sectors – i.e., SGs can get access to a wide range
of resources from the formal sector, while the formal sector could tap into the information that
informal groups have on the rural clientele to design services that fit poor people’s needs.
However, some viewpoints
6
suggest that linkages should be practiced with caution. This is
because there is a concern that linkages may damage the social systems that bind members
together and disrupt group dynamics, thus destroying the very flexibility and adaptability upon
which groups are founded (Bouman, 1977; Dercon, De Weerdt, Bold & Pankhurst 2006; Aliber,
2002).
Nevertheless, as reported in Figure 9, such linkages are still not common. Only 6% (15,604
groups) of the SGs in the dataset are linked to a formal financial institution. The remaining 94%
have no such relationship, neither savings or loans, with a formal financial institution. As shown
6
See https://www.cgdev.org/blog/conversation-hugh-allen-village-savings-loan-associations
Global Indicators of Savings Groups
22
further in Figure 9, majority of the linked groups (67%) are linked by only credit, 30% by only
savings and 3% by both credit and savings.
Figure 9: Financial Linkages
Financial linkages of Savings Groups
Freq.
Percent
Without financial linkages
246,036
94.04
With financial linkages
15,604
5.96
Total
261,640
100.00
Type of linkage
Freq.
Percent
Credit linkages
10,499
67.28
Savings linkages
4,674
29.95
Credit & savings linkages
431
2.76
Total
261,640
100
Note: Financial linkages were measured considering the debts and bank balance of the savings
groups in the database. Groups with positive debts were classified as having a credit linkage, while
groups with positive bank balance were classified as having a savings linkage. Groups that have
both, positive debts and positive bank balance, were classified as having credit and savings
financial linkages. The figure shows that only 6% of the savings groups in the database have
financial linkages, 4% corresponding to credit linkages, 1.8% to savings linkages and 0.2% to both.
94%
6%
Without financial
linkages
With financial linkages
67%
30%
3%
Credit linkages
Savings linkages
Credit and savings
linkages
Global Indicators of Savings Groups
23
IV. Performance of Savings Groups
Portfolio Indicators
Savings groups have assets, liabilities and capital/equity. In a typical group, assets consist of
cash (which is kept in a cash box including the emergency fund), possible bank balance and
loans to the members. A typical group (median group) has $87.90 US dollars in the cash box.
Some groups keep much larger amounts in the cash box (the maximum amount being $4,745.8)
making the average amount in the cash box $278.
Some groups keep an emergency fund (cash in other funds), which covers group members in
times of emergencies like drought, school fees and sickness or death of a loved one. The cash
in other funds of a typical group is $10. About 25% of the groups do not keep an emergency fund
or have no cash in the emergency fund.
The third component of cash consists of cash deposited into a bank account (bank balance). Few
groups actually have such a bank account. As reported in Table 2, the average SG keeps only
$4.30 in a bank account. The small number is a result of the fact that only about 2% of the
groups have a bank account.
The second and most important group of assets is, of course, the money that is lent out to group
members. These are funds borrowed from the group at a charge (interest). The average amount
outstanding is $304.70, while the median group lent out $140.30 to its members. As many as
10% of the groups have no loans outstanding (see percentile 25 in Table 2). This is further
reaffirmed by the number of loans outstanding in Table 3, suggesting that 23% of the groups
seem not to give out loans or are still early in their cycles and have thus not yet started lending
activities. Hence, when analyzing savings groups it is important to take into consideration how
far into a cycle a group has reached.
As in any bank, the SGs may encounter members having problems in repaying their loans. Thus,
the value of overdue loans must be added to the total assets of the group. However, the results
in Table 2 show that on average a group has only $0.60 in loans past due, and this average is to
a large extent driven by “an outlier” with $1477 in loans past due. In fact, only 1.1% of the groups
report having loans past due. Thus, there is an outstanding repayment record in these savings
Global Indicators of Savings Groups
24
groups. This same repayment discipline is shown by the value of loans write-off in Table 3 which
is only $0.10 for the average group. Only 0.8% of the groups have written off loans. Bad loans
are not common occurrences in SGs. For outsiders trying to understand these numbers it is
important that you keep in mind the close ties between members which make defaulting difficult.
Moreover, most groups practice strict credit screening and only allow members to triple their own
savings when deciding their loan amounts.
Table 2: Balance Sheet (in US dollars)
Cash in
group
box
Cash in
other
funds
Bank
balance
Value of
loans
out-
standing
Value of
loans
past due
Property
now
Debts
Equity
Mean
278.0
23.7
4.3
304.7
0.6
11.6
2.2
667.3
Median
87.9
10.0
0.0
140.3
0.0
0.0
0.0
401.5
Std. deviation
475.1
43.4
45.6
447.5
12.4
39.7
16.5
731.5
Minimum
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
Maximum
4745.8
3426.0
2584.8
4843.0
1477.0
4298.6
2319.3
4883.0
Percentile 5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
23.6
Percentile 10
2.7
0.0
0.0
0.0
0.0
0.0
0.0
52.1
Percentile 25
21.9
0.3
0.0
8.2
0.0
0.0
0.0
151.2
Percentile 75
311.3
30.3
0.0
399.5
0.0
13.0
0.0
927.5
Percentile 90
782.9
61.5
0.0
822.9
0.0
38.0
0.0
1678.9
Percentile 95
1246.9
89.8
0.0
1195.1
0.0
52.2
12.6
2236.9
Note: The upper values of the following variables were trimmed based on expert criteria: value
of total assets, cumulative value of loans, number of loans in the cycle, seed capital, cash in
other funds, number of loans outstanding, cash in a box, net value of savings and equity. Note:
The summarized means of the assets and the liability/equity do not balance because of the type
of trimming applied to the variables.
Global Indicators of Savings Groups
25
Some groups also report having physical assets (represented by “Property now” in Table 2).
These are normally small things like the cash box itself or a calculator, but it could also be seeds,
animals or even a piece of land. The typical group (median) holds no physical assets, while the
average holds physical assets worth $11.60. The largest amount reported in the dataset is
$4,298.
Moving onto the liabilities and capital/equity, savings groups are customarily financed by
members’ contributions and sometimes by loans from external sources, such as formal
financial institutions. Only 4% of the groups have borrowed from other sources outside their
members. The equity for an average savings group amounts to $667.3, while the median group
has $401.5 worth of equity.
Table 3: Other Financial Measures
Net value of
savings
(USD)
Number of
loans out-
standing
Property at
the start of
the cycle
(USD)
Total assets
(USD)
Write-off
since the
start of the
cycle (USD)
Mean
484.2
8.8
3.9
672.2
0.1
Median
290.9
7.5
0.0
390.5
0.0
Std. deviation
532.0
7.9
22.4
752.5
5.7
Min
0.1
0.0
0.0
0.1
0.0
Max
3296.1
74.0
4298.6
5654.6
1139.3
Percentile 5
20.8
0.0
0.0
22.4
0.0
Percentile 10
43.3
0.0
0.0
50.3
0.0
Percentile 25
117.2
1.0
0.0
144.2
0.0
Percentile 75
660.0
14.5
0.0
931.0
0.0
Percentile 90
1205.9
20.3
11.8
1725.6
0.0
Percentile 95
1628.4
23.8
23.4
2304.9
0.0
Note: The upper values of the following variables were trimmed based on expert criteria: value
of total assets, cumulative value of loans, number of loans in the cycle, seed capital, cash in
other funds, number of loans outstanding, cash in a box, net value of savings and equity. The
table shows that on average, the value of savings in the groups is around $484, but the assets
reach $672. The main reason for the difference between value of savings and value of assets is
because the groups have collected interest on loans to members.
Global Indicators of Savings Groups
26
Financial performance ratios
Table 4 & 5 show measurements of the financial performance of SGs calculated with information
from the SAVIX database. Table 4 shows that the average savings and assets per member of the
groups is $21.76 and $30.33 respectively. The collected interests on loans explain most of the
difference between the two numbers.
Table 4: Performance Measures
Equity per member
(USD)
Average savings per
member
(USD)
Average assets per
member
(USD)
Mean
30.22
21.76
30.33
Median
19.49
14.22
19.57
Std. deviation
30.68
21.98
30.69
Minimum
0.02
0.02
0.02
Maximum
161.15
116.55
161.15
Percentile 5
1.21
1.16
1.30
Percentile 10
2.80
2.44
2.96
Percentile 25
7.83
6.11
7.92
Percentile 75
42.29
29.73
42.40
Percentile 90
75.08
52.45
75.21
Percentile 95
98.08
70.00
98.19
Note:The performance indicators shown in the table were trimmed at the 1 and 99 percentiles.
The definitions of these variables can be found in the Appendix I at the end of this document.
Measuring Return on Savings and Return on Assets
Several metrics have been suggested to measure the financial performance of savings groups.
Returns on assets and returns on savings –annualized or unannualized– are among the most
used metrics of financial performance. In this research project, we are particularly interested in
Global Indicators of Savings Groups
27
what set of factors that can influence group performance. Thus, this project will study the effects
of (i) the dynamics, composition and characteristics of the group, (ii) the delivery mechanisms
and facilitation provided by development agencies, and (iii) the macro-economic environment of
the country where the group is located.
In terms of the dynamics, composition and characteristic of the group, variables as the
attendance of the members to the meetings and the gender composition of a group can
potentially influence the financial returns of savings groups if for example member’s attendance
or a larger share of women in a group can help to reduce the default rate of savings accumulation
during the cycle. Also, the number of members that are part of the group can affect the dynamics
of allocation and performance, since larger groups will tend to accumulate larger sums of joint
savings, but at the same time larger groups also favor the possibility of free-riding behavior.
Larger groups can also dramatically reduce the strong ties and the sense of community that are
a characteristic of small-to-moderate size of savings groups, typically of around 20 or 30
members.
The facilitating agencies apply fairly similar though still different methodologies. Thus, the
method applied by the facilitating agencies are interesting to study in more detail. Are for
example village agents a better way of mobilizing groups than field officers? And, in what type of
environment and for what type of groups are field agents most effective? These are some of the
questions that could potentially be studied in the SAVIX database.
The macro conditions (economic, political, social etc.) will also affect the returns and
sustainability of a savings group. For example, are savings groups affected by high levels of
inflation and political or economic instability? Or, is savings groups performance related to formal
banking penetration in a country? Maybe savings group have higher returns in context where
interest rates are high?
In table 5 below, the values for the Return on Savings (ROS) and Return on Assets (ROA) are
presented. The values of the Return on Savings and the Return on Assets (ROA) show that while
SGs tend to have an average return of 33.5% (ROS) or 23.8% (ROA), there is a great dispersion
in the returns of the groups, since some groups have returns that are close to zero, while in the
other extreme, some savings groups in the database have returns above 100%.
Global Indicators of Savings Groups
28
Table 5: Performance Measures
Returns on
Savings
(ROS)
Returns on
Assets
(ROA)
Annualized
Internal Rate
of Returns
(IRR)
Mean
33.5%
23.8%
116.9%
Median
21.2%
18.9%
63.9%
Std. deviation
42.0%
26.9%
197.4%
Minimum
-102.6%
-126.8%
-161.5%
Maximum
288.5%
118.1%
2128.3%
Percentile 5
0.0%
0.0%
0.0%
Percentile 10
0.0%
0.0%
5.4%
Percentile 25
0.5%
0.5%
27.2%
Percentile 75
52.1%
40.8%
122.9%
Percentile 90
89.7%
61.3%
246.4%
Percentile 95
116.7%
72.9%
412.2%
Note: To annualize the performance measures, we limited the sample to groups between 30 and
270 days into the cycle. Days into the cycle were calculated as the difference between date the
data was collected and the date the savings were started for that cycle. The performance
measures were then annualized by diving each financial indicator by the days into the cycle and
multiplying the result by 365. The internal rate of return was calculated by solving for in the
formula , where is the future value of an annuity, is the value of
weekly savings and represents the number of weeks that a group has been operating. The
weekly IRR was annualized by multiplying by 52.The performance indicators shown in the table
were trimmed at the 1 and 99 percentiles. The definitions of these variables can be found in the
Appendix I at the end of this document.
In any case, the numbers show that most groups provide loans to their members at interest rate
levels above typical bank rates. In fact, our experience is that SGs lend at higher rates than those
reported here. Generally speaking, 5% flat monthly rate, which per annum totals above 100%
effective interest, is common. Thus, the lower numbers reported here could be the results of
groups not being strict in interest rate recovery but rather focus on recovery of the principals.
But the levels reported are also highly influenced by the savings group model itself. Many groups
put little emphasis on lending and are “happy” in keeping most of the money in the cashbox.
Global Indicators of Savings Groups
29
Typically savings groups flourish in context with few business opportunities, thus borrowing for
business may not be wise particularly if the groups set interest rates at too high levels. Moreover,
most savings groups stop lending during the last couple of months of a cycle and each new cycle
they normally start from zero. Thus, at the beginning of a cycle no lending can take place and
when the end of a cycle is approaching the groups are only collecting loans and not giving out
new loans.
It is therefore not straightforward neither to understand nor to calculate the financial
performance of a savings group, Hence, in addition to traditional Return on Assets (ROA) and
Return on Savings (ROS) measures Table 5 presents the annualized returns calculated using the
Internal Rate of Return (IRR) approach. The IRR is a typical financial measure that estimates the
actual rate of return at the group-level by means of the well-known future value () of an
annuity-formula: , where the annuity-amount is the value of savings
expressed on a weekly basis and represents the number of weeks that a group has been
operating. Solving the equation for yields an estimate of the effective weekly internal rate of
return generated by the groups which can be converted into an annual rate of return. This
approach to calculating returns allows to quantify the extent to which the different income-
generating channels contribute to overall financial sustainability, because in the business model
of a savings group weekly savings inputs grow into a total asset value throughout the cycle of the
group.
v. Further Research
In summarizing the contents of the SAVIX database, CERSEM has identified several potential
ideas that could be addressed using this database. The benefits of the database (the first of its
kind on community managed microfinance) is that it gives an opportunity to study and compare
SG projects in different contexts. As is the case in any secondary database, however, the SAVIX
database is not free from limitations. First, there is selection bias in the sense that we mostly
observe groups operating under an initiated development program. In addition, only some donors
require facilitating agencies to report to the SAVIX database. Thus, some of the data is reported
on a voluntary basis which can lead to possible reporting problems. Moreover, we mostly observe
groups during an initial stage in their lifecycles. This means that in the analyses, data must be
Global Indicators of Savings Groups
30
treated with the needed caution and potentially additional data will have to be collected. Finally,
working with the SAVIX database is challenging. Not only is it comprised of several relational
tables, it also has three time dimensions, i.e. 1) the time of the year the information is uploaded
to SAVIX (we have grouped this into quarters of the year), 2) the number of cycle a group is in,
and 3) the number of weeks/months the group is into a current cycle.
Nevertheless, we do believe the data would allow researchers to tackle a number of interesting
research questions that, due to data limitations, have not been tackled before. CERSEM invites
other researchers to join and explore the SAVIX database. However, the database is complex to
use, and strong methodological and statistical knowledge as well as a deep understanding of
the saving group “business model” are required for those interested in partnering with CERSEM.
Global Indicators of Savings Groups
31
References
Aliber, M. (2002). Informal finance in the informal economy: promoting decent work among
the working poor (No. 993576903402676). International Labour Organization.
Allen, H., & Panetta, D. (2010). Savings groups: What are they? Washington DC: SEEP
Network, 2.
Bouman, F. J. (1977). Indigenous savings and credit societies in the third world. A message.
Savings and development, 181-219.
Bouman, F. J. (1995). Rotating and accumulating savings and credit associations: A
development perspective. World development, 23(3), 371-384.
Burlando, A., Canidio, A., & Selby, R. (2016). The economics of savings groups.
Demirguc-Kunt, A., Klapper, L., Singer, D., Ansar, S., & Hess, J. (2018). The Global Findex
Database 2017: Measuring Financial Inclusion and the Fintech Revolution. The World
Bank.
Dercon, S., De Weerdt, J., Bold, T., & Pankhurst, A. (2006). Group-based funeral insurance in
Ethiopia and Tanzania. World development, 34(4), 685-703.
Ghate, P. B. (1992). Interaction between the formal and informal financial sectors: The Asian
experience. World Development, 20(6), 859-872.
Goedecke, J., Guérin, I., D'espallier, B., & Venkatasubramanian, G. (2018). Why do financial
inclusion policies fail in mobilizing savings from the poor? Lessons from rural South
India. Development Policy Review, 36, O201-O219.
Madestam, A. (2014). Informal finance: A theory of moneylenders. Journal of Development
Economics, 107, 157-174.
Pagura, M., & Kirsten, M. (2006). Formal—informal financial linkages: lessons from developing
countries. Small Enterprise Development, 17(1), 16-29.
Piprek, G. (2007). Linking with Savings and Credit Cooperatives (SACCOs) to expand
financial access in rural areas: a case study of CRDB Bank in Tanzania. Rural Finance
Group of the Food and Agriculture Organization (FAO) of the United Nations.
Rippey, P., & Fowler, B. (2011). Beyond financial services-a synthesis of studies on the
integration of savings groups and other developmental activities. Geneva: Aga Khan
Foundation.
Seel, G. (2018). Where next after VSLA? A World Renew Research Paper Exploring the Value
of Linking Savings Groups with Financial Services Providers. World Renew, 1-46.
Tsai, K. S. (2004). Imperfect substitutes: The local political economy of informal finance and
microfinance in rural China and India. World Development, 32(9), 1487-1507.
Global Indicators of Savings Groups
32
Appendix I: Variables in the database
Variable
Definition
Active members time at visit
The number of active members of the group
Active men at time of visit
The number of active men in the group
Active women at time of visit
The number of active women in the group
Members at formation
The number of members when the group was formed
Members at start of the cycle
The number of members when the group initiated its financial activities
in the current cycle.
Dropouts since start of the cycle
The number of members who have left the group for any reason -
including death, expulsion, departure, personal reasons or any other
reason - since the start of the current cycle
Number of members attending meetings
The number of members present during the meeting at which the data is
collected
Cash in group box (USD)
The total amount of cash in the loan fund (in the box/bag, deposited in a
financial institution or stored on a mobile device)
Cash in other funds (USD)
The total amount of cash in all other funds (in the box/bag, deposited in
a financial institution or stored on a mobile device)
Bank balance (USD)
The total amount of cash in the group bank account
Value of loans outstanding (USD)
The total value of all balances of loans outstanding
Value of loans past due (USD)
The total remaining value of all loans that are late (i.e. past the agreed-
on date for full reimbursement)
Property now (USD)
The purchase price of physical assets that the group owns at the time of
data collection
Debts (USD)
The total value of group debts to external financial institutions and
individuals
Equity (USD)
Equity was calculated by subtracting groups’ debt to the sum of the cash
kept in a box, bank balance, the value of loans outstanding, property, the
cash in other funds and the value of loans past due.
Cumulative number of loans (this cycle)
The cumulative number of loans with any balance remaining unpaid
(whether on-time or late) in the current cycle
Cumulative value of loans (this cycle)
The cumulative value of all balances of loans outstanding in the current
cycle
Net value of savings (USD)
The total balance of member savings in the current cycle
Number of loans outstanding
The number of loans with any balance remaining unpaid (whether on-
time or late)
Property at the start of the cycle (USD)
The purchase price of physical assets (buildings, livestock, furniture,
agricultural equipment etc.) which the group owned at the start of the
current cycle
Total assets
Assets were calculated as the sum of the value of the cash kept in a
box, the bank balance, the property of the group and the value of loans
outstanding, the cash in other funds and the value of loans past due.
Write-off since the start of the cycle (USD)
The total value of all loans that the group has decided are uncollectible
Returns on Savings (ROS)
Group returns were calculated by adding the value of the cash kept in a
box to the value of the bank balance and the property of the group and
the value of loans outstanding, minus the sum of the net value of
Global Indicators of Savings Groups
33
savings, the property at the start of the cycle and the debts of the group.
These returns were divided by the average of the net value of savings.
Returns on Assets (ROA)
Group returns were calculated by adding the value of the cash kept in a
box to the value of the bank balance and the property of the group and
the value of loans outstanding, minus the sum of the net value of
savings, the property at the start of the cycle and the debts of the group.
These returns were divided by the average of the net value of assets,
which was calculated as the sum of the value of the cash kept in a box,
the bank balance, the property of the group and the value of loans
outstanding.
Internal Rate of Return (IRR)
The IRR measures the actual rate of return at the group-level by means
of the future value () of an annuity-formula:
, where the annuity-amount is the value of savings
expressed on a weekly basis and represents the number of weeks that
a group has been operating. Solving the equation for yields an
estimate of the effective weekly internal rate of return generated by the
groups which can be converted into an annual rate of return.
Equity per member (USD)
Equity was divided by the number of active members in a group to
obtain equity per member.
Average savings per member (USD)
Net value of savings divided by the number of active members in a
group
Average assets per member (USD)
Assets were divided by the number of active members in a group to
obtain assets per member.
Savings to loans ratio
Net value of savings divided by the value of loans outstanding
Risk
Write off since the start of a cycle plus the value of loans past due
Group status
A group may still be receiving training and supervision as a savings
group, in which case it is defined as Supervised, or it may be operating
independently and no longer being trained or supervised, in which case
it will be defined as Self-managed.
Location
Whether the group is domiciled in the rural or urban area
Group formation
A group may be formed by a field officer, village agent or form
spontaneously
Continent
The continent I which the group is located
Agency
The facilitating agency supporting the group
Region
The region in which the group is located
Country
The country in which the group is located
Financial linkages
Whether the group has a joint savings account or joint liability with a
formal financial institution
Provision of additional services (SG plus)
Whether the group provides additional services apart from savings and
credit
Global Indicators of Savings Groups
34
Appendix II: Detailed tables
Table A.1. Facilitating Agencies in the SAVIX database
N
Facilitating Agency
SGs
Percent
Cum. percent
1
Plan International
74,092
28.32
28.32
2
CARE
47,657
18.22
46.53
3
Aga Khan Foundation
27,995
10.70
57.23
4
Catholic Relief Services
25,221
9.64
66.87
5
Oxfam
25,097
9.59
76.47
6
World Vision
23,787
9.09
85.56
7
No Facilitating Agency
19,035
7.28
92.83
8
IED
7,054
2.70
95.53
9
SaveAct
2,496
0.95
96.48
10
World Relief Canada
1,765
0.67
97.16
11
Freedom from Hunger
1,678
0.64
97.80
12
ChildFund
1,127
0.43
98.23
13
Norwegian Association of Disabled (NAD)
1,030
0.39
98.63
14
Global Communities
934
0.36
98.98
15
World Relief
725
0.28
99.26
16
We Effect
560
0.21
99.47
17
Mercy Corps
520
0.20
99.67
18
World Education
211
0.08
99.75
19
PCI
201
0.08
99.83
20
Welthungerhilfe
157
0.06
99.89
21
Feed the Children
139
0.05
99.94
22
PATH
91
0.03
99.98
23
Food for the Hungry
50
0.02
100
24
FHI360
9
>0.01
100
Global Indicators of Savings Groups
35
Table A.2. Donors in the SAVIX database
N
Donor
SGs
Percent
Cum. percent
1
Master Card Foundation
38,972
26.86
26.86
2
Bill & Melinda Gates Foundation
35,377
24.38
51.25
3
Barclays Bank
14,291
9.85
61.1
4
Canadian International Development Agency
10,913
7.52
68.62
5
U.S. Agency for International Development
7,786
5.37
73.99
6
Department for International Development
5,908
4.07
78.06
7
Irish Aid
3,526
2.43
80.49
8
European Union (EU)
3,267
2.25
82.74
9
Danish International Development Agency
3,233
2.23
84.97
10
Barclays Corporation
2,768
1.91
86.88
11
Financial Sector Deepening Kenya
2,614
1.8
88.68
12
CIDA and the Norwegian Ministry of Foreign
2,300
1.59
90.26
13
DFATD Canada
2,268
1.56
91.83
14
Master Card Foundation, Canadian International cooperation
1,691
1.17
92.99
15
Norwegian Agency for Development Cooperation
1,654
1.14
94.13
16
Weberg Foundation
1,392
0.96
95.09
17
Australian Agency for International Dev
1,376
0.95
96.04
18
Inter-American Development Bank (IDB)
1,054
0.73
96.77
19
Swedish International Development Cooperation
725
0.5
97.27
20
Barclays UK
707
0.49
97.75
21
African Development Bank
675
0.47
98.22
22
Banca de las Oportunidades
355
0.24
98.46
23
Marshall Foundation
332
0.23
98.69
24
Plan USA
302
0.21
98.9
25
United Nations High Commissioner for Re
287
0.2
99.1
26
Plan International
279
0.19
99.29
27
Stromme Foundation
210
0.14
99.43
28
New Zealand Aid Programme (NZAID)
193
0.13
99.57
29
FSDT Kenya
158
0.11
99.68
30
VISA
86
0.06
99.74
31
Finnish International Development Agency
72
0.05
99.79
32
Austrian Development Agency (ADA)
64
0.04
99.83
33
L'Agence Francaise de Developpement
62
0.04
99.87
34
Canadian International Development Agency
49
0.03
99.91
35
ECHO
38
0.03
99.93
36
Taiwanese International Cooperation
33
0.02
99.96
37
United Nations Development Programme (UNDP)
26
0.02
99.97
38
Deutsche Gesellschaft fur International
23
0.02
99.99
39
EU + Plan
13
>0.01
100
40
Banque Mondial
3
>0.01
100
Global Indicators of Savings Groups
36
Table A.3. Savings groups in each country of the SAVIX database
N
Country
SGs
Percent
Cum. percent
1
Uganda
25,397
9.71
9.71
2
Mali
24,218
9.26
18.96
3
Tanzania
22,887
8.75
27.71
4
Burkina Faso
14,983
5.73
33.44
5
Ghana
13,680
5.23
38.67
6
Ethiopia
11,396
4.36
43.02
7
Mozambique
10,801
4.13
47.15
8
Senegal
10,786
4.12
51.27
9
Kenya
9,998
3.82
55.10
10
Afghanistan
8,214
3.14
58.23
11
Benin
8,137
3.11
61.34
12
Burundi
7,890
3.02
64.36
13
Niger
7,216
2.76
67.12
14
Colombia
7,145
2.73
69.85
15
Zimbabwe
6,781
2.59
72.44
16
Ivory Coast
6,612
2.53
74.97
17
Malawi
6,602
2.52
77.49
18
Zambia
5,919
2.26
79.75
19
Egypt
5,652
2.16
81.91
20
Togo
5,567
2.13
84.04
21
Sierra Leone
5,409
2.07
86.11
22
Rwanda
4,914
1.88
87.99
23
India
4,419
1.69
89.68
24
Cambodia
3,611
1.38
91.06
25
Philippines
3,398
1.30
92.36
26
Guinea
2,914
1.11
93.47
27
South Africa
2,516
0.96
94.43
28
Cameroon
2,281
0.87
95.30
29
Tajikistan
2,261
0.86
96.17
30
Madagascar
2,164
0.83
96.99
31
Nigeria
1,708
0.65
97.65
32
Pakistan
1,219
0.47
98.11
33
Guinea Bissau
1,020
0.39
98.50
34
Myanmar
635
0.24
98.75
35
Russia
621
0.24
98.98
36
Mongolia
402
0.15
99.14
37
Swaziland
384
0.15
99.28
38
Vietnam
302
0.12
99.40
39
Republic of the Congo
300
0.11
99.51
40
Namibia
261
0.10
99.61
41
Peru
232
0.09
99.70
42
Nicaragua
201
0.08
99.78
43
South Sudan
184
0.07
99.85
44
Sudan
147
0.06
99.91
45
Lesotho
95
0.04
99.94
46
Bangladesh
50
0.02
99.96
47
Angola
34
0.01
99.97
48
Sri Lanka
33
0.01
99.99
49
Guatemala
17
0.01
99.99
50
Somalia
10
>0.01
100
51
Costa Rica
4
>0.01
100
52
East Timor
3
>0.01
100
53
Bulgaria
1
>0.01
100
Global Indicators of Savings Groups
37
The Center for Research on Social Enterprises and Microfinance (CERSEM) is one of Europe’s
leading research groups on social enterprises and microfinance including Savings groups.
CERSEM specializes in microfinance research, where the unit of analysis is the provider of
financial services to unbanked populations. With unique access to global and other datasets,
combined with solid academic research, the goal of CERSEM is to deepen our understanding of
microfinance delivery mechanisms.
CERSEM is integrated into the research group on Emerging Markets at the School of Business
and Law, University of Agder, Norway. CERSEM also operates in close contact with industry
practitioners. To learn more about the current research projects at CERSEM, visit the webpage
at https://cersem.uia.no/.
The FAHU Foundation was established in 2007 to make a positive and continuous difference in
the world. FAHU is governed by a group of personally vested individuals, most of whom volunteer
their time. Since its inception, FAHU has supported initiatives in Latin America, Asia and Africa
that addressed the root causes of poverty and provided sustainable solutions for improving
livelihoods. The Foundation has promoted the inclusion of marginalized and vulnerable groups,
such as people with disabilities, people living with HIV/Aids, indigenous peoples, and illiterate
people, in its savings group projects.
FAHU’s current strategy is to partner with organizations on innovative savings group projects. For
more information, visit the webpage at http://www.fahufonden.dk/.
