







































Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 1 

 

Do Savings Groups Work? Six Years of 

Evidence from the DRC 
 

Courtney Purvis 

 

 

 
Using a unique model for community development called the Church Empowerment Zone, World 

Relief partners with the local church in the Democratic Republic of the Congo to implement a high 

impact, sustainable economic development program called Savings for Life. An innovative 

longitudinal data collection tool brings data from six years of program implementation and reveals 

continued positive financial growth for beneficiaries, even after program staff left the area. World 

Relief collected 673 datasets from 196 savings groups to learn how savings groups evolve and grow 

over time; members average over 350% increase in savings after six years while group sizes remain at 

maximum membership. The data also reveal the program’s positive outcomes on future savings 

groups: later-joining members saved nearly three times as much in their first year as early savings 

groups had in their first year. 

 

 
Overview 

World Relief, a Christian relief and development 

organization operating in nine countries internationally, 

uses a unique implementation model, called the 

Church Empowerment Zone, to create sustainable 

programing that grows in positive outcomes over time. 

One such program, dedicated to promoting village 

saving and enterprise, is called Savings for Life (SFL). 

Using the Savings for Life methodology, World Relief 

has trained over 400,000 members in eleven different 

countries to save together, and to give and repay loans 

from group savings. 

For purposes of monitoring and assessment, 

World Relief utilizes a technological longitudinal data 

collection system called Text-it alongside the SFL 

program in the Democratic Republic of the Congo. 

Analysis of Text-it data reveals insights on the economic 

outcomes of the SFL program in the lives of savings 

group members. Key findings are: 

  

1. There is high demand for SFL, evidenced by 

nearly 100% of groups maintaining maximum 

membership. 

2. Members’ savings increase over time; the most 

mature groups’ savings increased more than 350% 

over six years. 

3. Newer savings groups increase the share value, 

the unit of weekly savings, over 80% more than 

what older savings groups used in previous years 

during the same cycle.  

4. The Covid-19 pandemic has had a limited 

negative impact on SFL members’ ability to save. 

70% of savings groups that shared out after saving 

during the pandemic distributed more per 

member than before Covid-19.  

5. There is over a 2600% return on donors’ 

investment after eight savings group cycles. 

 

The paper begins by providing the context for how 

World Relief implements technical programs, using the 

Church Empowerment Zone as an operating model. 

Second, the paper discusses World Relief’s 

community-managed savings group program and the 

longitudinal data collection tool World Relief used 

from 2016 to 2021 to gather economic data on its 

savings groups in the DRC. Third, the paper presents 

and explains five key findings from savings program 

implementation. Finally, the paper notes some research 

gaps and suggests ideas for further research. 

 

Introduction and Program Implementation 

Context  
Influenced by the early work of the Lausanne 

Movement and the 2001 Micah Declaration on Integral 

Mission, World Relief believes in integral mission, the 

practical demonstration of the gospel that bears social 

consequences alongside the verbal proclamation of the 

gospel (Lausanne Movement 2010; Micah Network 

2001; Stott 1974). Moreover, World Relief believes that 

the local church must be at the center of development 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 2 

work in order to produce lasting change.
1

  Therefore, 

in its basic operating model, World Relief walks 

alongside local churches in geographic areas called 

Church Empowerment Zones (CEZ) to implement 

technical programs, such as child development, health 

and nutrition, peace and reconciliation, and economic 

development. In the CEZ, World Relief follows a 

multiyear process to empower networks of local 

churches, training key leaders and volunteers in the 

implementation of technical programs. When the time 

is right, World Relief hands responsibility for 

continuation of the programs over to the local church.  

The Church Empowerment Zone process has four 

implementation phases. In the first phase, mobilization, 

World Relief works to build a network of local 

Protestant and Catholic churches. Living in the same 

geographic area is often not enough to bring churches 

or church leaders together. Instead, churches may feel 

like they cannot work together sometimes like they are 

in competition with each other. Through a long process 

of sensitization, identification, and introduction, local 

churches come to know each other in new ways and 

learn to behave more like teammates than competitors. 

World Relief also helps group local churches together 

in meaningful geographic areas, calling each group a 

Church Network. Each Church Network elects leaders, 

called the Church Network Committee (CNC) to 

represent 15-40 churches in their geographic area. 

During this first introductory phase, the goal is to bring 

churches together so they can coordinate their work in 

a variety of key community ministries. 

In the second phase, World Relief conducts a 

needs assessment and baseline survey to understand the 

starting point of need in the community. After this, 

World Relief aims to transform mindsets, especially 

around the ideas of empowerment and poverty 

alleviation.
2

 In this phase, the goal is to help local 

church leaders take a hard look at the capacities and 

assets they already have within the community along 

with the needs they would like to address. Continuing 

to focus on the empowerment of churches, World 

Relief challenges the network of churches to address 

one tangible need in their community using only their 

own resources.
3

   World Relief has seen churches from 

various denominations work together to repair roads, 

rebuild widows’ homes, and even construct a new 

church building for one of the denominations in the 

 
1

 World Relief is not alone in this belief; The Accord Network, and its member organizations, affirm this in the 

“Principles of Excellence in Integral Mission,” which serves as guiding ideals for Christian organizations operating in 

international development. https://accordnetwork.org/about/principles/  
2

 Gil Odendaal’s book, Standing with the Vulnerable (2016) includes lessons like the ones World Relief uses in the 

field, contextualized for a North American audience.  
3

 World Relief adapted this concept of seed projects from Disciple Nations Alliance and the Global CHE Network 

(2012) and Bob Moffitt’s “Seeds and Seed Projects” (2007). 

community. As churches begin to see each other as 

teammates instead of competitors, they work together 

across denominational lines on common projects that 

benefit the whole community, not just the members of 

one church.  

In the third phase, World Relief focuses on 

capacity development of church leaders and volunteers 

as they introduce technical programs to address 

community needs discovered in phase two. World 

Relief field staff work alongside church leaders to select 

community-based volunteers to train in programs like 

child development, outreach groups for health and 

hygiene, agriculture, and savings groups. In this phase, 

World Relief staff take an active role in program 

implementation, always keeping in mind the final 

objective of handing the program over to the local 

church. World Relief’s goal in this third phase is to 

multiply outreach in the community by training 

committed church-based volunteers to serve the 

broader community.  

Once technical programs develop and mature, 

World Relief begins phase four: reflection, evaluation, 

and handover to the local church. By this point, teams 

of local volunteers have gained enough expertise in 

program implementation to carry on program growth 

with the support of church leaders. Many programs 

create sustainability mechanisms, such as local program 

committees, which act as the program authority and 

provide leadership and ongoing training for other 

volunteers in the community. As World Relief exits the 

community, local churches assume the responsibility of 

continuing valued technical programs as church 

ministries. 

 

The Savings for Life Program in the Church 

Empowerment Zone 
World Relief applied this Church Empowerment 

Zone model in 2015 in the Rutshuru region in the 

Eastern part of the Democratic Republic of the Congo. 

Prior to introducing technical programs, World Relief 

networked churches together, conducted baseline 

surveys and needs assessments, provided envisioning 

seminars with church leaders and challenged local 

churches to take on a seed project in the Kako region 

of Rutshuru. In the Kako CEZ, there are approximately 

36,000 households with a total population of nearly 

215,000; 89% of the total population lives under the 

about:blank


Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 3 

poverty line. World Relief partnered with 107 churches 

and organized them into seven CNCs that provide 

leadership for program activities in their geographic 

areas.  

In 2015, after six months of phase one and two 

work, the Kako CEZ became ready to implement 

technical programs. Bearing in mind the needs 

identified in phase two, church leaders decided one of 

the first issues they wanted to address was material 

poverty. World Relief staff linked their request to the 

economic development program, Savings for Life 

(SFL).  

World Relief developed the Savings for Life 

program out of its deep history of starting and operating 

microfinance institutions (MFIs). Around 2008, World 

Relief intentionally refocused on partnering with the 

local church for community-based programming and 

identified MFIs, largely operating in urban centers, as 

outside their target focus. Instead, World Relief 

redirected their economic development efforts towards 

community-based savings groups. At that time, the 

Chalmers Center4  was the thought leader for faith-

based savings programming, while secular organizations 

were starting to follow the Village Savings and Loan 

Association (VSLA) methodology, developed by 

CARE international
5

 and popularized by Hugh Allen of 

VSL Associates.
6

 World Relief piloted both the 

Chalmers model and VSLA in separate countries from 

2008-2010. Experience in the field found more 

cohesive, financially successful savings groups who used 

VSLA, leading World Relief to employ only VSLA for 

future savings groups operations. In 2013, World Relief 

wrote narrative sections to complement the operations 

of the VSLA methodology and named this curriculum 

Savings for Life. The SFL curriculum brings a biblical 

understanding of stewardship and poverty alleviation 

that draws upon Bryant Myers’ ideas that poverty is 

caused by brokenness in four key relationships (Myers 

2011).  

Once church leaders decided to address material 

poverty through SFL, World Relief followed a three-

year process to hand over the SFL ministry to the local 

church. In the first year, World Relief field staff formed 

 
4

 When Helping Hurts (Corbett and Fikkert 2009) and From Dependency to Dignity (Fikkert and Mask 2015) are two 

of the Chalmers Centers’ most helpful resources on savings programs. The Chalmers Center has their savings group 

training materials available online at https://chalmers.org/training/restore-savings/  
5 https://www.care.org/our-work/education-and-work/microsavings/vsla-101/  
6

 The VSL Associates’ website, https://www.vsla.net/ has a full description of the VSLA methodology, including 

downloadable training guides. Early work by Allen and Panetta (2010) summarizes VSLA-style savings group 

programs. The Savings-Led Working Group of the SEEP Network serves as a convener for implementing 

organizations in the savings group sector, worldwide. The website www.mangotree.org houses a resource library of this 

working group. 
7

 World Relief’s volunteer structure largely follows the VSLA methodology’s Village Agent approach. See 

https://www.vsla.net/ for information and training guides for utilizing Village Agents.  

and trained 24 savings groups in the SFL methodology. 

Through SFL, members contribute weekly to a 

common savings fund through share purchases, setting 

an initial share value at the beginning of the cycle. 

Members must buy a minimum of one share and a 

maximum of five shares at every weekly meeting. Initial 

savings groups in the Kako CEZ set a share value of 500 

Congolese Francs, giving members the opportunity to 

save 500-2,500 Congolese Francs ($0.25-$1.25), in 

increments of 500, at each weekly meeting. This 

common savings fund provides loan money for group 

members, repayable to the group with an interest rate 

set by the group. In Kako CEZ, groups have chosen 

between 5-10% interest per month with a maximum 

loan term of three months. Groups also manage a social 

fund that acts as micro-insurance, providing small grants 

in times of need. Groups set a fixed amount for weekly 

contributions, ranging between 100-500 Congolese 

Francs ($0.05-$0.25) for the social fund and agree upon 

the amount of a grant for a member facing a financial 

crisis, such as an unexpected hospitalization of a child. 

The group has a metal box with three locks to hold all 

money for savings, loans, and the social fund. Three 

separate members hold keys for the box which may 

only be opened at SFL group meetings. After a period 

of 9-12 months, groups hold an annual distribution of 

funds, called a share out, at which each member takes 

home the amount s/he saved plus the portion of interest 

earned from loan activity (typically around 25-30% 

return on one’s savings). After each share out, groups 

start a new cycle of savings and lending. The first twenty-

four groups saved together and found great success 

following the SFL methodology; the average member 

saved $36.15 but took home $47.71 at the share out, a 

32% return on their investment. 

After a successful first cohort of savers, World 

Relief then selected volunteers
7

 from among those first 

groups and trained them on how to form, train, and 

support new savings groups in the community. World 

Relief’s field staff then moved into a position of 

supporting teams of volunteers as the volunteers 

formed and trained new groups. In total, World Relief 

Congo now has thirty-three community-based 

about:blank
about:blank
about:blank
about:blank
about:blank
about:blank


Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 4 

volunteers working with the SFL program in the Kako 

CEZ. These volunteers receive annual refresher 

training from World Relief, a bicycle to help them get 

to savings groups, and, because the weather in Kako is 

often overcast and rainy, a raincoat and rain boots. 

Beyond this, there are no other financial incentives 

from the organization, churches, or savings groups. 

World Relief designed the volunteer structure with the 

eventual handover to the local church in mind, meaning 

in part that they did not want to pass on any financial 

cost to the church. Yet, even without financial 

incentives, the SFL program thrives, with volunteers 

typically meeting or exceeding target numbers of new 

savings groups to form and train.  

After two years of implementation, World Relief 

started setting up the sustainability mechanism: the SFL 

Committee. This committee is a group of “super 

volunteers” who take on the roles and responsibilities 

of the field staff and sit under the authority and 

leadership of the CNC. This is the first step of World 

Relief phasing out of the community. World Relief 

organized four SFL Committees throughout the Kako 

area. These report to the CNC and support the 

remaining SFL volunteers as they continue to work with 

new and existing savings groups. In SFL 

implementation year three, the World Relief staff work 

to hand over their responsibilities, ensuring SFL 

Committee members understand and can follow their 

roles successfully. Though World Relief staff phased 

out of the Kako CEZ in 2018, the SFL Committee 

continues supporting both groups and volunteers, and 

CNC members support and encourage SFL 

Committee members in order to keep SFL work going. 

At the time of the handover from World Relief staff to 

the SFL Committee in 2018, World Relief staff and 

volunteers had established 125 SFL groups. From 2018 

until today, the SFL Committee continues to support 

these 125 existing groups, but have also added 72 new 

groups with 1,801 new members. The total number of 

SFL groups in Kako CEZ is now 197 with 4,906 

members.
8

 Overall, group members save an average of 

$62 annually and earn 27.4% return on their savings 

(Savix 2021).  

Many international organizations that implement 

VSLA-style savings groups, including World Relief, use 

the same management information system, the Savix. 

This shared platform allows organizations to compare 

key performance indicators. Table 1 shows indicators 

from the Savix comparing World Relief’s SFL program 

in Kako CEZ to other organizations’ savings program 

in the Congo, East Africa, all of Africa, and worldwide. 

World Relief’s lower average outstanding loan with 

higher return on savings indicates numerous turnovers 

of small loans to generate increased profit to members.
9

  

 

 
Savings Groups Programs Operating in: 

Key Performance Indicator 

Kako CEZ 

(World 

Relief only) 

Congo (all 

implementers) 

East  

Africa (all 

implementers) 

Africa  (all 

implementers) 

World  (all 

implementers) 

Attendance rate 96.6% 87.0% 86.0% 87.0% 87.0% 

Average savings per 

member 
$62  $28  $25  $25  $28  

Average outstanding loan 

size 
$27  $38  $37  $37  $41  

Return on savings 27.4% 15.7% 19.3% 17.3% 16.6% 

Table 1 

The context of implementing SFL in the CEZ is 

one positive factor enabling the program’s success. 

Volunteers are committed to training and supporting 

groups with the encouragement of the CNC. 

Community members not yet in a SFL group see the 

 
8 The Kako CEZ savings groups are one subset of the entire groups in the DRC. Throughout the DRC, World Relief 

has trained and formed 400 savings groups with nearly 10,000 members. 
9 Researchers cannot know individual groups’ interest rates. VSL Associates promotes 10% interest rate per month, 

which many organizations also share with groups due to its ease of calculation. Ultimately, groups choose their own 

interest rate. A majority of World Relief SFL groups in Kako CEZ use 10% interest rate.  

profit earned by current groups, which motivates and 

encourages them to form their own group. Although 

World Relief staff officially left the area in 2018, SFL 

Committees remain in the community to sustain the 

work carried on by volunteers. The framework of the 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 5 

CEZ, especially the initial work of church networking, 

lays the groundwork for the SFL program to have 

ongoing positive outcomes in community members’ 

lives. 

 

Savings for Life Longitudinal Research Using 

the Text-it System 
In order to monitor and asses the SFL program, in 

2016 World Relief partnered with an innovative, 

longitudinal data collection service called Text-it to 

supply group-level data about the savings groups in 

Kako CEZ. The Text-it system is a SMS-based bounce 

back service that allows group members to send 

quantitative data via text message for World Relief to 

analyze. Each group has a unique identification 

number, allowing World Relief to follow each group’s 

incremental financial progress. 

Since starting to use Text-it in 2016, World Relief 

has had five cohorts of savers, one starting each year, as 

depicted in Figure 1.  

 

 
   Figure 1 

Each cohort is a cluster of savings groups that started 

their saving cycle roughly around the same time and 

therefore shared out roughly around the same time, 

approximately nine months later. Cohorts progress to 

the next cycle as a new cohort starts cycle 1. Cohort 1, 

the first 24 groups to use Text-it, completed eight cycles 

by the time of this data collection. World Relief 

collected 673 datasets from 196 groups across all 

cycles.
10

 Figure 2 shows the number of datasets World 

Relief collected per cycle. 

 

 
   Figure 2 

 
10 This research would not be possible without the diligent work of Joseph Kalegamire, Emily Mugisha, and Sam 

Ferguson, as well as many field staff and volunteers in the Kako CEZ. 

On the day of the share out and in the presence of 

the full group, a member of the savings group, usually 

the record keeper, sends a missed call to a local Text-it 

number that initiates a series of seven questions in text 

message form, written in the local language. Each 

question has a numeric answer with a preset acceptable 

range. If the response to the prompt is within the 

acceptable range, the system sends the next question. If 

an answer comes back outside the acceptable range, the 

system sends a response indicating the answer is outside 

the normal range and asks the member to verify and 

send a different response. The questions Text-it sends 

are: 

 

1. What is your group number? (this is preassigned 

to each group and does not change cycle to 

cycle) 

2. What cycle number has your group just 

completed? 

3. What was the share value for your group this 

cycle? 

4. What was the total number of shares collected 

this cycle? 

5. What was the total amount shared out this cycle? 

6. How many members did your group have this 

cycle? 

7. Does your group plan to start a new cycle? 

 

All of the data for this study came from within the 

Kako CEZ. Because World Relief’s standard is to 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 6 

implement technical programs within the context of the 

CEZ, this study does not have similar data from outside 

the CEZ to compare. Similarly, there is a lack of other 

longitudinal research of this nature to compare to 

World Relief’s data. Therefore, this research does not 

attempt to show causality to the CEZ model or any 

other single factor for reported outcomes; instead, the 

research simply observes the context in which these 

outcomes take place. 

 
Key terms: 

Shares SFL members save in the form of 

shares, saving one to five shares at each 

weekly meeting. 

 

Share 

Value 

The share value is the cost of one share, 

for example, 500 Congolese Francs 

(about 25 cents in US currency). 

Cycle Each saving group saves, gives loans and 

repays loans within the group for 9-12 

months, which constitutes one cycle of 

saving and lending. 

Share 

Out 

At the end of each cycle, groups have a 

distribution of funds, called the share 

out. 

New 

Cycle 

After the share out, groups have the 

opportunity to start a new cycle with a 

starting balance of zero. 

Table 2 

Key Findings 

Key Finding 1: There is high demand for SFL  

Savings for Life appears to meet a strong need for 

members. Across all 673 datasets, 100% of the groups 

started a new cycle. Additionally, 98.2% of groups have 

the maximum group size of 25 members (Figure 3). 

While World Relief does not expect that all group 

members are exactly the same from the time the group 

formed (as members may have moved away or dropped 

out of the program), this means the group replaced any 

departing member with a new member to keep the 

group size at 25. Finally, data from the management 

information system, the Savix, reveal the average weekly 

attendance rate is 97%. This rate is higher than averages 

from other implementers in the Congo, Africa, or 

worldwide (Table 1) and also the savings group sector 

standard of 85%, set by Hugh Allen, the founder of 

VSL Associates (Allen 2007, Mersland et al. 2019). The 

high attendance rate of SFL members suggests that they  

 
11 This finding reports savings in local currency only. The DRC experienced greatly varying exchange rates from 2016 

to 2021, as low as 950 CF to 1 USD to as high as 1969 CF to 1 USD, respectively. Even so, rural SFL members likely 

experienced little impact from these exchange rate changes; for example, the standard daily wage for a farm laborer or 

are 

eager and willing to attend group meetings regularly. 

From 2016 until 2018, while World Relief set annual 

targets for new groups, field staff and volunteers 

regularly exceeded the targeted numbers, reflecting the 

demand from the community. Many new members 

testified that their reasons for joining a savings group 

included the simplicity of the saving and loan recording 

processes, the ease of applying for and obtaining loans, 

and also the testimony of current savings group 

members who earned a significant profit on their 

savings.  
The geographic context of the Kako CEZ may 

influence the high demand for SFL; there is virtually no 

other financial institution available for savings or loan 

activities for community members. The closest formal 

financial institution is approximately 70 kilometers 

away. Local moneylenders charge up to 50% monthly 

interest rate and demand significant collateral, one’s 

home for example. In addition, no other development 

organization implements a savings group program in 

the Kako region. The shortage of other available 

financial services, combined with the continued success 

of SFL groups, as indicated by regular 25-30% return 

on savings, contributes towards the high demand for 

SFL in Kako CEZ. 

 

Key Finding 2: Members’ savings increase over time  

The data from Cohort 1 (groups 1-24) show 

significant savings growth over sequential cycles. The 

graph below shows the average amount a member in 

Cohort 1 saved and the amount s/he took home at the 

share out (savings plus profit from loan interest) for 

each of the eight cycles. An average member in Cohort 

1 increased their savings 351% from cycle one to cycle 

eight, as depicted in Figure 4. 
11

    

Figure 3 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 7 

This increase in savings occurred in two ways. First, 

members increased the number of shares they 

purchased as they progressed in cycles, as shown in 

Figure 5. At the beginning of each cycle, members set 

the share value, then purchase a minimum of one to a 

maximum of five shares at each weekly meeting. The 

average number of weekly shares saved increased from 

1.9 in cycle one to 3.4 in cycle 3 and then remained 

relatively steady over the next five cycles, reaching a 

maximum of 3.7 in cycle 8 (Figure 5). Second, as groups 

progressed from one cycle to the next, they increased 

the group share value from the previous cycle, as seen 

in Figure 6. After the share out, members who regularly 

saved the weekly maximum of five shares, challenged 

the group to increase the share value so that, in the next 

cycle, they would be able to save more. Many times, 

members attempted to save the same number of shares 

each week as the previous cycle, even though the share 

value increased, leading to much higher savings. All 

groups in Cohort 1 started their first cycle with a share 

value of 500 Congolese Francs (CF) and by the eighth 

cycle, the share value had increased to 1000 CF for 71% 

of groups and 2000 CF for 29% of groups, bringing the 

average share value in cycle eight for Cohort 1 to 1,285 

CF.   

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
           Figure 4 

 
 
 
 
 
 
 
 
 
 
 
 
 
             

       Figure 5 

 

 
the price for locally grown foods remained unchanged during these years. Therefore, observing the change in saving 

amounts over the years in local currency, not USD, is a better representation of the SFL member experience. 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 8 

 
 
 
 
 
 
 
 
 
 
 
 
              Figure 6 

 
Key Finding 3: Share values increase for successive 

cohorts 

A different way to analyze the change in share value 

is to look at how it changes within the same cycle over 

successive years and cohorts. When SFL started in the 

DRC, 100% of groups used 500 CF as their share value. 

Savings group implementers may expect to observe 

increasing share values as groups progress in cycles 

(Key Finding 2) but one notable finding from this study 

is that new cohorts started the same cycle but with a 

higher share value. For example, when Cohort 4 started 

their first cycle, 37% of groups opted to increase the 

share value to 1000 CF or 2000 CF from the previous 

standard of 500 CF, Figure 7. This increase is even 

greater comparing cycle two data between Cohort 1 and 

4; in 2017, when Cohort 1 held the second cycle share 

out, again 100% of groups used a share value of 500 CF. 

By 2020, when Cohort 4 held their second cycle share 

out, the average share value increased to over 900 CF, 

Figure 7. Only 35% of Cohort 4 groups maintained the 

500 CF share value; 56% of groups increased their 

share value to 1000 CF and 9% of groups increased to 

2000 CF. 

 

            
               Figure 7 

World Relief has a few hypotheses on why newer 

cohorts start cycles with share values higher than the 

preceding cohorts. First, when World Relief introduced 

SFL in the community, group members were uncertain 

about the program and if or how it would benefit them. 

After the initial cohort held a share out with 25-30% 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 9 

return on their savings, testimony spread throughout 

the community. This may have led future cohorts to 

trust the SFL system and lead them to save more than 

the initial savers. When members in Cohort 1 start new, 

advanced cycles, with a share value of 1000 CF or more, 

and new members join a freshly formed cycle-one SFL 

group, these new members may look to the seasoned 

members for modeling and setting expectations. New 

groups may choose to save with a 1000 CF share value 

simply because they observe it modeled by their peers. 

Another possible explanation is that, when the initial 

cohort started saving, field staff reported that members 

did not believe in their own ability to save. Once 

members shared out with profit and started the next 

cycle with a higher share value, however, others were 

also encouraged and started saving at a higher share 

value. Yet another explanation may be that a general 

increase in small business activities, which World Relief 

definitely noticed, contributed to the members’ ability 

to save more. A final possibility is that some members 

in early cohorts may have joined later cohorts to be part 

of a second savings group, even though World Relief 

does not encourage this practice. When this happens, 

the more mature members may encourage the entire 

group to start with a higher share value because they 

know it is possible.  

While World Relief does not know the exact 

reasons behind the higher share value, the increasing 

share value across successive cohorts does seem to 

demonstrate members’ increased appreciation for and 

trust in the SFL program. 

 

Key Finding 4: Covid-19 had limited negative impact on 

SFL members’ ability to save 

World Relief assigned each savings group a unique 

number in the Text-it system so it could compare 

savings and share out amounts from cycle to cycle. At 

first glance, data appear to show a strong negative 

outcome for members saving during the Covid-19 

pandemic: only 70% of groups increased the amount 

they shared out in 2020 (when a majority of savings 

occurred during Covid-19) compared to the share out 

the year before, 2019, and 30% of groups decreased the 

amount they shared out during this period. In contrast, 

during the year before Covid-19 (2019), 83% of groups 

increased their share out over the share out in 2018 

while only 17% of groups decreased their share out 

during this period (Figure 8). 

 

 
Figure 8 

Still, although fewer groups increased their share 

out amounts during the pandemic, the average amount 

each member earned at share out actually increased 

compared to the share out pre-pandemic. The average 

change in share out, across groups in all cycles, 

comparing the 2020 share out to the 2019 share out was 

22,099 Congolese Francs (CF). This is slightly higher, 

1,104 CF more, than the share out differential between 

2019 and 2018 (Figure 9). Although fewer groups had 

an overall increase, those that did increased at a greater 

rate, making the average change higher for savings 

during the Covid-19 pandemic compared to savings 

from the previous year. 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 10 

 
Figure 9 

Figure 10 depicts the percent of groups that 

increased or decreased their share out in three different 

cycles: 2019 compared to 2018 (savings prior to Covid-

19, orange bars), 2020 compared to 2019 (savings 

during the first half of Covid-19, blue bars) and 2021 

compared to 2020 (savings during the second half of 

Covid-19, grey bars). While there were groups that 

increased in 2019 (orange bars), the increase was 

relatively low, with a majority of groups only increasing 

0-20,000 CF per member. This contrasts with a greater 

percentage of groups that increased their share out in 

2020 (blue bars) or 2021 (grey bars) but did so at a 

higher rate, with a majority of groups in 2020 increasing 

20,000-100,000 CF per member and a majority of 

groups in 2021 increasing 20,000-50,000 CF per 

member.  

 
 

 
Figure 10 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 11 

Government-imposed Covid-19 restrictions may 

explain these varying increases and decreases of 

members’ savings. When the government imposed a 

lockdown, community members could not travel from 

the countryside to the city. This affected farmers who 

now could not travel to sell their produce in densely 

populated city markets. Instead, many farmers 

attempted to sell their produce in countryside markets 

where, due to the competition, prices were low. On the 

other hand, some community members may have 

financially benefitted from the travel restrictions, 

because they did not spend money on travel expenses 

and were available to work on others’ farms as day 

laborers. Overall, these changes in share out amounts 

indicate that although the Covid-19 pandemic may have 

impacted some group members, causing them to save a 

bit less, overall, the general progression has continued; 

as members progress through cycles of savings, the 

amount saved in each cycle increases. 

 

Key Finding 5: There is a strong positive return on 

donors’ investment 

The average cost for World Relief to implement 

SFL in the DRC is $21.56 per member12 (as indicated 

by the star in Figure 12). The amount of money the 

average member saves in the SFL program far 

outweighs the cost running the program. This research 

does not have counterfactual data to compare savings of 

non-SFL members in Kako to the savings of SFL 

members, so this paper cannot assign causality to the 

savings rates.
13

 But World Relief does not know of any 

other external factors that would drive up savings rates 

of SFL members such as savings groups or other 

economic development programs offered by 

development organizations or government programs. 

Therefore, this research compares the cost per 

member to the amount the member saved in the cycle 

for the return on the donor’s investment, using the 

following return on investment (ROI) calculation of: 

Figure 11 

 

In cycle one, the average member in Cohort 1 

earned $47.71 at the time of the share out. Using the 

above calculation, there is already a 121% ROI on the 

donors’ investment. Because the investment for the 

member is a one-time cost (World Relief directs future 

investments to future cohorts), any savings beyond the 

cost of $21.56 is a positive return, as indicated by the 

circled sections on Figure 12.  

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 Figure 12 

 
12

 Cost per member is one standard ratio, calculated by the management information system, the Savix. The Savix 

calculates this as the prorated cost of the program divided by total number of savings group members. Implementing 

organizations have not agreed upon what to include in the program cost, making this a ratio difficult to compare across 

programs. World Relief includes all field costs (staff salaries, training, monitoring and evaluation costs, and a portion 

of field office expenses) but not US home office overhead. 
13 Other randomized control trials from the savings group community confirm that savings groups contribute towards 

increased household savings (Gash and Odell, 2013). 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 12 

As the cycles continue and share out amounts increase, 

the return on the donors’ investment rises to 2661% by 

cycle eight (calculated by adding up all the share out 

amounts over cycles 1-8, subtracting the cost of 

investment, $21.56, then dividing by the static cost of 

investment, $21.56), as shown in Figure 13.
14

  

 

 

 

  
 Figure 13 

  

The implications of this finding may encourage donors 

to know how the outcomes of their giving exponentially 

increases over years, as members continue to save more 

and more in new cycles. 

 

Research Gaps and Call for Further Research 
This research has the obvious limitation in that 

there is no control group, or a group similar in 

demographics where World Relief does not implement 

the SFL program, to compare data with. Because of this 

limitation, this research is not able to show definitively 

that SFL is the factor driving positive outcomes. It is 

possible, for example, that other factors are playing a 

role in the saving experiences of SFL members. Given 

the village context, however, and World Relief’s 

knowledge of that context, it is not unreasonable to 

conclude that the SFL program has been a major factor 

in the change. Still, in order to prove causality, World 

Relief, or perhaps an independent researcher, would 

need to conduct further research using a randomized 

control trial (RCT), designating a control group to 

compare to the treatment group of SFL members.  

Other savings group implementing organizations 

have conducted RCTs to link savings groups to the 

causality of economic impacts in treatment areas (Gash 

and Odell 2013; Gash 2017). Most research, however, 

has a limited timeframe, such as one to three years, 

because RCTs are difficult to conduct over a long 

 
14 Admittedly, this figure ignores any net present value calculations. Its purpose is only to show that return on donors’ 

initial investment is indeed significant. 

period. These studies reveal increases in savings and the 

use of credit in treatment areas. Researchers have also 

examined household expenditures to determine if 

increased savings came about because members simply 

reduced household expenses and therefore had more 

money to save. Yet researchers found no reduction in 

household expenses, suggesting that the increase in 

savings did not negatively affect household 

consumption. Some RCTs have also found an increase 

in total household expenditures for the treatment 

group, suggesting that savings groups may allow for 

increased household purchases (Ksoll et al. 2013; 

Annan et al. 2013). Finally, for RCTs that measure asset 

ownership, nearly all studies find reasonable increases 

for savings group members, while no study finds 

negative effects on asset ownership (Gash 2017).  

Evidence from these RCTs suggests a story similar 

to what World Relief believes has occurred in the DRC 

with the Savings for Life program; participation in the 

savings group program leads to higher savings among 

participants. The full extent of these impacts, however, 

remains uncertain until a fuller RCT is conducted.  

Another limitation of this research is the lack of 

analysis into members’ utilization of SFL funds for 

business, health, education, or other outcomes. 

Although researchers may accomplish this with robust 

RCT work, the complicated nature of money 

management likely calls for a mixed methods approach 



Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 13 

to research (Rutherford 2003). In the qualitative 

section, researchers may find financial diaries as an 

effective technique to chart and monitor community 

members’ income and expenditure and how 

participation in a savings group may change these 

amounts (Collins, Morduch et al. 2009). A further 

analysis into the use of SFL funds would provide a 

deeper understanding into how the SFL program 

impacts important everyday aspects of community 

members’ lives.  

 

Conclusion 
Following the financial data of savings groups in the 

DRC from 2016 until today, World Relief has 

witnessed significant growth in members’ savings. 

Savings groups continue to increase their share values 

as they move to subsequent cycles, which leads to 

higher weekly savings. Groups continue to have high 

profits from internal loans and distribute those profits 

at the annual share out. New members continue to join 

SFL groups, saving at higher rates than the earlier 

groups. 

World Relief believes lasting community change 

only happens when the community, and especially 

community leaders, are not only champions of change 

but are also leading the change. This is why World 

Relief invests so much time in the Church 

Empowerment Zone doing church networking, 

envisioning seminars, and mindset transformation 

trainings before introducing technical programs in the 

community. Like the SFL, the network of cooperating 

churches continues after World Relief hands the 

program over to them. With this framework in place, 

churches are able to select community and church-

based volunteers who are the main trainers for 

programs like Savings for Life. Within SFL, the 

volunteer SFL Committee continues to support other 

volunteers and train new groups in the SFL 

methodology. 

Although World Relief staff phased out of the 

Kako CEZ SFL program in 2018, testimonies reveal 

ongoing positive economic outcomes, such as 

volunteers continuing to form new groups with the 

maximum number of members, groups increasing their 

share value as they progress into new cycles, and 

continued increased savings shared out at the annual 

distribution. With its holistic understanding of human 

development, World Relief knows that progress 

involves much more than increased saving, but it also 

knows that increased saving is an important step in the 

direction of better lives in the economic sphere. 

Importantly, as this article tries to show, World Relief’s 

SFL program is helping communities take that step.  

 

 

 

References 

 

Accord Network. “Integral Mission Principles.” 

https://accordnetwork.org/about/principles/   

Allen, Hugh, and Panetta, David. 2010. Savings 
Groups, What Are They? The SEEP Network: 

Washington, D.C.  

Allen, Hugh and The SEEP Network Savings-led 

Financial Services Working Group: Ratios sub-

group. 2007. Financial Ratio Analysis of 

Community-managed Micro-finance Institutions. 
The SEEP Network.  

Annan, Jeannie, Tom Bundervoet, Juliette Seban, and 

Jaime Costigan. 2013. Urwaruka Rushasha (New 

Generation): A Randomized Impact Evaluation of 
Village Savings and Loans Associations and 

Family-Based Interventions in Burundi. Final 
Evaluation. International Rescue Committee. 

Collins, Daryl, Jonathan Morduch, Stuart Rutherford, 

and Orlanda Ruthven. 2009. Portfolios of the 
Poor: How the World’s Poor live on $2 a Day. 

Princeton, NJ: Princeton University Press.  

Corbett, Steve and Brian Fikkert. 2009. When 

Helping Hurts: How to Alleviate Poverty Without 
Hurting the Poor…and Yourself. Chicago, IL: 

Moody Publishers. 

Fikkert, Brian and Russell Mask. 2015. From 

Dependence to Dignity: How to Alleviate Poverty 
Through Church-Centered Microfinance. Grand 

Rapids, MI: Zondervan. 

Gash, Megan. 2017. Understanding the Impact of 

Savings Groups. The SEEP Network: 

Washington, D.C. 

Gash, Megan and Kathleen Odell. 2013. The 

Evidence-Based Story of Savings Groups: A 

Synthesis of Seven Randomized Control Trials. 

The SEEP Network: Washington, D.C. 

Ksoll, Christopher, Helene Bie Lilleør, Jonas Helth 

Lønborg, and Ole Dahl Rasmussen. 2013. Impact 
of Village Savings and Loans Associations: 

Evidence from a Cluster Randomized Trial. 
Study Paper No. 56. Rockwool Foundation 

Research Unit. 

Lausanne Congress. 2010. “The Cape Town 

Commitment: A Confession of Faith and a Call to 

Action.” Lausanne Movement. 

https://lausanne.org/content/ctcommitment#capet

own  

Medical Ambassadors International and Global CHE 

Network. 2012. “Planning a Seed Project.”   

https://chenetwork.org/_dvd/03%20CHE%20TO

Ts/TOT%20Lessons%20Archive/TOT%201%20

Single%20Column/1-

TOT%201%20in%205%20days/1-

TOT%20I%20Lessons/TA200%20Seed%20Proje

cts_SC.doc  

about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank#capetown
about:blank#capetown


Christian Relief, Development, and Advocacy 3(1), Summer 2021  

Purvis, Do Savings Groups Work? Six Years of Evidence from the DRC 14 

Mersland, Roy, Bert D’Espallier, Rolando Gonzales, 

and Linda Nakato. 2019. What are Savings 

Groups? A description of savings groups based on 
information in the SAVIX database. Hørsholm, 

Denmark: FAHU Foundation. 

Micah Network. 2001. “Micah Declaration on Integral 

Mission.”  

Moffitt, Bob. 2007. “Seeds and Seed 

Projects.”  https://disciplenations.org/wp-

content/uploads/2020/04/Seeds-and-Seed-

Projects_Moffitt.pdf.   

Myers, Bryant. 2011. Walking with the Poor: The 

Principles and Practices of Transformational 
Development. Maryknoll, NY: Orbis Books. 

Odendaal, Gil. 2016. Standing with the Vulnerable: A 
Curriculum for Transforming Lives and 

Communities. World Relief. Downers Grove: 

InterVarsity Press.  

Rutherford, Stuart. 2003. The Poor and their Money. 

Oxford, NY: Oxford University Press.  

Savings-Led Working Group, The SEEP Network. 

https://seepnetwork.org/Savings-Led-WG  

Savix: The Savings Group Information Exchange. 

https://thesavix.org/ 

Stott, John. 1974. “The Lausanne Covenant.” 

Lausanne Movement. 

https://lausanne.org/content/covenant/lausanne-

covenant#cov 

Text-it: The Leading Bot Platform. https://textit.com/  

The Mango Tree: The Global Resource on Savings 

Groups. The SEEP Network. 

https://mangotree.org/   

VSL Associates. https://www.vsla.net/   
 

 

Courtney Purvis has served as the Senior Program 

Advisor for World Relief’s Savings for Life program 

since 2011. Her recent work focuses on digitization of 

savings groups and promoting women’s empowerment 

and gender equality through savings groups. 

 

Author email: cpurvis@wr.org  
 
 

about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank
about:blank#cov
about:blank#cov
about:blank
about:blank
about:blank
about:blank

