MIDTERM PERFORMANCE EVALUATION OF 
THE FEED THE FUTURE RWANDA NGURIZA 
NSHORE ACTIVITY
November 12, 2021
November 2021
This document was produced at the request of the United States Agency for International 
Development. It was prepared independently by ME&A, Inc.
MIDTERM PERFORMANCE
EVALUATION OF THE FEED THE 
FUTURE RWANDA NGURIZA 
NSHORE ACTIVITY 
November 12, 2021
Prepared for the United States Agency for International Development under USAID Feed the Future 
Global Program Evaluation for Effectiveness and Learning (PEEL)
Contract Number: AID-OAA-TO-16-00008 
Submitted to:
USAID/Rwanda
Fina Kayisanabo 
fkayisanabo@usaid.gov
and
USAID/Bureau for Resilience and Food Security 
Lindsey Anna 
lanna@usaid.gov
Submitted by:
ME&A 
1020 19th Street NW, Suite 875
Washington, DC 20036
Tel: (240) 762-6296 
Photo Credit: Workers at Masaka Creamery, Kigali Rwanda.
DISCLAIMER
The authors’ views expressed in this publication do not necessarily reflect the views of the United States 
Agency for International Development or the United States Government.
ABSTRACT 
This midterm performance evaluation report of the Feed the Future Rwanda Nguriza Nshore (NN) 
activity was commissioned by the United States Agency for International Development (USAID). The 
evaluation used a mixed-methods approach to assess the activity’s employment multiplier model; its efforts 
to improve financial institution lending for micro, small, and medium enterprises (MSMEs); and its efforts 
to facilitate investment in MSMEs. The evaluation identified several issues causing inaccuracies with the 
activity’s jobs creation model. These include issues related to properly attributing jobs created to activity
interventions, failure to count jobs saved, the model’s extreme sensitivity to classification of MSME by 
sector, and the activity’s use of “job years” to count jobs created (compared to unique jobs created). 
Further, staff turnover at NN resulted in the loss of institutional knowledge about how its jobs creation 
model works. The activity’s efforts to train Savings and Credit Cooperatives (SACCOs) has successfully 
improved their institutional capacity. Partnering with larger financial institutions to increase their 
orientation towards MSMEs has been more of a challenge, and progress has been slow. However, 
partnerships with non-banks have successfully facilitated access to financial services for MSMEs who are, 
in turn, satisfied with those services. The activity’s investment facilitation efforts have led to investments 
in MSMEs that have increased employment and revenue at beneficiary MSMEs. However, MSMEs going 
through NN’s investment pipeline reported a lack of responsiveness by the activity and the pipeline 
process has been time consuming, taking 10 months on average to procure investment.
TABLE OF CONTENTS
Executive Summary ................................................................................................................................. i
EVALUATION OVERVIEW................................................................................................................................. i
ACTIVITY BACKGROUND ................................................................................................................................. i
EVALUATION DESIGN AND LIMITATIONS.................................................................................................. i
FINDINGS AND CONCLUSIONS.......................................................................................................................ii
RECOMMENDATIONS ......................................................................................................................................iii
1.0 EVALUATION OVERVIEW...........................................................................................................1
1.1 EVALUATION PURPOSE AND AUDIENCE...................................................................................... 1
1.2 EVALUATION QUESTIONS.................................................................................................................. 1
2.0 ACTIVITY BACKGROUND ...........................................................................................................1
2.1 ACTIVITY DESCRIPTION .................................................................................................................... 1
2.2 TARGET AREAS AND GROUPS.......................................................................................................... 2
2.3 APPROACH AND IMPLEMENTATION............................................................................................. 2
2.3.1 The Three activity Components ................................................................................................................... 2
2.3.2 Employment Multiplier Model ...................................................................................................................... 3
3.0 EVALUATION METHODS AND LIMITATIONS........................................................................3
3.1 DATA COLLECTION METHODS......................................................................................................... 3
3.1.1 Document Review ............................................................................................................................................... 3
3.1.2 KIIs ............................................................................................................................................................................. 4
3.1.3 Online Survey ........................................................................................................................................................ 4
3.2 METHODOLOGICAL LIMITATIONS ................................................................................................. 5
3.2.1 COVID-19 and Remote Fieldwork ................................................................................................................ 5
3.2.2 Response Rates .................................................................................................................................................... 5
3.2.3 Limited Representativeness of Financial Transaction Sample ....................................................... 6
3.2.4 Limited Available Respondent Expertise on Employment Multiplier Model ........................... 6
4.0 FINDINGS AND CONCLUSIONS..................................................................................................6
4.1 EQ 1a: How accurate is Nguriza Nshore’s measurement of its direct job creation 
impact likely to be?........................................................................................................................................... 6
4.1.1 Background – How NN Defines and Measures Direct Jobs ............................................................... 6
4.1.2 Finding – Lack of Indicator Definitional Clarity and Overly Aggressive Finance 
Facilitation Attribution Leads to Overestimation of Direct Job Creation ................................................ 7
4.1.3 Finding – Not Accounting for Jobs Saved Leads to Underestimation of NN’s Direct Job 
Creation ................................................................................................................................................................................ 8
4.1.4 Finding – Sector Differences Drive Differences in Job Creation Estimates ............................... 9
4.1.5 Finding – Jobs vs. Job Years .......................................................................................................................... 10
4.1.6 Conclusions .......................................................................................................................................................... 10
4.2 EQ 1b: How appropriate and well-implemented is the macro employment multiplier 
model?................................................................................................................................................................ 11
4.2.1 Background .......................................................................................................................................................... 11
4.2.2 How Appropriate is the Macro Model? .................................................................................................... 12
4.2.3 How Well Implemented Is the Macro Model? ...................................................................................... 12
4.2.4 Conclusions .......................................................................................................................................................... 13
4.3 EQ 2: HAVE FIs Improved lending for MSMES?....................................................................... 14
4.3.1 Background .......................................................................................................................................................... 14
4.3.2 Finding – Trainings Influenced SACCOs to Make Positive Changes ............................................ 14
4.3.3 Finding – NNs FI Partner Clients Are Satisfied, but Partnerships Have Stalled or Been 
Slow to Start ...................................................................................................................................................................... 16
4.3.4 Conclusions .......................................................................................................................................................... 17
4.4 EQ 3: how effective are NN’s Investment facilitation efforts ............................................. 17
4.4.1 Background .......................................................................................................................................................... 17
4.4.2 Finding – Investment Pipeline Is Effective but Slow, and Communication Could Be 
Improved ............................................................................................................................................................................ 18
4.4.3 Finding – Investments Increase Income and Revenue ..................................................................... 19
4.4.4 Conclusions .......................................................................................................................................................... 20
5.0 RECOMMENDATIONS...............................................................................................................20
5.1 EQ1c – How Can the Employment Model Be Improved? ..................................................... 20
5.2 EQ2 – Financial Institutions .......................................................................................................... 22
5.3 EQ3 – Investment Pipeline............................................................................................................. 22
6.0 ANNEXES .....................................................................................................................................24
ANNEX 1: PEEL TASK ORDER EXPRESSION OF INTEREST (EOI) – PERFORMANCE 
EVALUATION .................................................................................................................................................... 25
ANNEX 2: BIBLIOGRAPHY............................................................................................................................ 39
ANNEX 3: LIST OF KEY INFORMANT INTERVIEWS (KIIS)................................................................. 40
ANNEX 4: INTERVIEW GUIDES................................................................................................................... 42
ANNEX 5: ONLINE SURVEY QUESTIONNAIRE........................................................................................ 51
ANNEX 6: SIGNED CONFLICT OF INTEREST (COI) AND NON-DISCLOSURE AGREEMENT 
(NDA) FORMS................................................................................................................................................... 54
LIST OF TABLES
Table 1: KIIs by Type of Stakeholder
Table 2: Estimation of Upward Bias in Direct Job Creation Figures for 2020
Table 3: Jobs Created and Saved (n = 16)
Table 4: Direct Jobs Created – Fieldwork vs. Employment Model (n = 21)
Table 5: Disaggregation of Direct Jobs Created by Sector
Table 6: Job to Employment Multipliers
Table 7: SACCO Training Learning and Implementation (n = 80)
Table 8: Increase in SACCO Lending Post Training
Table 9: MSME Job Creation and Revenue Increase Post Investment (n = 9)
LIST OF ACRONYMS
Acronym Description
AMIR Association of Microfinance Institutions in Rwanda
EQ Evaluation Question
ET Evaluation Team
FI Financial Institution
GOR Government of Rwanda
I-O Input-Output 
KII Key Informant Interview
MFI Microfinance Institution
MSME Micro, Small, and Medium Enterprise
NN Nguriza Nshore
PIRS Performance Indicator Reference Sheet
SACCO Savings and Credit Cooperative
SAM Social Accounting Matrix
SME Small and Medium Enterprise
SOP Standard Operating Procedures
USAID United States Agency for International Development
USG United States Government
i 
EXECUTIVE SUMMARY
EVALUATION OVERVIEW
This midterm performance evaluation of the Feed the Future Nguriza Nshore (“Lend So I May Invest”) 
(NN) activity in Rwanda was commissioned by the United States Agency for International Development 
(USAID). The five-year activity, which began in 2018, aims to increase access to finance for micro, small, 
and medium enterprises (MSMEs). The purpose of this evaluation was to assess the progress of and learn 
from NN, to assess the accuracy of the employment multiplier model used to measure the activity’s job 
creation impact, and to provide evidence and recommendations to improve USAID’s future access to 
finance programming. The evaluation questions (EQs) are shown in abbreviated form below.
● EQ1a. How accurate is NN’s measurement of its direct job creation likely to be?
● EQ1b. How appropriate and well-implemented is the macro model used by NN to estimate 
indirect and induced job creation?
● EQ2. To what extent have partner financial institutions (FIs) supported by NN been interested in 
and capable of improving lending services and tools for MSMEs? Has support from NN increased 
the orientation of FIs towards MSMEs and increased the amount they lend to MSMEs?
● EQ3. How effective are NN’s investment facilitation efforts (including all four stages) at increasing 
revenue and employment? 
ACTIVITY BACKGROUND 
NN aims to facilitate access to finance for MSMEs with the potential to drive rural job creation, economic 
growth, and poverty reduction. The activity aims to do this through three components.
Component 1 – Improve the capacity of FIs to finance MSMEs. NN provides technical 
assistance to partner FIs on a demand-driven basis to improve their capacity to finance MSMEs.
Component 2 – Increase investment in MSMEs. NN aims to increase private and 
commercial investment in MSMEs in Rwanda. The activity works with investment advisor partners 
to link MSMEs with investors through an investment facilitation pipeline. 
Component 3 – Strengthen the policy environment for MSME finance. This component 
is excluded from the evaluation.
The activity commissioned the creation of a custom employment multiplier model to track its job creation 
impact. The model uses an Input-Output (I-O) model, which uses national macroeconomic data to
estimate the total job creation impact of NN. The total estimated job creation number generated by this 
model includes three types of jobs created:
● Direct jobs – Jobs created with NN beneficiary MSMEs as a direct result of NN activity. 
● Indirect jobs – Jobs created in the same value chain from increased levels of income, employment, 
and spending at activity-supported MSMEs.
● Induced jobs – Jobs created an economy wide outside of the value chains of supported MSMEs 
due to increased levels of income, employment, and spending induced through activity support. 
EVALUATION DESIGN AND LIMITATIONS
This evaluation team (ET) used a mixed-methods approach to answer the EQs. They used desk research 
to inform the evaluation design and contributed to answering the EQs. The ET also reviewed the activity’s 
Microsoft Excel-based employment multiplier model. Qualitative data collection methods included key 
informant interviews (KIIs) with MSMEs, FIs, activity staff, investment advisor partners, investors and
others. Quantitative data collection methods included an online survey of activity-assisted Savings and 
Credit Cooperatives (SACCOs) and a quantitative component embedded in the MSME KII guides 
measuring job creation. Due to COVID-19, the ET conducted their fieldwork remotely, doing interviews 
ii
via phone or alternate telecommunications. In total, the ET conducted 71 KIIs, while 80 of 173 (46.2 
percent) of the SACCOs invited to take the online survey completed it.
The evaluation was subject to several limitations. One was COVID-19 and, more specifically, the 
implications of remote fieldwork. The primary implication was the low response rates for certain 
respondent types, particularly MSMEs. This was partially offset by data gained through the duplication of 
some questions across respondent types and data collection tools. Another limitation was the lack of 
representativeness of information received from MSMEs about NN-facilitated financial transactions. This 
was a function of low response rates by MSMEs with which the ET requested interviewers and the difficulty 
asking MSMEs about each of the (at times numerous) financial transactions they had completed with NN
facilitation, given time limits and the difficulty of generating this information using remote interviewing 
technologies. Finally, there was limited knowledge about the employment multiplier model among NN 
staff given that the people responsible for developing the model had moved on to other employment
outside NN.
FINDINGS AND CONCLUSIONS
EQ 1a: How Accurate Is Nguriza Nshore’s Measurement?
Findings: The lack of indicator definitional clarity and overly aggressive finance facilitation attribution led 
to overestimating direct job creation under components 1 and 2. Several partners and beneficiary MSMEs 
questioned NN for taking credit for certain financial deals. In the data collected by the ET, this led to an 
overestimation of direct job creation by nearly 100 percent. On the other hand, the activity had not taken 
credit for jobs saved through the increased financial access it facilitated during the COVID19 pandemic. 
These jobs accounted for 37 percent of jobs created and saved in the evaluation sample. Comparing direct 
jobs created reported during fieldwork with what NN’s employment model predicted, the ET found that 
the model underestimated employment creation by about one-third compared to the numbers found 
through this evaluation’s fieldwork. This was likely accounted for by differences between the composition 
of NN’s food-crop sector and the corresponding sector composition at the national level on which NN’s 
employment model is based. Finally, NN reports its progress as job years rather than unique jobs. The 
former number is nearly six times higher than the latter, but this distinction was not made explicit in 
performance reports. 
Conclusions: On net, the ET found that the different sources of over- and under-estimation of direct 
job creation led to NN’s employment model counting 106 out of every 100 direct jobs created. However, 
this exercise had many methodological limitations. Its utility was seen by the ET to be in identifying sources 
of over- and under-estimation in NN’s direct employment counting approach and model rather than the 
specific degree to which the model was over- or under-counting direct job creation. 
EQ 1B: How Appropriate and Well Implemented Is the Macro Model?
Findings: NN’s employment multiplier was created using the commonly implemented I-O model, which 
links output across all sectors in an economy. The model has two components: a macro model used to 
create the multipliers, which convert the direct job creation figures to indirect and induced job creation 
figures. In turn, these multipliers are inputted into the employment micro model to produce the final job 
creation figures for direct, indirect, and induced jobs created. This model is reasonably simple to 
implement but is sufficiently rigorous to be implemented by national governments and donor agencies. 
The staff who developed and initially implemented the model, however, have left their respective positions, 
thus the model serves effectively as a “black box” for NN. Implementation was further hindered by unclear 
business sector definitions, which had a significant impact on model estimates.
Conclusions: The I-O employment multiplier model was an appropriate choice for the activity because 
of its relative ease of creation and rigor. However, implementation was hindered by two factors: lack of 
existing staff’s understanding of the model and high sensitivity to vague sector definitions.
iii
EQ 3: HAVE FIs Improved Lending for MSMES?
Findings: SACCOs were trained on governance, loan management, and business continuity. The 
evaluation found that the training for SACCOs was extremely effective—with between 68 and 99 percent 
of SACCOs reporting on the online survey that they had made improvements based on what they learned 
during training (depending on the training topic), and over 80 percent reporting that their lending volume 
increased post training. MSME clients of NN’s FI partners were generally pleased with the service they 
received. Some FI partnerships had ended by the time of this evaluation, while partnerships with 
commercial banks were difficult to initiate and slow to develop.
Conclusions: NN struggled to engage the interest of commercial banks—in part because it could not 
bring capital to the table because of activity design features. However, SACCOs and other non-bank FIs 
were more motivated to participate. NN did build the capacity of SACCOs through highly successful 
trainings; and built the capacity of partner FIs through targeted strategic and technical support. SACCOs
increased lending as a result of this support, while NN was also instrumental in raising funds for its FI 
partners. 
EQ 3: How Effective Are Nguriza Nshore’s Investment Facilitation Efforts? 
Findings: NN’s investment pipeline features four stages: sourcing, selection, investment readiness 
support and deal making, and deal closure. The pipeline was generally viewed as successful by respondents, 
with several caveats. Communication between investment advisors and MSMEs was sometimes seen as 
weak, and deals were not always structured conveniently. The pipeline was also very time consuming, with 
an average time gap between application and deal closure of 10 months, due in part to delays related to 
the COVID-19 pandemic. Nonetheless, investments did indeed create jobs and increase revenues. On 
average, creating one job cost under $7,000, and each investment increased SME revenue by 50 percent.
Conclusions: NN’s investment pipeline led to investments which increased MSME revenues and job 
creation. However, it could operate more effectively by addressing the above findings and by possibly 
better structuring incentives for its investment advisors. Investment advisors are currently paid by NN for
training and by SMEs for closing deals, whereas for commercial work they are paid for closing deals only. 
The current arrangement may lead to more training and less deal closing.
RECOMMENDATIONS
The following are recommendations for EQs1a and EQ1b:
● NN should amend its Monitoring, Evaluation and Learning Plan with guidelines for what financing 
it can take credit for having facilitated. The guidelines should be used going forward and to audit 
past investments and resulting job creation. 
● NN should validate direct job creation figures through follow-up data collection with all MSMEs 
that received financing credited to NN. This should also include an accounting of jobs saved.
● Future USAID/Rwanda activities should be encouraged to focus on measuring direct jobs created 
only, measured via direct counting rather than projected through a model.
● USAID/Rwanda should consider developing and maintaining its own multiplier model. The model 
could be used to estimate the indirect and induced impacts of its portfolio and, perhaps more 
importantly, to target interventions by sector to have the broadest impact.
The following are recommendations for EQ2:
● NN should work with the Association of Microfinance Institutions in Rwanda to develop and 
disseminate copies of the very successful training course to all SACCOs. These could be hard 
copies or online, low-bandwidth, interactive trainings. 
● USAID/Rwanda should develop an activity with an MSME finance challenge fund to develop more 
FIs catering specifically to MSMEs.
iv
The following are recommendations for EQ3:
● MSMEs starting but not completing the investment pipeline should be given templates for 
documents used to attract investments. These can be used to help them attract financing 
elsewhere.
● Future activity designs should pay investment advisors through higher deal closing fees rather than 
based on the number of MSMEs trained.
● USAID/Rwanda should design investment activities for MSMEs intended to better incentivize 
attracting commercial investments in high-growth, upside MSMEs. These investments have high 
impact potential but require the willingness and ability for activities to take some risks.
1 
1.0 EVALUATION OVERVIEW
1.1 EVALUATION PURPOSE AND AUDIENCE
The United States Agency for International Development (USAID) commissioned a midterm performance 
evaluation of the Feed the Future Nguriza Nshore (NN) activity (“Lend So I May Invest”) in Rwanda. The 
five-year NN activity, which began in 2018, aims to increase access to finance for micro, small, and medium 
enterprises (MSMEs). The purpose of this evaluation was to assess progress and learn from two of NN’s 
three components, to assess the accuracy of the employment multiplier model used to measure the 
activity’s job creation impact, and to provide evidence and recommendations to improve USAID’s future 
access to finance programming.
This performance evaluation was conducted about two-thirds of the way through NN implementation, 
which has been drastically affected by the corona virus disease 2019 (COVID-19) pandemic and resulting 
Government of Rwanda (GOR) lockdown measures. Because of these two factors, this evaluation is not 
primarily intended to be an assessment of NN performance to date, with recommendations for NN 
second-half changes. Rather, the main purpose is to evaluate NN’s design and implementation to inform 
and improve future USAID/Rwanda programming. However, this evaluation will address any issues 
identified that can and should be addressed in the activity’s remaining 18 months. The primary audience 
for this evaluation is USAID/Rwanda.
1.2 EVALUATION QUESTIONS
The NN midterm performance evaluation answers the following three Evaluation Questions (EQs).
1. How accurate is the NN multiplier model?
a. How accurate is NN’s measurement of its direct job creation impact likely to be?
b. How appropriate and well implemented is the macro model used by NN to estimate indirect 
and induced job creation?
c. How can the model or its use be improved to increase the accuracy and utility of activity￾reported data? (Note: This will be addressed in the recommendations.)
2. To what extent have partner financial institutions (FIs) supported by NN been interested in and 
capable of improving lending services and tools for MSMEs? Has support from Nguriza Nshore 
increased the orientation of FIs toward MSMEs and increased the amount they lend to MSMEs?
3. How effective are the NN activity's investment facilitation efforts (including all four stages) at
increasing revenue and employment? 
2.0 ACTIVITY BACKGROUND
2.1 ACTIVITY DESCRIPTION
MSMEs account for 97 percent of businesses in Rwanda, contribute about 55 percent of the total gross 
domestic product, and employ 41 percent of the population. A 2017 study found that while 90 percent of 
MSMEs had an account in a formal institution, 79 percent of them were unable to meet their financing 
needs.1 Several challenges constrain access to adequate financing. In general, FIs have a limited range of 
products suitable for MSMEs, while MSMEs often lack the paperwork and collateral requirements to secure 
loans or adequately sized loans. 
1https://www.microsave.net/files/pdf/BN_170_MSME_Finance_in_Rwanda_Status_and_Opportunity_for_Financial_Institutions.
pdf
2 
NN seeks to address these constraints and facilitate access to finance for MSMEs, especially agribusinesses
with the potential to drive rural job creation, economic growth, and poverty reduction. The activity aims 
to do this by working with FIs to improve their capacity to lend to MSMEs, facilitating MSME access to 
investment, and improving the enabling environment for MSMEs.
2.2 TARGET AREAS AND GROUPS
NN’s geographic focus prioritizes the 13 districts of the new Feed the Future Zone of Influence. These 
districts include Bugesera, Burera, Gakenke, Gatsibo, Karongi, Kayonza, Ngoma, Ngororero, Nyabihu, 
Nyamasheke, Nyamagabe, Rubavu, and Rutsiro. The activity also works in other areas, notably Kigali, to 
maximize impact.
A consortium of implementing partners led by DAI Global implements NN. DAI provides overall vision, 
leadership, and management for the activity, while the partner investment advisors led by the BID 
Network source MSMEs, provide them with technical assistance, and link them to investors. The activity’s 
other partners and beneficiaries are described below:
● FIs: The activity provides technical assistance and consulting to FIs (i.e., banks, non-bank FIs, and 
Microfinance Institutions [MFIs]) to bolster their know-how and capacities to deliver financial 
products and services to MSMEs.
● MSMEs: The activity provides technical assistance and consulting support to MSMEs, which are 
the direct beneficiaries of activity support. 
● Vanguard Economics: Vanguard Economics helped develop the employment multiplier model.
● The GOR, donors, donor-funded activities, and industry groups: The activity shares 
resources and expertise and coordinates with the GOR, domestic industry associations, and 
donors/implementers to better achieve shared objectives for MSME access to finance.
2.3 APPROACH AND IMPLEMENTATION
2.3.1 The Three activity Components
NN is implemented through three components, described below.
Component 1 – Improve the capacity of FIs to finance MSMEs. Nguriza Nshore provides technical 
assistance to partner FIs on a demand-driven basis to improve their capacity to finance MSMEs. Activities 
include:
● Providing training to 170 Savings and Credit Cooperatives (SACCOs) provided through the 
Association of Microfinance Institutions in Rwanda (AMIR). These trainings covered governance, 
loan management, and business continuity planning.
● Providing technical assistance to FIs to help them service MSMEs better, including helping to create 
MSME strategies.
● Facilitating access to finance for FIs or providing loan guarantees for MSMEs to enable them to 
access commercial finance from FIs.
Component 2 – Increase investment in MSMEs. NN aims to increase private and commercial 
investment in Rwandan MSMEs. The activity works with transaction advisor partners to link MSMEs with 
investors through an investment facilitation pipeline. This involves the four-stage process described below.
● Stage 1: Sourcing – NN identifies MSMEs through public events and referrals.
● Stage 2: Screening and selection – NN selects businesses to continue through the pipeline based 
on their management, established client base and revenue, and potential for impact through 
increased revenue and job creation.
3 
● Stage 3: Investment readiness support and deal making – NN investment advisors provide 
technical assistance and training to MSMEs in areas like business plan development, financial 
modeling, and pitch deck preparation to help MSMEs become investment ready. 
● Stage 4: Deal closure – NN investment advisors introduce MSMEs to investors and attempt to 
help them to close deals, including attending meetings with investors. 
Component 3 – Strengthen the policy environment for MSME finance. NN works to strengthen 
the policy environment and finance and investment ecosystems. Activities under Component 3 include:
● Support GOR in developing and passing the Entrepreneurship Development Policy, incorporating 
input from many stakeholders to make GOR more responsive to the private sector. 
● Work to improve regulations for attracting MSME investors and technical assistance to the 
Business Development Fund. 
Note that Component 3 interventions are not included as part of this evaluation.
2.3.2 Employment Multiplier Model
NN commissioned the creation of a custom input-output employment-multiplier model to track the 
activity’s job creation impact. This model uses national macroeconomic data to estimate NN’s total job 
creation impact. The total estimated job creation number generated by the model includes three types of 
jobs:
● Direct jobs – Jobs created with NN beneficiary MSMEs as a direct result of NN activity. 
● Indirect jobs – Jobs created in the same value chain due to increased levels of income, employment, 
and spending at activity-supported MSMEs.
● Induced jobs – Jobs created economy wide outside of the value chains of supported MSMEs due 
to increased levels of income, employment, and spending induced through activity support. 
The activity inputs either expected jobs created, revenue changes, or the amount of investment facilitated 
into the model, which, in turn, produces a total employment estimate. Direct jobs are the subject of EQ1a, 
whereas indirect and induced jobs are the subject of EQ1b.
3.0 EVALUATION METHODS AND 
LIMITATIONS
The Evaluation Team (ET) consisted of three members, including a team leader access to finance specialist, 
and evaluation specialist. The ET conducted fieldwork in August 2021 using a mixed-methods approach 
to answer the EQs. Qualitative data collection methods included a document review (also used to inform 
the evaluation design) and key informant interviews (KIIs) with MSMEs, FIs, project staff, investment 
advisor partners, investors, and others. Quantitative data collection methods included an online survey of 
SACCOs and a quantitative component of MSME KIIs measuring job creation. Each of the above methods 
is described in greater detail below. 
3.1 DATA COLLECTION METHODS
3.1.1 Document Review
Document Review
The ET consulted four types of documents for this evaluation. They consulted documents and data about 
Rwanda’s economy and access to finance to better understand the context for the activity and evaluation. 
They reviewed project documents, including annual performance reports, to understand project design 
4 
and implementation. The ET further reviewed literature on employment multiplier models as part of the 
process of answering EQ1b. Finally, the ET reviewed the Microsoft Excel-based employment multiplier 
model created for NN by Vanguard Economics, which divided the model into two parts in separate 
worksheets, each with multiple tabs. (See a bibliography of documents reviewed in Annex 2.)
3.1.2 KIIs
Stakeholders Interviewed
Because of the COVID-19 pandemic, the ET conducted KIIs remotely via phone or via alternative distance 
communications technologies, after which they typed up their uploaded to the cloud each day using 
Google Forms, which allows notes to be downloaded in tabular format by EQ. Overall, the ET conducted 
70 KIIs, which are enumerated by stakeholder type in Table 1 below. (See Annex 3 for a list of stakeholders 
interviewed.)
Table 1: KIIs by Type of Stakeholder
Respondent type Total KIIs
SACCOs 20 
MSMEs – Component 1 12 
MSMEs – Component 2 (no investment) 8 
MSMEs – Component 2 (investment) 11 
FIs 4 
Investment advisors 3 
Investors (through investment advisors) 2 
NN staff or consultants 6 
Associations 4 
Total 70
The ET conducted KIIs with SACCOs and MSMEs using standardized interview guides, which ensured that 
they asked the same information type across similar respondent types. The interview guides for 
Component 1 and Component 2 MSMEs included a quantitative job creation section used to identify the 
relationship between access to finance and job creation at the MSME level. The team leader conducted 
the remaining KIIs with FIs, investment advisors, investors, NN staff or consultants, and association 
members, which were more conversational and less structured to facilitate a greater level of 
responsiveness to the unique stakeholder types. (See Annex 4 for copies of the KII guides.)
KII Questions on Job Creation
Because many MSMEs that received financing through the NN activity received multiple rounds of 
financing, it proved too confusing to ask respondents to associate any job creation to all funding received. 
Instead, the KII guides asked about the largest single financing the MSME had received through the activity. 
They asked respondents to not only provide the number of jobs created through NN-facilitated financing, 
but also to provide causal qualitative explanations to link the financing to the claimed job creation. 
Analysis of KII Data
The ET extracted quantitative data on financing and job creation from the interview notes with MSMEs 
and entered them into Excel for quantitative analysis. They reviewed the qualitative responses recorded 
in the interview notes by interview question and then excerpted, summarized, and paraphrased them by 
evaluation question into a detailed final evaluation report outline. 
3.1.3 Online Survey
The ET administered an online survey to SACCOs that received training through NN. (NN had only four 
mature FI partners, one of which disputed its partnership status with the activity. In contrast, NN 
5 
supported about 200 SACCOs, with training provided by AMIR. Thus, the online survey focused on 
activity-supported SACCOs.) The survey asked respondents to evaluate how much they learned through 
the training and if they made any changes based on the training. The ET emailed a link to the online survey
to representatives at173 SACCOs trained by NN, along with a letter from their umbrella organization, 
AMIR, requesting their participation. Eighty persons responded to the survey for a response rate of 46.2 
percent. The ET analyzed the survey responses using the statistical software program Stata. (See Annex 5 
for a copy of the online survey questionnaire.)
3.2 METHODOLOGICAL LIMITATIONS
3.2.1 COVID-19 and Remote Fieldwork
Because of the COVID-19 pandemic, the ET carried out its fieldwork remotely, which had implications 
for multiple aspects of fieldwork. The first was building rapport between team members to allow them to 
function well together and overcome inevitable challenges. The ET addressed this limitation by holding 
regular meetings before fieldwork and having all team members discuss the fieldwork plan and provide 
inputs. Second, the management of fieldwork had to be remote. This was overcome through features of 
the evaluation design. One was the requirement for daily uploading of notes so the team leader could 
monitor them online. Also, the team held regular meetings via WhatsApp to recap progress and address 
challenges in real time.
3.2.2 Response Rates
Perhaps the most significant fieldwork limitation was interview response rates. Businesses are often 
reluctant to speak to evaluators, especially about business details like investment and employment. Those 
involved with NN were no exception. Remote fieldwork posed further challenges, as respondents often 
prefer face-to-face meetings to phone conversations, especially when they are busy, and interruptions are 
inevitable. To increase response rates, NN or partner FIs facilitated introductions for the ET to targeted 
stakeholders. Nevertheless, soliciting responses from certain stakeholders was a challenge, as detailed 
below.
● SACCOs were relatively responsive, and the ET was able to meet its target of 20 interviews. The 
SACCO members who were contacted but not responsive had moved to different jobs, had not 
participated in training, or were unreachable at the listed phone numbers. 
● MSMEs for Component 1, who had received debt financing from NN partner FIs, had a response 
rate of 100 percent; all 12 contacts were successfully interviewed. 
● The most challenging type of interview was with MSMEs that had been selected through 
Component 2 but had not received investment. This type of respondent was especially 
unresponsive, and the ET was only able to interview eight of 45 introduced to them by NN, falling 
short of the target of 20 completed interviews. It is not surprising that these stakeholders were
unresponsive, as many did not receive the level of investment they had expected. Several did not 
respond to email or were unreachable by phone. Moreover, others who did respond had not 
heard of NN or said they had not applied for financing with NN. Discussions with NN revealed 
that many events where MSMEs were identified for possible investment opportunities were GOR 
events where activity branding was not permitted. 
● The response rate for MSMEs who received investment through Component 2 was also relatively 
low, although not as low as for those who did not receive investment. The ET interviewed 11 of 
21 MSMEs introduced to them by NN, with some unreachable by phone and others declining to 
be interviewed. 
● The online survey response rate of 47 percent was consistent with other online surveys 
conducted by the team leader. 
6 
The extent to which this non-responsiveness biases the data depends on the extent to which those who 
responded differ in meaningful ways from those who did not. This can only be subject to speculation since 
it is difficult to learn much about the potential respondents that were unwilling or unable to respond. 
Based on what the ET knows about the prospective respondents that did not respond, its sense is that 
the non-respondents did not differ systematically from respondents. 
In anticipation of this non-responsiveness, the evaluation fieldwork plan included some redundancy to 
avoid gaps or information losses due to unresponsiveness. For example, MSME KIIs and the online survey 
included some similar questions, and MSMEs and FIs both provided perspectives on the MSME orientation 
and FI capabilities. 
3.2.3 Limited Representativeness of Financial Transaction Sample
The ET was able to find data for only a small share of the financial transactions NN takes credit for 
facilitating. For example, for Component 2, NN collected data on just 10 of the 76 NN-facilitated 
investments. For Component 1, it was a similar story. Compounding this challenge was the non￾responsiveness of Component 2 beneficiaries. Finally, many MSMEs had multiple transactions facilitated by
NN, some more than 10, and the time required to ask about each one would have been prohibitive. 
3.2.4 Limited Available Respondent Expertise on Employment Multiplier Model 
Another limitation of this evaluation was that the people most knowledgeable about the employment 
multiplier model are no longer with their former employers. This includes the key NN staff members and 
the person leading the development of the model for Vanguard Economics, which developed the model 
for NN. As a result, it was difficult for the ET to get answers to all but the most basic questions about the 
model, which invariably limited the depth of analysis of the model. 
4.0 FINDINGS AND 
CONCLUSIONS
This section presents the findings and conclusions for each EQ along with some background to help 
contextualize them. While there are multiple findings for each EQ, there is only one conclusion per EQ. 
These conclusions aim to provide direct, explicit answers to the EQs through analysis and synthesis of the 
findings. 
4.1 EQ 1A: HOW ACCURATE IS NGURIZA NSHORE’S MEASUREMENT OF ITS 
DIRECT JOB CREATION IMPACT LIKELY TO BE?
4.1.1 Background – How NN Defines and Measures Direct Jobs
Direct jobs created are those that result from financing facilitated by NN, which counts two types. The 
first is jobs created at MSMEs by NN’s partner FIs through the lending activity for which NN takes credit. 
These types of jobs fall under Component 1. The amount of financing provided by partner FIs for which 
NN claims credit is entered into the employment multiplier model, which then provides employment 
estimates based on the amount of financing provided. 
The second type of direct job created by NN is those jobs created by investments made in MSMEs through 
the investment pipeline in Component 2. These jobs are an average of the expected jobs estimated in the 
future from the investment and the estimated jobs based on the expected revenue change. Shortly after 
deal closure, the investment advisors provide both projections, employment and revenue created, to NN. 
In cases where jobs created or revenue projections are unavailable, NN inputs the investment amount 
described with Component 1 above to produce a projection of job creation. 
7 
4.1.2 Finding – Lack of Indicator Definitional Clarity and Overly Aggressive Finance 
Facilitation Attribution Leads to Overestimation of Direct Job Creation
Nguriza Nshore lacks a clear and detailed definition for its custom indicator “Number of jobs created 
with USG assistance.” According to the Performance Indicator Reference Sheet (PIRS) for this custom 
indicator, the number of jobs created with United States Government (USG) assistance, “… counts direct 
employment created as a result of direct assistance from Nguriza Nshore....” However, in the same PIRS, 
direct jobs are also defined as “those jobs that are created in the small, medium, and large firms with 
which Nguriza Nshore or one of the activity’s partners are [sic] directly working.” The difference here is 
subtle but significant as the former definition requires attribution to the activity (“as a result of”), whereas 
the latter does not. Through interviews with activity staff and beneficiaries, it became clear that the former 
definition—which requires an attempt to make a claim that job creation is caused by NN and not just 
coincidental—is the definition that was implemented by the activity. 
Attribution is often a challenge when trying to account for an activity’s impact. In this case, the challenge 
is defining what constitutes direct NN assistance when it comes to job creation attributable to NN. The 
activity has no explicit guidelines defining what access to finance it can claim credit for and what it cannot. 
The result is definitional ambiguity that has, in the view of the ET and certain activity partners and 
beneficiaries, allowed the activity to be overly aggressive in taking credit for job creation. Below are four 
cases in which NN appears to have overestimated its role in job creation. (Note that because of 
confidentiality agreements with respondents, the ET cannot disclose the identities of these partners and 
beneficiaries.)
1. One partner FI took issue with the nature of the partnership claimed by NN. Specifically, it claimed 
that its collaboration with NN was limited to attending an entrepreneurship policy workshop. NN
took credit for a large sum of funding, which the partner FI claims to have procured through a 
preexisting relationship with a funder unrelated to its participation in the workshop or any other 
NN activities. For its part, NN contests this characterization of events and claims that the previous 
executive director of the FI partner had an agreement with NN that would justify NN taking 
credit for the funding. The FI partner raised its concern with both NN and USAID. Its concern 
was not that NN’s job creation impact might be overstated but, rather, that it was not receiving 
due credit for raising financing in the small MSME financing ecosystem in Rwanda. From the 
perspective of responding to EQ1a, this contested large funding amount taken credit for by NN 
had a significant impact on NN’s reported job creation. (The amount of funding is not revealed to 
protect the respondent’s anonymity.)
2. Another FI partner questioned the assumptions in the quarterly reports that NN sent it to sign. 
The contested assumptions related to the amount of its lending that NN took credit for facilitating. 
The FI deemed the assumptions in question to be inaccurate and biased in a direction that would 
positively favor NN job creation results. For example, the forms included assumptions of 100 
percent recovery of loans from capital that NN had facilitated for the FI. In fact, the recovery rate 
was lower, especially during the pandemic when only about 50 percent of loans have been 
recovered. They also included assumptions that all capital was to be disbursed as productive loans 
when, in fact, both parties had agreed that some was to be used for overhead. Finally, the forms 
had assumptions of a cycle-through rate—the time from loan disbursement, recovery, and 
disbursement—that was deemed to be significantly shorter than was realistic. These assumptions, 
both individually and collectively, led to higher job creation estimates.
3. One MSME, for which NN has taken credit for facilitating multiple and significant investments, was 
unaware that NN had facilitated access to these investments. It had approached one of NN’s 
investment advisors independently seeking its help to raise funds. As it turns out, the investment 
advisor was partnering with NN and received payment for investment support. The support this 
partner received consisted of helping the MSME put together a pitch slide deck to present to 
8 
investors. The MSME paid the investment advisor a deal closing fee, which it thought was full 
renumeration for the support it had received to attract the investment. 
4. Another MSME interviewed had no issues with NN taking credit for some investments it had 
attracted but took issue with others. For certain investments, it viewed NN’s support as 
instrumental, but for others, it viewed NN’s support as incidental claiming that the MSME did 
most of the work and would have been able to attract that same investment without NN. 
Table 2 provides a rough estimate of the extent to which NN overestimated job creation (i.e., claimed 
credit or attribution for jobs created that were generated through other means) during 2020, for reasons 
discussed above. The table provides separate estimates for FI partners (Component 1) and MSMEs 
receiving financing (Component 2). (Note that because of anonymity agreements with respondents, FI or 
firm-level real numbers are not used, as they could identify respondents.)
For the FI partners described in Cases 1 and 2 above, the ET derived the estimate of overcounting by 
subtracting the disputed capital from 2020 figures for FI total lending reported by NN. For Case 1, the 
entire amount of capital for which NN took credit was removed. For Case 2, the percentage disputed 
was removed from that partner’s total. For Component 2, the disputed financing for 2020, including Case 
3 above, was removed. In both cases, the new figures for financing credited to NN was compared with 
the original. That ratio was then multiplied by the direct jobs created figures to produce the adjusted jobs 
figures.2
These adjustments result in a significant decrease in the financing credited to NN, and consequently job 
creation. Table 2 below shows that the originally reported figures overestimate direct jobs creation by an 
estimated 92 percent over the adjusted figures. 
Table 2: Estimation of Upward Bias in Direct Job Creation Figures for 2020
Component 1 Component 2 Total
Direct jobs created – original 538 377 915 
Original finance credited to NN 
as a percentage of adjusted
247% 146% 192% 
Direct jobs created – adjusted 218 259 477 
4.1.3 Finding – Not Accounting for Jobs Saved Leads to Underestimation of NN’s 
Direct Job Creation 
Just as with a job created, a job saved means that someone has a job who would not otherwise have one. 
However, this definition was not included by NN in its job creation measurement methodology. This 
omission led to an undercounting of NN job creation, especially during the COVID-19 pandemic when
working capital was scarce and access to financing was desperately needed to keep businesses afloat. 
For example, one MSME said it would have gone bankrupt last year if not for the investment received 
through NN. This would have meant about 90 employees, including 60 full-time employees, losing their 
jobs. Two other MSMEs were able to use the loans to pay salaries. Another MSME had put two full-time 
employees on temporary leave because it could not pay them, but when it received financing through NN, 
it was able to rehire them. Yet another MSME said it would have had to lay off one full-time worker and 
40 casual workers had it not received the NN-facilitated investment. 
2 Because the employment multiplier model is anonymized, the ET assumes here that the ratio between original access to financing 
and the adjusted figure is the same as the ratio of original direct jobs created to the adjusted figure for direct jobs created. This 
assumption is not necessarily true given that the mix of MSME sectors might change, and there are different employment 
multipliers for different sectors. Nevertheless, the use of this assumption is not deemed likely to significantly change the results 
of this exercise. 
9 
Table 3 shows the reported full and casual (including part-time) jobs created and saved, according to the 
data collected through this evaluation. In summary, 37 percent of the total jobs reported during the 
fieldwork were jobs saved. This provides evidence of the significant extent of the undercounting caused 
by not including jobs saved, especially during the COVID-19 pandemic. 
Table 3: Jobs Created and Saved (n = 16)
Full-Time Part-Time + 
Casual
Total Jobs
Jobs created (and % of total) 83 (54%) 169 (69%) 252 (63%)
Jobs saved (and % of total) 71 (46%) 76 (31%) 147 (37%)
Total 154 245 399
4.1.4 Finding – Sector Differences Drive Differences in Job Creation Estimates
The key parameter that influences the accuracy of direct job creation in NN’s employment model is the 
ratio between financing accessed by MSMEs and the job creation that resulted. The ET used this ratio to 
estimate all the direct jobs created under Component 1 and some of the jobs created for Component 2 
where the MSMEs or investment advisors were unable to provide estimated job or revenue creation 
expected to result from the financing. In total, during 2020, this method estimated 73 percent of the direct 
jobs counted by NN.
The employment model calculates this ratio based on sector-level national data from surveys conducted 
by the GOR. The key inputs are the assets-to-revenues ratio, the employment expenses-to-revenues 
ratio, formal and informal employee earnings, and the share of workers in each sector (which are formal 
versus informal). In other words, the model says that an investment will increase an MSME’s assets, which 
will, in turn, increase its revenue in proportion to the average asset-to-revenue ratio in that sector. This 
increase in revenues will next increase employment in proportion to the employment expenditure-to￾revenue ratio in that sector. Finally, this increase in employment expenditure is divided by the average 
wages in that sector to produce the number of jobs expected to result from investment.3 
The ET tested the investment-to-jobs-created ratio produced by the model by comparing predicted direct 
jobs created through investments with those reported by respondents. It made this comparison at the 
investment level with investment amounts entered in the model for each business and matched with their 
sector. Matching by sector is important as the employment model produces a wide range of direct jobs 
produced by sector.4 
Table 4 below shows the comparison of the number of jobs that evaluation respondents reported were 
created by financing facilitated by NN with the number of jobs the NN employment model projected 
based on the financing they received. The total amount of financing received, on which these jobs created 
are based, was $7 million. The table shows that the employment model predicts just under two-thirds of 
the total jobs reportedly created, with 158 predicted based on NNs employment model versus 252 
reported by respondents during fieldwork. 
Table 4: Direct Jobs Created – Fieldwork vs. Employment Model (n = 21)
Full-time Part-time + 
Casual
Total Jobs
Model 53 105 158 
Fieldwork 83 169 252 
3 Direct jobs created = ([Investment / assets to revenue ratio] x [employment expenses-to-revenue ratio]) / ([formal employment 
expenses x % of jobs formal] + [informal employment expenses x % of jobs informal]). 
4 For example, for a $100,000 investment, the range of total jobs generated by sector is between 0.6 and 10.2.
10
Disaggregating the above results by sector reveals that one sector was the source of the underestimation 
by the employment model: the food crops sector (see Table 5). For these three businesses, the actual job 
creation far exceeded what the model predicted based on the financing they received (137 vs. 22). On the 
other hand, for the remaining 18 businesses not in food crops, the model slightly overestimated the 
number of jobs created (115 vs. 136). 
Table 5: Disaggregation of Direct Jobs Created by Sector
Full-time Part-time + 
Casual
Total Jobs
Model – food crops sector (n = 3) 5 17 22 
Fieldwork – food crops sector (n = 3) 7 130 137 
Model – not food crops sectors (n = 18) 48 88 136 
Fieldwork – not food crops sectors (n = 18) 76 39 115 
This discrepancy is likely caused by the difference between the average food crop producer in Rwanda, 
who is a smallholder with little employment generation potential, and the larger food crop-producing 
companies for which NN has facilitated investment. Further, the fieldwork focused on a respondent’s 
largest financing received through NN—which for several respondents in the food crop sector happened 
to be for large employment-generating projects, including providing salaries to former subsistence farmers. 
While this previous issue relates to the difference between the population-level data used in the model 
and the businesses for which NN helped facilitate financing, another measurement issue is the assumption 
of linearity when converting loans to job creation estimates. For example, the model assumes that 100 
$300 loans will generate the same number of jobs as one $30,000 loan, but this assumption may not be 
accurate. The ET interviewed several small farmers who secured FI loans with activity facilitation who 
reported only using family labor. They reported that the small loans they received had not caused them 
to hire anyone. However, this potential source of overestimation is likely far less significant than the 
sources discussed above. 
4.1.5 Finding – Jobs vs. Job Years
The jobs created figures reported in NN’s annual performance reports are job years. In other words, 
rather than reporting unique jobs created, the NN performance reports include jobs per year projected 
over a six-year period. For example, access to finance through NN may be projected to generate seven 
jobs in Year 1, eight in Year 2, nine in Years 3, nine in year 4, nine in year 5 and nine in year 6. While in 
this example the activity would report the total direct jobs generated as 51 (7 + 8 + 9 + 9 + 9 +9), only 
nine people have new jobs due to the investment.
This reporting decision leads to a nearly six-fold increase between what NN reports for jobs created and 
what might commonly be understood as a job created. The indicator PIRS does not reflect this substitution 
of job years created for unique jobs created, nor do performance reports explain it. 
4.1.6 Conclusions 
As has been shown in the previous findings, the answer to this EQ depends on the net effect of different 
sources of under- and over-estimation of direct job creation. Combining the sources of under- and over￾estimation of direct job creation documented above produces an overall estimation of job creation under￾or over-estimation, at least from the sources identified here. Table 2 above shows that the reported value 
for total direct job creation is 192 percent of the adjusted or “true” value, whereas Table 3 shows that 
11
figure to be 63 percent. Table 4 shows that the model reported value (158) is also 63 percent of the true 
value. This implies that, on net, the activity counts 106 direct jobs created for every 100 created.5 
Note that this estimate ignores the small overcounting bias likely introduced through counting financing 
to smallholder farmers. It also does not include the difference between unique jobs created and job years 
reported, a difference of nearly 600 percent.
The above calculations, however, have several limitations that should be noted. First data collection on 
investment and jobs during the evaluation fieldwork did not cover all the financing facilitated by NN. 
Neither were all those receiving financing interviewed nor were all the loans or investments discussed 
with all those who were interviewed. As has been shown, the extent of over- or under-estimation can be 
significantly influenced by a single transaction. Examples include a large investment in which NN’s role is 
disputed, a loan that prevents a business from bankruptcy, or an investment that, for whatever reason, 
generates far more or fewer new jobs than the parameters of the model predicted. The previous estimates 
are extremely sensitive to changes that could have taken place had the ET been able to interview every 
MSME that accessed financing through NN or to spend hours that would have been required to interview 
respondents that had received multiple loans or investments. Further, this evaluation was not able to test 
the job creation estimation method used for about one-quarter of the jobs created claimed by NN. Finally, 
it may be that some of the disagreements about NN’s role in facilitating financing might be adjusted in 
NN’s favor. 
Because of these limitations, the above calculation is more a result of the process of identifying sources 
of jobs creation measurement inaccuracy rather than a precise measure of the inaccuracy itself. The 
sources of inaccuracy identified here will be addressed further through recommendations to improve 
NN’s job measurement accuracy. 
4.2 EQ 1B: HOW APPROPRIATE AND WELL-IMPLEMENTED IS THE MACRO 
EMPLOYMENT MULTIPLIER MODEL? 
4.2.1 Background
NN’s employment multiplier model has two components. The macro model, which is the subject of this 
evaluation question and the findings below, is used to create the multipliers, which convert the direct job￾creation figures discussed above to indirect and induced job creation figures. Indirect jobs are those jobs 
in the same sector as the direct job, and induced jobs are those jobs created in the entirety of the 
economy—excluding the direct-job sector. These multipliers are put into the employment micro model 
to produce the final job creation figures for direct, indirect, and induced jobs created.
NN’s macro model, created by Vanguard Economics, follows the classic Input-Output (I-O) model 
developed by Wassily Leontief, who won a Nobel prize for developing the model. The I-O model is based 
on input-output tables, or supply use tables, which represent the interdependence of different sectors in 
an economy. Specifically, the tables show the value of output from each sector in the economy used by 
each of the other sectors in the economy, including the final consumer. For example, an IO table shows 
the value of output from the food crops sector used by the hotel and restaurants sector. I-O models are 
used to estimate the impact on the economy of investments, policies, and projects. Employment impacts 
are among the different impacts that I-O models can be used to assess. 
The I-O model in this case is used to create employment multipliers that estimate indirect and induced 
jobs associated with NN’s direct job creation. For example, for each job that NN generates in a business 
classified in the food crops sector, 0.2 additional indirect jobs are created in that same sector and another 
5 Calculated as the unweighted average of the three biases or, rather, the average of three biases weighted by the same true jobs 
number. Note that the possible inclination to multiply the three biases by one another would produce the total bias of biases. 
Further, these sources of under- and over-estimation are appropriate to combine as the they are mutually exclusive. Tables 2 
and 3 reveal different sources of over- and under-counting, and these are not captured in Table 4. 
12
0.02 induced jobs are created in the economy at large. Overall, across the 23 sectors included in NN’s 
jobs multiplier model, an average of 0.81 indirect jobs and 0.13 induced jobs are created for each direct 
job created. In total, the employment multiplier model used by NN projects an average of less than one 
total indirect and induced job to be created for every direct job created by NN. 
4.2.2 How Appropriate is the Macro Model? 
There are two other commonly used approaches besides I-O to estimate economy-wide employment 
impacts. One is the Social Accounting Matrix (SAM), which is an extension of the I-O model. SAM allows 
for modeling of distributional impacts as they include data on expenditure patterns of households, 
businesses, and government. However, like I-O models, SAM models are static, among other limitations. 
Their static nature means, among other things, that prices do not change based on changes in demand. 
Time-series data can be used to make both types of models dynamic, but this increases complexity and 
data requirements immensely. Computable General Equilibrium models combine data with a series of 
equations that allow them to dynamically project how the economy will respond to external changes. 
They are also more customizable and, because of this, can be programmed to relax some of the 
assumptions in the I-O and SAM models.6
The I-O model is the simplest to implement of the above three types of models because it requires the 
least data, expertise, and computational power. In fact, the model can be generated and run simply using 
Microsoft Excel, as is the case for the model used by NN. Because of its relative ease of implementation, 
the I-O approach has been popular with development projects and has been used by USAID before, 
including for a study to measure job impacts in the Haitian apparel sector.7
4.2.3 How Well Implemented Is the Macro Model?
The model appears to be well implemented technically as the ET did not identify any significant technical 
errors. However, NN does not have the in-house expertise to understand the model, and the person 
responsible for assembling it left the consulting firm. As a result, NN staff plug in the inputs and report 
the results but have little ability to adjust, correct, or improve the model, or to explain its assumptions 
and implications. Also, while there is a document describing the model at a high level, there is no manual 
to provide a detailed explanation of it. Thus, the employment multiplier model operates as a “black box” 
at NN. 
A further limitation is related to the data reporting format coming from the GOR. Specifically, the sectors 
used to classify data shown in Table 6 below are not entirely consistent with NN’s needs. For example, it 
is not clear how to classify a seed producer or wholesaler. The implications of sector classification are 
large. If you classify the seed producer under food crops, you get only 0.04 indirect and induced jobs for 
each direct job generated. If, however, you classify them under wholesaler trade, maintenance, and repair, 
the result is nearly six indirect and induced jobs for each direct job created.
6https://www.ilo.org/wcmsp5/groups/public/---ed_emp/---emp_policy/---invest/documents/publication/wcms_426586.pdf
7https://www.marketlinks.org/sites/default/files/resource/files/LEVE-Technical-Report-Apparel-Sector-Multiplier_508.pdf
13
Table 6: Job to Employment Multipliers8
Sector Direct Indirect Induced
Food crops 1.00x 0.02x 0.02x
Export crops 1.00x 0.02x 0.02x
Livestock & livestock products 1.00x 0.02x 0.02x
Forestry 1.00x 0.00x 0.01x
Fishing 1.00x 0.00x 0.02x
Mining & quarrying 1.00x 0.21x 0.06x
Food manufacturing 1.00x 1.98x 0.05x
Beverages and tobacco 1.00x 0.88x 0.04x
Other manufacturing 1.00x 0.33x 0.03x
Electricity 1.00x 1.21x 0.19x
Water & waste management 1.00x 0.30x 0.06x
Construction 1.00x 0.13x 0.03x
Wholesale trade and maintenance and repair 1.00x 4.65x 1.10x
Transport 1.00x 0.40x 0.08x
Hotels & restaurants 1.00x 2.95x 0.11x
Telecommunication 1.00x 1.20x 0.33x
Financial services 1.00x 1.79x 0.33x
Professional, scientific & technical activities 1.00x 0.52x 0.14x
Administrative & support service activities 1.00x 0.46x 0.09x
Public administration & defense; compulsory social security 1.00x 0.73x 0.07x
Education 1.00x 0.29x 0.09x
Human health & social work activities 1.00x 0.43x 0.23x
Other services 1.00x 0.12x 0.03x
Sector average 1.00x 0.82x 0.13x
4.2.4 Conclusions
The employment multiplier macro model chosen to measure NN’s indirect and induced job creation is 
appropriate for its application. The model combines an internationally recognized and rigorous 
methodology used by national governments and multilateral donors with relatively straightforward 
implementation, which is appropriate for a development project such as NN. 
The management of the model was hamstrung by staff turnover both at NN and the consulting firm 
responsible for creating the model. While the ET did not identify any pressing issues with the model that 
would require addressing, there was also no one available who could answer technical or even conceptual 
questions. The lack of staff knowledge about the model means that they would not be able to update it to 
account for, say, the shock created by COVID-19. A detailed user manual could have addressed this issue
by helping to transfer technical know-how given staff turnover. 
Another implementation issue was the vagueness of the sector definitions and the extreme sensitivity of 
the model to sector classification. This is a common challenge in working with data from national accounts, 
8 Note that the multiplier results reported here are the actual figures used in NN’s micro model to report job creation. These 
figures are different than in the document produced for USAID titled “Feed the Future Rwanda Nguriza Nshore activity 
Employment Multiplier,” dated October 15, 2019, and the macro model shared with the ET. For both these two latter sources, 
the multipliers are identical. The ET’s assumption is that the multipliers used in this document and the macro-model were created 
with Rwanda’s Supply Use Tables for 2014, and that the multipliers used in the micro model were created using the updated 2017 
Supply Use Tables. However, at the time of draft submission, the ET was not able to confirm this assumption. The multipliers are 
similar for most sectors, varying most dramatically for “wholesale trade and maintenance and repair.” As compared to the indirect 
and induced multipliers for this sector above, the multipliers from these other two sources are 0.23 and 0.06 respectively. This 
drastic change over the presumed three-year period raises questions about the source data and/or estimation of multipliers for 
this sector. 
14
which may or may not be classified according to a project’s specific needs. It is unclear whether the raw 
data could have been obtained and classified in a more useful way for NN’s application.
4.3 EQ 2: HAVE FIS IMPROVED LENDING FOR MSMES? 
4.3.1 Background
MSMEs are seen by banks as having low levels of sophistication, poor financial statements, and poor 
collateral compared with larger businesses. Banks, therefore, see them as risky and are reluctant to lend 
to them. Initially, NN hoped to partner with more banks than the one it currently partners with, but it 
faced lack of interest from potential commercial bank partners. The activity found that to get commercial 
banks interested in partnerships, it would have to bring capital to the table, which was not a part of its
design. However, international development banks extending loans to commercial banks increasingly 
require commercial banks to make efforts at MSME lending, which is what motivated one bank to partner 
with NN. Further, the economic consequences of the COVID-19 pandemic and lockdown measures may 
have forced banks to be more creative and risk taking to maintain their profitability. Recently, NN signed 
an agreement to work with a large bank to help it develop an agricultural lending program. It is also helping 
two MFIs to develop agribusiness and MSME strategies, although these are not yet completed. 
While NN always intended to work in microfinance, the initial struggles to partner with larger institutions 
meant that training SACCOs took on a larger share of Component 1. To date, NN has trained over 170 
SACCOs and plans to double this number. This assistance is provided through training delivered by AMIR, 
an umbrella organization that does advocacy and capacity development work for its over 400 members. 
The focus of training has recently turned to helping them use the Bank of Rwanda’s Economic Recovery 
Fund, which in part is aimed to support the recovery of micro and small businesses through SACCOs.
4.3.2 Finding – Trainings Influenced SACCOs to Make Positive Changes
AMIR provided trainings to over 170 SACCOs. These trainings were initially in person but were shifted 
online once the COVID-19 pandemic started. The five-day trainings covered three topics:
1. Governance subtopics, including SACCO governance structure, organization, responsibilities of 
membership, mission, leadership, transparency, and accountability.
2. Loan management subtopics, including product design and launch, member outreach, branding and 
marketing, loan analysis, risk management and follow-up, and delinquent loans.
3. Business continuity planning, focusing on staying in business during the COVID-19 crisis and 
contingency planning for future crises.
This evaluation found that these trainings were extremely well received by trainees. SACCO staff reported 
not only learning through the trainings but making changes in their SACCOs as a result. These results are 
shown below in Table 7 based on 80 completed SACCO surveys. The results show that, in all three 
training topics, respondents reported improved knowledge post training. Not only that, but most 
respondents reported that their SACCOs made improvements based on what they learned during 
trainings. The range of those reporting having made improvements based on training was 68 percent for 
loan management training and 99 percent for governance training. 
Table 7: SACCO Training Learning and Implementation (n = 80)
Question Result
Knowledge and capacity on governance topics before training (1 to 5)9 3.1
Knowledge and capacity on governance topics after training (1 to 5) 4.1
Knowledge and capacity on loan management topics before training (1 to 5) 3.3
Knowledge and capacity on loan management topics after training (1 to 5) 4.2
9 On the five-point scale, a higher score (i.e., closer to 5) indicates a more positive result.
15
Question Result
Knowledge and capacity on business continuity topics before training (1 to 5) 2.9
Knowledge and capacity on business continuity topics after training (1 to 5) 3.8
% SACCOs implementing governance improvements based on training 98.6 
% SACCOs implementing business continuity improvements based on training 75.6 
% SACCOs implementing loan management improvements based on training 68.3 
KIIs conducted with SACCOs revealed the types of post-training changes that SACCOs made. Following 
governance trainings, respondents reported making the following changes:
● Reviewed job descriptions and standard operating procedures (SOPs) or created SOPs to 
describe the roles of members.
● Held meetings between elected committee and operational management to clarify roles and 
responsibilities. Some held meetings once, some quarterly, and some monthly. 
● Improved communications with members through social media and WhatsApp groups.
SACCOs participating in loan management trainings reported making the following changes:
● Established loan repayment plan and organized home visits for business coaching and advisory 
services to reduce risks of payment default.
● Made changes to collateral requirements, including allowing land from a third party with both 
parties signing with a notary. 
● Sped up loan application assessments from two weeks to one week.
● Introduced loan restructuring for those affected by COVID-19, specifically principal without 
interest and extension.
● Changed loan product offerings to be more attractive to clients, including larger loans with longer 
repayment periods, lower interest rates, and specialized products based on sector, including 
agricultural loans based on seasonality. 
● Created a loan recovery team.
● Improved the loan application approval process using what they learned about the 5Cs—capital, 
character, capacity, collateral, and conditions.
Finally, SACCOs participating in business continuity training reported making the following changes: 
● Held awareness meetings with borrowers to help them apply for the ERF.
● Extended repayment periods because of COVID-19. 
● Providing advisory services for businesses affected by COVID-19. 
● Review of loan policy and incorporation contingency planning elements.
While feedback on the SACCO trainings was positive, there were several common suggestions related to 
challenging aspects of the training. Several respondents found that not enough time was allocated to go in 
depth on topics. Some had gone through a single-day training and others a five-day training with each 
session half a day in length. Others had network issues with online training that caused interruptions. 
Some found it difficult to escape their everyday SACCO responsibilities, which caused interruptions in 
their training participation.
Table 8 shows the reported changes in lending by SACCOs after training; 81 percent of SACCOs 
responding to the online survey increased lending post-training by an average of 18 percent. Note that 
these results should be interpreted with caution. Not only is the sample not representative—neither 
based on size or randomness of selection, but there is no counterfactual. The extreme shock of COVID￾19 means that timing of training was especially important. Those trained before the pandemic were likely 
to see a decrease in lending, whereas those trained at its peak may have reported an increase. 
16
Nevertheless, these results do provide some evidence that training increased the value of lending for 
SACCOs.
Table 8: Increase in SACCO Lending Post Training
Question Result
% Reporting that SACCOs lending increased post trainings (n = 73) 80.8 
% Increase in SACCO lending value post training (n = 58) 18.0 
4.3.3 Finding – NNs FI Partner Clients Are Satisfied, but Partnerships Have Stalled or 
Been Slow to Start
NN has three primary FI partners with whom it has worked closely enough to claim credit for some of 
their job creation.10 Because these partners and partnerships are so distinct from one another, this finding 
will address them each separately. 
Factoring
One partner is a startup that does factoring for MSMEs, using their invoices marked by buyers as delivered 
to extend short-term loans of between $5,000 and $50,000 to borrowers. This bridges the gap between 
delivery and payment for MSMEs, a gap that can sometimes be up to 90 days. Historically, they have had 
60 clients. Their clients are not able to access bank financing, so their only other alternative would be 
money lenders who charge very high interest rates. NN helped this FI partner through an initial grant at 
start-up by facilitating access to financing and through legal assistance. However, this collaboration stopped 
in 2020. 
The partner’s clients interviewed by the ET were all very satisfied with its service. They noted the limited 
paperwork requirements, absence of collateral requirements, quick delivery time for the loans of less than 
five days after application, and low interest rates, all advantages over other sources of financing. The only 
complaint voiced was that the loan amounts were too small. 
MFI
Another partner is an MFI, which NN helped to facilitate a relationship between the MFI and a poultry 
company to de-risk lending to company’s supplier farmers. Previously, the poultry company would advance
capital to its producers to buy hatchlings and inputs, but this exposed the company to risk and limited its
supply because it lacked capital to finance large numbers of poultry farmers. NN helped set up an 
arrangement whereby the MFI would handle the lending to farmers and be paid by the poultry company 
once the farmers sold their broilers. NN also subsidized the lending to make it less costly for farmers. In 
total, the MFI loaned to about 350 poultry farmers through this arrangement. However, this activity has 
stopped since COVID-19 started in 2020, although NN has continued to provide the MFI with support 
to help it raise more funds.
Loan recipients received enough financing to purchase chicks and inputs and raise them for 45 days before 
selling to the poultry company. Borrowers interviewed found the process simple; all they had to do was 
sign a contract, and, unlike other FIs, no collateral was required, and the money arrived quickly. Most of 
the farmers interviewed had not received agricultural financing before, so they had little basis for 
comparing their experience. Overall, their experiences were positive, except for one respondent who 
encountered a delay in payment.
Commercial Bank
The third FI is a large commercial bank. It was already lending to some MSMEs, albeit through its corporate 
finance department, which was not well set up to service MSMEs. When NN initially contacted it, the FI 
10 Note that as discussed in the findings for EQ1a, one of the four FI partners disputes the nature of its relationship with NN. It 
is omitted from discussion here. Also, NN is in the preliminary stages of partnering with several other FIs.
17
was seeking a loan from the African Development Bank, which required them to have an MSME strategy 
to receive the loan. This formed the basis of their partnership with NN. The activity helped it to draft this 
strategy, which will inform the development of its MSME department. This strategy has not yet been 
implemented as there have been multiple rounds of revisions. In mid-2021, after a long bureaucratic delay, 
the bank hired an embedded advisor paid for by NN to help support its MSME borrowers. Despite the 
delays in finalizing the SME facility and advisor, the SME initiative still has the support of bank management. 
In addition to the MSME strategy and embedded advisor, NN has helped the bank to meet due diligence 
requirements and access loan guarantee funds to de-risk lending for MSMEs. It is through this latter 
assistance that NN has thus far contributed to increasing the bank’s MSME lending, with four MSMEs 
accessing funding to date. At least one of these was referred to the bank after failing to attract investment 
through NN’s Component 2 activities. The one MSME client interviewed for this evaluation said that the 
loan application process was comparatively easy, but the processing time was long. The client found the 
loan terms convenient and flexible during the pandemic but experienced problems with the disbursement 
schedule. Overall, however, the client said it would return for future funding needs. 
4.3.4 Conclusions
FIs, including SACCOs, who participate in NN are interested in improving their services to MSMEs. 
Obviously, those less motivated to service that market segment would be less inclined to participate with 
NN. As described in the background to this section, NN has struggled to form partnerships with 
commercial banks who generally see MSME lending as too risky. Part of this is that NN was not designed 
to bring capital to the table, often a prerequisite for working with commercial banks.
On the capability front, NN made significant progress in improving SACCO capability as evidenced by the 
number of improvements they made after completing NN training. FI partner capabilities have not been 
strengthened through training but, rather, through targeted legal and strategic technical assistance. It is 
unclear how effective this was at improving the capacity of the FIs themselves; nevertheless, this technical 
assistance appears to have been instrumental in implementing programs with FI partners. 
NN has contributed to increasing lending to MSMEs through money it helped its FI partners raise or by 
subsidizing lending in the case of its MFI partner. In the case of its commercial bank partner, it has more 
generally increased the bank’s orientation towards MSMEs through the MSME strategy and embedded 
advisor. Although it is difficult to isolate the impact of NN, the FI partners’ clients interviewed for this 
evaluation expressed satisfaction with the services they received from the partners. These financial 
services fill an underserved market segment (i.e., MSMEs who may lack collateral) and offer more 
convenient services with, for example, less paperwork and a shorter turnaround time. SACCOs, for the 
most part, increased lending and made improvements following NN training, including developing services 
that were more useful for their clients. 
4.4 EQ 3: HOW EFFECTIVE ARE NN’S INVESTMENT FACILITATION EFFORTS
4.4.1 Background
NN’s Component 2 links MSMEs with investors through a four-stage pipeline process outlined in Section 
2. There are three NN investment advisor partners who provide investment readiness support and deal￾making support to selected MSMEs. These investment advisors work throughout East Africa and have 
networks of investors, mostly from Europe, Asia, and North America, who they link to MSMEs. Investors 
are often impact investors, who are willing to accept a lower financial return if they see that their 
investment is having a positive social impact. They are also more patient than typical investors in that they 
are willing to wait for a return on their investment, about twice as long as conventional investors.
The context in which Component 2 was implemented was challenging. For one, Rwanda is a new market 
for MSME investment where MSMEs are generally small and inexperienced with investors and used to 
receiving cheap or free donor capital. Further, COVID-19 introduced significant uncertainty in business 
18
growth projections, increasing risk for investors. Finally, COVID-19 meant that investors were reluctant 
to travel to Rwanda for face-to-face deal-closing meetings.
4.4.2 Finding – Investment Pipeline Is Effective but Slow, and Communication Could 
Be Improved 
Answering EQ 3 requires not only estimating the revenue and employment impact of investments 
facilitated through the pipeline, but it also requires determining how well the pipeline functions at each of 
its four stages: sourcing, selection, investment readiness support, and deal closure.
Sourcing
NN has identified potential MSMEs to enter the investment pipeline, primarily through the four sources 
listed below with an approximation of their share of sourced MSMEs (in parenthesis): 
● Referrals from banks (10 to 15 percent)
● Referrals from other projects (10 to 15 percent)
● Network events including roadshows (50 percent)
● Website (25 percent)
During the evaluation fieldwork, several respondents also said they had learned about the activity from 
friends, a news advertisement, or through pre-existing relationships with one of the investment advisors. 
The website has been abandoned because it was inefficient and was attracting too many startup businesses, 
which NN does not support because of the risk. 
Selection
In selecting businesses, NN looks at whether the applicant is linked to agriculture, has a client base and 
revenue, can provide two years of financial statements, has a capable and driven owner, has a board or 
governance system, possesses market opportunities, and is aligned with NN’s impact objectives. 
Stakeholder feedback, including from those not receiving investment, was positive, with most considering 
the selection process fair and transparent. 
However, some businesses appear to have been selected and then later rejected for criteria that could 
have been identified during selection. For example, some were seeking loan sizes that were too small and, 
therefore, could not continue in the pipeline. Another was later disqualified because it was a nonprofit 
and not a private-sector business. 
Investment Readiness Support
When working on a commercial basis, the investment advisors usually charge the MSME when they close 
a deal. The NN activity, however, pays them to provide investment readiness support to selected MSMEs. 
The deals investment advisors have closed for NN typically are in the range of $100,000 to $1 million. To 
date, NN takes credit for facilitating investment for 27 companies worth over $30 million. In 2019 and 
2020, NN sourced a total of 465 MSMEs, selecting 198 to enter the pipeline. Twenty-one MSMEs closed 
deals with investors and received investments, with some receiving multiple investments following the first 
deal closing.
Once selected, MSMEs reported receiving investment readiness support on the following topics:
● Training on business analysis and financial modeling, which was useful for improving their 
existing business plans.
● Support to structure their business and organize financial documents.
● Strategic and business planning advice, which was helpful to orient their business.
● Help with financial modeling and negotiating a good contract.
● Help to better understand the impact of their products on their customers.
19
● Assistance putting together pitch decks; and pitching and presenting to investors. 
This support was nearly universally regarded as useful by the MSME participants interviewed, even if they 
did not close deals through investors linked to them by NN. In fact, half of those interviewed who did not 
close deals through NN reported later accessing financing independent of support from NN’s investment 
advisor partners. Among those able to gain access to financing elsewhere, the most useful skills they gained 
were bookkeeping, organizing financial statements, confidence and presentation skills, and an appreciation 
of the value of honesty when dealing with an FI or investor. 
SME Participants’ Feedback on the Overall Pipeline Experience
A common complaint from participants was the lack of communication from NN or its investment advisor 
partners. This complaint related to different stages of the pipeline. At the selection stage, some said that 
the next steps were not made clear, or they did not know who the investment advisor was when they 
first contacted the SME. Others never heard back from NN, despite contacting it, and were left assuming 
they would not proceed further in the investment pipeline. 
Another issue cited by respondents was the long duration of the process and the work involved. Among 
respondents, the average reported duration between application and exit from the pipeline was nearly six 
months. For those receiving investments, the average reported period between application and deal 
closure was 10 months. This is consistent with the experience of the investment advisors with other 
clients. Following deal closure, it still took nearly a half-year more on average for MSMEs to receive their 
first disbursement. However, some of this lengthy experience in the pipeline can likely be accounted for 
by COVID-19. 
Several respondents also pointed out that installment plans with the investors had too many installments 
disbursed over time. Disbursement by installments rather than lump sum caused delays in the respondents’ 
work since they needed the full or close-to-the-full amount to purchase equipment. 
Investor Perspectives
The ET spoke with two investors who had very different perspectives on the businesses they had been 
introduced to by NN’s partner investment advisors. One had invested in two businesses through NN and 
was pleased with the quality of the information memorandum, financial model, and management 
presentations, which led them to make the investments. Another, however, had not yet made any 
investments, although it had met with several businesses through NN’s investment advisors. It found 
businesses it was introduced to were looking to make strategic moves and were, therefore, too risky, 
lacked professional documentation, and were unable to understand their own profitability, or they were 
looking for subsidized funding. The former investor was an impact investor, whereas the latter was a 
commercial investor. The difference in expectations for these two types of investors may explain their 
different perceptions of the caliber of NN-supported businesses. 
4.4.3 Finding – Investments Increase Income and Revenue 
The data collected by the ET suggests that NN’s investment facilitation did lead to increased revenue and 
job creation. Table 9 below shows these results, based on data collected through the ET’s fieldwork. The 
table combines jobs created and saved as reported by nine MSMEs interviewed who closed investment 
deals through NN’s investment pipeline. The results show that the $2.2 million in investments included in 
discussion for the fieldwork created or saved 331 jobs. This implies an investment cost per job generated 
of $6,648. Also, respondents reported that the investments led to an average revenue increase of 52 
percent.
Table 9: MSME Job Creation and Revenue Increase Post Investment (n = 9)
Result
Investment amount (USD) 2,200,332 
20
Result
Total jobs created and saved 331 
Investment dollars per job created or saved (USD) 6,648 
Monthly or annual revenue change attributed to investment received (%) 52 
Respondents who received investments were asked to explain how the investment generated jobs or 
increased revenue. Below are some of the most common responses:
● Finance was used to purchase new machinery, which required additional staff to operate or new 
staff to source supply for expanded production. 
● Finance was used for establishing new division/product lines, requiring more staff.
● Finance was used to expand post-harvest processing, which necessitated more staff. 
● Finance was used to expand production, which required hiring more staff.
4.4.4 Conclusions 
The previous findings present evidence that the NN investment facilitation efforts have had a significant 
impact on the employment and revenue at the MSMEs that received investments. On average, a job was 
generated or saved for less than $7,000, and businesses reported substantial average revenue increases 
following investment of around 50 percent. 
It is unsurprising, however, that MSMEs successfully procuring investments would generate employment 
and revenue. What is a more interesting question is whether the investment pipeline effectively connected 
MSMEs with investors and helped them to close deals. The first step is sourcing, where it appears NN 
used a variety of approaches to identify prospective MSMEs. The selection stage appears to have selected 
some MSMEs that could have been screened out during selection. Assuming that these businesses were 
forthright in their applications, this represents inefficiency through unnecessary engagement with 
unqualified applicants. Further, some had a negative impression of NN afterward, which might make them 
less likely to refer more suitable candidates. 
Overall, investment readiness support was well regarded by the MSMEs receiving them. A number who 
did not procure investments were later able to use their improved capacity to access financing elsewhere. 
However, one investor had a negative view of the sophistication and preparedness of businesses to which 
they were introduced. 
The structure of incentives for investment advisor partners is a potential weakness. Investment advisors 
are paid by the activity based on the number of MSMEs to whom they provide investment readiness 
support. This training payment is in addition to the typical deal closure fee paid by the successful investee. 
This is different than standard commercial payment model, in which investment advisors are paid only 
when they help close a deal. NN’s investment advisor payment model then potentially incentivizes 
investment advisors to provide investment readiness support to businesses that may not have realistic 
chances of closing investment deals and to provide lower quality support. To be clear, the ET did not find 
direct evidence that this is the case. The result of these payment design incentives may be more businesses 
passing through the investment pipeline but lower-quality potential investments for investors. On the 
other hand, some of these businesses gained enough capacity to attract other investments, even if not 
with the originally intended investors. In this way, paying per business supported is like the typical non￾governmental organization business training program. 
5.0 RECOMMENDATIONS
5.1 EQ1C – HOW CAN THE EMPLOYMENT MODEL BE IMPROVED?
The high-level strategic recommendation made here is that future activities should focus on measuring 
direct jobs, and doing so accurately, and not be responsible for also measuring indirect and induced job 
21
creation, with the latter two being the responsibility of the Mission. Along these lines, USAID/Rwanda 
should consider creating country-level macro-multiplier models to estimate the indirect and induced 
impacts of their activities and, perhaps more importantly, to help the Mission better target its work by 
sector to best achieve its impact goals. The following provides recommendations for achieving this 
strategic recommendation in the short term (the remainder of NN) and the medium term (subsequent 
activities). 
Short-Term Recommendations
Recommendation – NN should create guidelines for what financing it can take credit for having 
facilitated. These guidelines and criteria should be created in consultation with key FI partners and MSMEs 
which have already received investment, as well as USAID/Rwanda’s monitoring, evaluation and learning 
experts. The guidelines should be oriented by the following simple counterfactual question: “Would the 
finance deployed by a partner FI or finance received by a supported MSME have been deployed or received 
without NN’s support?” They should also be used to audit the past financing for which NN has claimed 
credit. Moreover, NN should keep forms for all financing they take credit for stating that the recipient 
understands and agrees that the financing can be credited to the activity. 
Recommendation – NN should validate reported direct-job creation through follow-up data collection 
with all MSMEs that have received financing justifiably credited to NN. These interviews should be in depth 
and in person and include both reviews of business records and a narrative recounting how the financing 
was used and how that use created jobs. Because of its relationship with these MSMEs, NN and its partners 
are better positioned than an external consultant to get the time and access needed from MSMEs to do a 
thorough assessment of actual job creation. 
These in-depth validation interviews should also include a component to assess jobs saved due to financing 
justifiably credited to NN. This is a significant but overlooked portion of NN’s impact, especially during 
the COVID-19 pandemic, and NN and USAID should quantify this impact and take credit for it. 
Medium-Term Recommendations
Recommendation – Future activities should be encouraged to focus on measurement of direct jobs 
created. Where possible, direct jobs should be measured directly rather than projected through a model. 
That is, they should be measured by collecting data from beneficiaries on job creation each year after 
receiving the financing until the end of the activity. In some cases, such as clients of FI partners that the 
activity lacks access to, a simplified micro model can be used. If desired, a simple multiplier can be taken 
from secondary sources to provide an estimate of how many indirect or induced jobs are created with 
each direct job created. For example, the model reviewed here shows that, on average, 0.95 indirect plus
induced jobs are created for each direct job created. Given the sensitivity of the full model to sector 
specification, such a simple economy-wide multiplier should be sufficient to provide a general indication 
of an activity’s broader impacts. This should not be a required part of activity reporting, however.
Recommendation – USAID/Rwanda should consider developing and maintaining its own multiplier 
model. The model could be used to model employment changes and other economic outcomes, such as 
income. The model could also be used to estimate the indirect and induced impacts of its portfolio and, 
perhaps more importantly, to target interventions by sector to have the broadest impact. Creating this 
model should draw on NN’s experience, and a detailed user manual should also be created to ensure that 
the model can be understood and updated even amid staff turnover. 
One advantage for USAID/Rwanda of implementing a multiplier model is the opportunity to influence the 
GOR to create more useful and clear sector categories for its macro-economic data. This has large 
implications for the accuracy of a model. It also has implications for the ability to strategically target 
investments by sector to maximize impact. Should USAID consider developing a multiplier model, it should 
seek to create an expanded sector classification schematic with precise definitions for each.
22
5.2 EQ2 – FINANCIAL INSTITUTIONS
As with the previous recommendations, these are divided into short-term recommendations for NN and 
medium-term recommendations for USAID/Rwanda. At this stage, the ability of the activity to make 
significant changes to this component looks limited. Partnerships with FIs have taken a long time to 
materialize, and the activity’s design limits its ability to bring significant capital to the table and incentivize 
partnerships with larger FIs. 
Short-Term Recommendations
Recommendation – NN’s training delivered through AMIR was widely found to be successful. The 
activity intends to expand the number of SACCOs trained and continue training until activity closure in 
2023. The activity should work with AMIR to develop and disseminate copies of the training courses to 
all SACCOs. These could be hard copies or online, low-bandwidth, interactive trainings. In addition to 
ensuring that the well-regarded training will be available to all SACCO staff beyond the activity period, 
this recommendation also addresses two complaints about the training. It would overcome bandwidth 
limitations for certain branches, and it would also be flexible by allowing staff to complete it on their own 
time rather than miss parts due to business interruptions.
While supporting SACCOs and MFIs is useful, they mostly service micro-enterprises and are unlikely to 
generate significant growth and job creation. Commercial banks have slow-moving bureaucracies and a 
limited appetite for the risk inherent in SME lending, preferring corporate financing. Instead of focusing on 
these two segments of the financial services marketplace, USAID/Rwanda should focus on growing a third 
segment of innovative financial intuitions that specialize in serving the “missing middle,” small and medium 
enterprises (SMEs). Factoring is one financial service suitable for SMEs, and NN’s factoring partner was 
extremely well regarded by its clients. Other innovative financial services will need to be brought to 
market to meet the financing needs for Rwanda’s SMEs. 
Medium-Term Recommendations
Recommendation – USAID/Rwanda should consider implementing an MSME finance innovation 
challenge fund activity. This activity would provide winners with initial capital as well as fundraising, legal, 
and technical assistance. Winners would be selected based on the promise of their model for providing 
financial services to MSMEs and their capacity for implementing the model. Existing FIs, including 
commercial banks, should be eligible. 
5.3 EQ3 – INVESTMENT PIPELINE
These recommendations include one short-term recommendation for the remaining NN implementation 
period and one medium-term recommendation for USAID/Rwanda to consider for subsequent 
programming. 
Short-Term Recommendations
Recommendation – NN should investigate and potentially address the concerns about the pipeline 
identified through interviews with MSMEs. These include issues with selection, e.g., 
● selected applicants are later rejected by criteria that could have been identified earlier in the 
process; 
● poor communication between investment advisors and MSMEs; 
● and deals closed where the structure is not well suited for the purposes of the investment. In 
addition, this simple short-term recommendation could expand the impact of NN’s positive work 
with the investment pipeline at low cost. 
Recommendation – MSMEs not completing the investment pipeline should be given templates for 
documents used to attract investments. As part of investment readiness support, MSMEs receive 
23
assistance to create a range of documents needed to attract financing. These include financial models and 
pitch decks, which the investment advisors help to produce based on templates. Having access to these 
templates could allow MSMEs to better attract investment elsewhere. Ideally, these would be posted 
online, but failing that, they could be distributed to MSMEs on cheap pen drives at the last meeting with 
their investment advisors.
Medium-Term Recommendations
Recommendation – activity designs with investment advisors in the future should feature modified 
commercial payment structures. Rather than provide a per-MSME fee for investment advisors to provide 
investment readiness and deal closure support, an activity should add a percentage bonus to deal closing 
fees. This would maintain the original commercial incentives for investment advisors to successfully close 
as many deals as possible undistracted by the incentive to train MSMEs who might not have much 
investment potential. Capacity for MSMEs not closing deals could be increased at low cost by providing 
investment form templates and guidance through a website. 
Recommendation – USAID/Rwanda should design MSME investment activities intended to better 
incentivize attracting commercial investments in potential high-growth MSMEs. This would be in addition 
to the lower-risk impact investments, which were most common with NN. Rwanda is a new market for 
investors, which means that many MSMEs are not ready for impact investors, let alone commercial 
investors. However, it also means there likely exist some uninvested in high-upside investment 
opportunities. These high-upside investment opportunities would have the potential to generate large 
impacts, such as job creation, for the country as well as returns for the investor. One can imagine one or 
two of these higher-upside investments being successful and outperforming the rest of an activity’s 
investment portfolio. On the other hand, they could all fail, and the activity would have nothing to show 
for its time. Future activity design should make space for some risk taking by setting aside some portion 
of resources to attempt to facilitate investment in high-growth-potential MSMEs. Performance indicator 
targets and USAID/Rwanda expectations should also reflect the likelihood that this portion of the activity
may very well produce no results.
24
6.0 ANNEXES
25
ANNEX 1: PEEL TASK ORDER EXPRESSION OF INTEREST (EOI) – PERFORMANCE 
EVALUATION
PEEL TASK ORDER
EXPRESSION OF INTEREST – PERFORMANCE EVALUATION
I. BACKGROUND INFORMATION
A) Identifying Information 
1. Project/activity Title: Ngurisa Nshore 
2. Award Number: Contract/720-696-18-C00001 
3. Award Dates: 4/24/2018 – 4/23/2023 
4. Project/activity Funding: $14,946,186
5. Implementing Organization(s): DAI Global LLC
6. Project/activity COR/AOR: Fina Kayisanabo
B) Development Context
1. Problem or opportunity addressed by the project/activity being evaluated
The purpose of Nguriza Nshore (“lend so I may invest”) (NN) is to foster rural economic growth by 
facilitating the emergence of a dynamic small- and medium-enterprise (SME) and agribusiness sector. By 
alleviating constraints to investment and increasing access to finance, Nguriza Nshore will accelerate the 
creation and growth of off-farm SMEs as entry points for widespread economic growth and social 
development—increasing productive employment for rural populations, creating opportunities for 
entrepreneurs, and reducing poverty throughout the country.
2. Intended results of the project/activity being evaluated
Nguriza Nshore (NN) facilitates finance and investment (debt and equity) for SMEs that are not yet 
realizing the benefits of Rwanda’s strong economic performance—contributing to the country’s economic 
growth, social development, and job creation. By working with financial institutions, investors, consulting 
firms, transaction advisors, and a variety of government ministries, NN functions as a facilitative catalyst 
that strengthens existing and emerging public-private platforms by supporting investment—leading to a 
more efficient financial services ecosystem. Nguriza Nshore fosters rural economic growth, social 
development, and employment creation by facilitating the emergence of a dynamic agribusiness sector in 
Rwanda. Activities focus on three key components as described below.
26
Component 1 – Expand Capacity of Financial Institutions (FIs) to Lend to SMEs
To achieve its objectives, NN’s Component 1 is: 
● Oriented toward shifting the competitive landscape—to push banks and non-bank lending 
institutions toward customer-oriented growth because Financial Institutions (FIs) typically design 
products best suited for themselves rather than their potential clientele. Hence, their growth is 
happenstance and not customer oriented. Institutionally, they do not understand the needs of 
their clientele and, therefore, they do not design and customize products for them. Nguriza 
Nshore will support FIs to understand client needs and customize products for them. As part of 
this process, NN will design de-risking mechanisms to reach SMEs. 
● Focused on top-down change (i.e., corporate to SME), which would come from customer￾oriented growth competitive pressures pushing banks to constantly seek new and emerging 
market segments, including bottom of the pyramid. Many, if not all, banks are focused on 
corporate clientele—not SMEs. Nguriza Nshore will help banks separate the two functions into 
separate departments with associated policies, procedures, and de-risking mechanisms. 
● Geared toward encouraging banks to compete for the bottom of the pyramid (entrepreneurs, 
micro-entrepreneurs, and SMEs) via a full range of tailored products—targeting SMEs, 
entrepreneurs, and households—that include both savings and lending-based products. 
● Intentional about amplifying bottom-up change (solidarity groups/Micro, Small, and Medium 
Enterprises [MSMEs] to larger SMEs) which would come from customer-oriented growth￾competitive pressures pushing/pressurizing non-bank lending organizations to expand their service 
offers, tailor products, and evolve toward larger, growth-oriented SMEs. 
● As a result, Nguriza Nshore will facilitate change at the top by bifurcating corporate from SMEs; 
creating customized products and services for SMEs; and, at the bottom, by focusing directly on 
micro-entrepreneurs with banks and non-banks. 
Component 2 – Increase Private and Commercial Investment in SMEs
Nguriza Nshore works with Rwandan SMEs to attract both local and international investors in five stages, 
ranging from initial meetings with SMEs to eventually closing investments. The five stages include: sourcing, 
due diligence and selection, contracting and co-creation, investment readiness support, and deal making. 
These stages are not necessarily linear, and each SME moves at its own pace. 
To achieve its objectives, Nguriza Nshore’s Component 2 is: 
27
● Oriented toward shifting the competitive landscape to favor agribusinesses and related SMEs that 
commit to a customer value-add-oriented growth strategy, including firms that focus on mass 
markets/bottom-of-the-pyramid customer segments. In other words, NN will focus a large part 
of its efforts on identifying and supporting SMEs that cater their products and services to ordinary 
Rwandan citizens—since they are the bulk of the population and represent the largest growth 
opportunities. The activity’s efforts will also focus on companies that offer employment 
opportunities for ordinary citizens and economically disadvantaged populations as the businesses 
grow. These prospects are best represented in agriculture and agriculture value chains where 
most of the population works. Supporting SMEs in these two areas will improve the lives, 
livelihoods, and resiliency of people who need it most. 
● Focused on shifting the balance of larger start-up and growth capital services away from bank and 
microfinance and toward risk-sharing/venture capital mechanisms since private capital is a new 
concept in Rwanda and an excellent source of growth investment. 
● Innovative in encouraging the emergence of tailored risk-mitigating investment-banking 
products—including derivatives, revenue sharing models, and impact investments. Nguriza Nshore 
will explore more opportunities to support companies that require Monitoring, Evaluation, and 
Learning (MEL) assistance—because impact investors require grassroots-level information, and 
companies are either unable to or not geared toward designing appropriate systems and collecting 
data. 
● Geared toward facilitating the emergence of more robust exit options that could link to existing 
regional exchanges, private placement services/matchmakers, and mergers/acquisitions since 
investors consider these options as important de-risking mechanisms.
Component 3 – Strengthen the Enabling Environment for the Expansion of SME Agribusinesses 
and Support Firms
To achieve its objectives, NN’s Component 3 is:
● Oriented toward helping the political system to become more constituent-oriented where 
ordinary citizens have a voice in policy development and implementation oversight—since the 
process is generally top-down and driven and managed in Kigali by external consultants. Post 
approval, policies tend to sit on the shelf and are not effectively implemented, monitored, and 
reported on. Specific changes in the functions that NN will advocate for include policy 
development, policy translation, public education, implementation, and monitoring. 
● Focused on catalyzing a set of support services, including civil society, that is funded locally and 
constituent driven—such as research firms and public relations firms and media. 
By the end of the activity, it is expected that NN will have achieved the following: 
activity-Level
● $15M in loans accessed due to United States Government (USG) interventions 
● 30,000 jobs created with USG assistance
● $30M in new USG commitment and private sector investment leveraged 
Component 1: Expand Capacity of Financial Institutions (FIs) to Lend to SMEs
● $5.9M in loans leveraged using de-risking tools 
● One hundred ninety-eight (198) financial and non-FIs receiving USG assistance that extended loans 
and other financial products to SMEs 
Component 2: Increase Private and Commercial Investment in SMEs
● Twenty (20) percent increase in value of annual sales of farms and firms receiving USG assistance
● 500 firms using USG-funded technical assistance to improve business performance 
28
Component 3: Strengthen the Enabling Environment for the Expansion of SME 
Agribusinesses and Support Firms
● Fifty-four (54) milestones in improved institutional architecture for food security policy achieved 
with USG support
Additional target information includes: 
Table 1: activity Objectives and Targets
Indicator FY20 
Actuals
FY20 
Target
Percent 
Achieved 
FY20
Life of 
activity
Actuals
Life of 
activity % 
Achieved
Life of 
activity
Target
Q1 – Q4 FY20 % FY18-FY20 % Value
1. EG.3.2-27 Value of 
agriculture-related 
financing accessed due 
to USG assistance 
[IM-level] 
$21,193,878 $3,600,000 589% $22,446,839 149% $15,000,000
2. Number of jobs 
created with USG 
assistance (Custom)
$14,645 $6,558 223% $16,572 55% $30,000
3. EG.3.1-14 Value of 
new USG 
commitments and 
private-sector 
investment leveraged 
by the USG to 
support food security 
and nutrition [IM￾level] 
$7,939,132 $6,600,000 120% $9,714,443 33% $30,000,000
3. Target areas and groups
Nguriza Nshore’s geographic focus prioritizes the 13 districts of the new Feed the Future (FtF) Zones of 
Influence (ZOI). However, to leverage economies of scale in urban/peri-urban locations and expedite 
finance/investment, NN pursues opportunities outside of these priority districts to create a successful 
track record of pulling in new financiers and investors. 
Target beneficiaries include off-farm SMEs, FIs (commercial banks, microfinance), investors, consulting 
firms, transaction advisors, and a variety of government ministries.
4. Approach and implementation
Nguriza Nshore is implemented through a consortium of partners. Consortium members consist of DAI 
Global, LLC; and BiD Network. The consortium members’ roles and responsibilities include:
● DAI Global, LLC – Provide overall vision, leadership, and management of the activity in 
collaboration with the Government of Rwanda (GOR) and USAID/Rwanda.
● BiD Network – Act as the key investment partner—sourcing and assessing SME opportunities 
through a standardized pre-investment readiness process. 
The consortium members foster key partnerships within the following areas: 
Partner #1 – Banks and non-banks will act as key middlemen that deliver financial products and services 
to SMEs. The activity will provide technical assistance and consulting to this industry to bolster their know￾how and capacities.
Partner #2 – SMEs will be direct beneficiaries of NN technical assistance and consulting support.
29
Partner #3 – Transaction advisors/facilitators and business service providers will support both banks and 
non-banks and SMEs by identifying opportunities, structuring transactions, and providing technical 
assistance and consulting services. 
Partner #4 – GOR, donors, donor-funded activities, and industry groups will serve as partners to share 
the resources and expertise necessary to coordinate on a size and scale not possible when working alone 
to implement initiatives.
C) Existing Data
Nguriza Nshore can provide the following documents:
Table 2. Documents for Nguiriza Nshore Midterm Review (MTR) 2017–2019
Name of Document Author Year of Publication
Project Contract + modifications DAI/USAID 2018
Monitoring, Evaluation and 
Learning Plan
DAI/NN 2018 (Y1), 2018 (Y2), 2019 (Y3)
Work Plans
Year 1 Work Plan DAI/NN 2018
Year 2 Work Plan DAI/NN 2018
Year 3 Work Plan DAI/NN 2019
Annual Performance Reports
Year 1 Annual DAI/NN 2018
Year 2 Annual DAI/NN 2019
Quarterly Reports11
Year 1: Q1, Q2, Q3 DAI/NN 2018
Year 2: Q1, Q2, Q3 DAI/NN 2018
Year 3: Q1, Q2, Q3 DAI/NN 2018
Other Documents
Bi-weekly Reports to COR All 2018–2020
Minutes for the Steering 
Committee Meetings (Y1 and Y2)
All 2018–2020
MOUs (Y1–Y3) All 2018–2020
Contract Modifications USAID 2018–2020
In addition to these documents, the following documents from GOR would be helpful for an evaluation 
team. These include:
● Economic Recovery Fund (ERF)
● Entrepreneurship Development Policy (EDP)
II. EVALUATION RATIONALE
A) Evaluation Purpose
This is a midterm performance evaluation. The purpose of this evaluation is to:
● Assess progress made to date in achieving the activity’s components, which are to: Expand 
Capacity of Financial Institutions to Lend to SMEs; Increase Private and Commercial Investment 
in SMEs; and Strengthen the Enabling Environment for the Expansion of SME Agribusinesses and 
Support Firms.
● Assess the accuracy of the NN job multiplier model. 
11 Note: Q4 reports will be the annual reports.
30
● Assist USAID in designing future access to finance activities by providing an evidence base on the 
efficacy of NN approaches and interventions, so that future work can build on successful 
approaches and pivot away from unsuccessful ones.
B) Audience and Intended Uses
The primary audience for this evaluation is USAID/Rwanda. The findings will be used by USAID/Rwanda 
to prioritize interventions and effective approaches to be used in future activity design work. Given that 
this performance evaluation is occurring after the midterm point in NN activity implementation and the 
activity is already shifting significantly to adapt to COVID-19 and other contextual changes, this evaluation 
is not meant to be an evaluation of Contractor performance resulting in changes to the activity
implementation in the final years of the project. Rather, it will be used to test the assumptions and 
approaches of NN’s design and implementation to effectively inform future USAID/Rwanda programming. 
C) Evaluation Questions 
The evaluation will respond to the following questions and provide recommendations, where appropriate. 
It will assess implementation from activity initiation to the present time and (1) develop conclusions based 
on findings, (2) identify opportunities and challenges, and (3) make recommendations for improving future 
programming.
The evaluation team may submit up to two additional evaluation questions for consideration, and the final 
consolidated list of five questions will be decided between USAID and the successful bidder.
1. Employment creation is a main goal of NN and other ongoing and future USAID/Rwanda 
Activities, because Rwanda needs to add 200,000+ jobs per year to successfully integrate new 
market entrants. Nguriza Nshore uses a jobs multiplier—which estimates the direct, indirect, and 
induced jobs created when a business secures a loan or investment. The contractor should 
propose a comprehensive methodology (using a sample of NN partner firms) to assess the 
accuracy of the multiplier model that NN is using. This portion of the scope will ultimately answer 
the following inter-related questions:
A. How closely do the numbers of increased direct jobs resulting from reported NN 
interventions using the multiplier model match the actual number of new, direct jobs created 
at NN-supported enterprises after receiving a new loan or investment by the end of FY 2020 
(using interviews and employment data from supported firms)?
B. How closely do the numbers of increased indirect jobs resulting from reported NN 
interventions using the multiplier model match the number of actual indirect jobs in the field 
(using outcome harvesting, tracing, and interviews with transporters, buyers, and other parties 
involved within the supply chain of supported firms)?
C. Are the methodology’s assumptions related to induced jobs within the multiplier 
reasonable and reflective of the Rwandan economic context? (Bidders are encouraged to 
propose convincing approaches to measuring the accuracy of the induced job section of the 
multiplier.) 
D. Based on the answers to Questions 1A, 1B, and 1C, how can the model or its use be 
improved to increase the accuracy and utility of activity-reported data?
2. To what extent have partner FIs supported by NN been interested in and capable of increasing 
or improving lending tools targeted at SMEs? Has NN’s support resulted in an increase of the 
proportion of lending value to SMEs compared to large business lending for the partner financial 
institutions? 
31
3. Nguriza Nshore supported SMEs to attract local and international investors via technical 
assistance at five stages: sourcing, due diligence and selection, contracting and co-creation, 
investment readiness support, and deal making. Is there any difference in outcomes (such as 
disbursed funds, increased sales, new jobs created, more socially and environmentally friendly 
business management approaches, etc.) for businesses supported at different stages and, if so, what 
is the optimal level/stage to facilitate investment in SMEs?
III. TIMEFRAME AND TRAVEL
A) Timeframe
The Period of Performance for this evaluation is approximately six (6) months after approval of the work 
assignment. 
B) Travel
The evaluation team implementing this evaluation is required to travel to Kigali and outside of Kigali—
including a sample of the 13 districts, as needed, based on a final list of subjects for Key Informant 
Interviews (KIIs), Focus Group Discussions (FGDs), and other methods of inquiry as determined in the 
Evaluation Design stage. The Evaluation Contractor is responsible for the administrative support and 
logistics required to fulfill this task; this includes all travel arrangements; appointment scheduling; 
secretarial services; report preparation services; and printing, duplicating, and translation services. USAID 
will assist the Contractor in obtaining any additional program documents and contacts necessary to fulfill 
the task. 
IV. DELIVERABLES & DESIGN
A) Deliverables
The following deliverables are required. All written deliverables must be in English and submitted in 
electronic format. The Evaluation Contractor must submit all draft documents to the Contracting Officer’s 
Representative (COR) at USAID/Washington, D.C., and the USAID/Rwanda evaluation activity manager. 
The activity manager will provide comments as noted below. The Evaluation Contractor must incorporate 
these comments and submit the final deliverables to USAID. The Evaluation Contractor must submit an 
ADS 508-compliant electronic copy to the Development Experience Clearinghouse at http://dec.usaid.gov
and any datasets to the Data Development Library (DDL) at www.usaid.gov/data. 
Evaluation Work Plan
Within 15 business days of the award of the activity, the lead evaluator shall complete and present a draft 
work plan for the evaluation to the activity manager. 
The work plan will include: 
1. Draft schedule and logistical arrangements.
2. Members of the evaluation team, delineated by roles and responsibilities.
3. Evaluation milestones.
4. Anticipated schedule of evaluation team data-collection efforts.
5. Locations and dates for piloting data collection efforts, if applicable.
6. Proposed evaluation methodology, including selection criteria for comparison groups, if applicable.
7. Evaluation Report outline (if different from the attached template). 
USAID offices and relevant stakeholders are asked to take up to five business days to review and 
consolidate comments through the activity manager. Once the evaluation team receives the consolidated 
comments on the work plan, they are expected to return a revised evaluation design and work plan within 
three business days.
32
33
Evaluation Design and Protocol 
Within 25 business days of approval of the work assignment, the evaluation team must submit an 
evaluation design to the activity manager. The design will become an annex to the evaluation report. 
The evaluation design will include: 
1. Detailed evaluation design matrix that links the Evaluation Questions from the Statement of Work 
(SOW) (in their finalized form) to data sources, data collection methods, and the data analysis 
plan. 
2. Draft questionnaires and other data collection instruments, or their main features.
3. List of potential interviewees and sites to be visited and proposed selection criteria and/or 
sampling plan (must include sampling methodology and methods, including a justification of sample 
size and any applicable calculations) by district.
4. Limitations to the evaluation design.
5. Dissemination Plan (designed in collaboration with USAID).
Unless exempted from doing so by the NN COR, the evaluation design will be shared with partner country 
stakeholders, as well as the implementing partners, for comment before being finalized.
The data analysis plan should clearly describe the evaluation team’s approach for analyzing quantitative 
and qualitative data (as applicable)—including proposed sample sizes, specific data analysis tools, and any 
software proposed to be used, with an explanation of how/why these selections will be useful in answering 
the evaluation questions for this task. Qualitative data should be coded as part of the analysis approach, 
and any code books used should be included in the appendix of the final report. Gender, geographic, and 
role (beneficiary, implementer, government official, Non-Governmental Organization [NGO], etc.) 
disaggregation must be included in the data analysis where applicable. 
All dissemination plans should be developed with USAID and include information on audiences, activities, 
and deliverables—including any data visualizations, multimedia products, or events to help communicate 
evaluations (findings/conclusions/recommendations). See the Evaluation Toolkit for guidance on 
Developing an Evaluation Dissemination Plan.
If applicable based on the Disclosure of Conflict of Interests Forms submitted with the awardee’s proposal, 
the evaluation design will include a Conflict of Interest (COI) mitigation plan. 
USAID offices and relevant stakeholders are asked to take up to 10 business days to review and 
consolidate comments through the activity manager. Once the evaluation team receives the consolidated 
comments on the initial evaluation design, they are expected to return a revised evaluation design and 
work plan within seven business days.
In-Briefing
If proposing in-country travel, within 10 business days of arrival in Rwanda (or five business days after the 
start of Evaluation Work Plan implementation for remote work), the evaluation team will meet with the 
USAID/Rwanda Economic Growth and Program offices for introductions and to discuss the team’s 
understanding of the assignment, initial assumptions, evaluation questions, methodology, and work plan, 
and/or to adjust the SOW, if necessary.
Draft Report
The draft evaluation report should be consistent with the guidance provided. The report will address each 
of the evaluation questions identified in the SOW and any other issues. Any such issues can be included 
in the report only after consultation with USAID. The submission date for the draft evaluation report will 
be determined in the evaluation work plan. Once the initial draft evaluation report is submitted, USAID 
will have 10 business days in which to review and comment on the initial draft, after which point the 
34
activity manager will submit the consolidated comments to the evaluation team. The evaluation team will 
then be asked to submit a revised final draft report within 10 business days.
This performance evaluation will comply with USAID evaluation policy requirements for evaluations, 
which is available in Annex A of the Evaluation Policy and at the following link: 
http://www.usaid.gov/sites/default/files/documents/2151/USAIDEvaluationPolicy.pdf. 
The format of the Performance Evaluation Report must include:
● Cover page using the standard Feed the Future Branding Template.
● Table of Contents.
● Acronyms.
● List of tables or charts (if any).
● Executive summary (not to exceed two pages).
● Background/Introduction (not to exceed five pages).
o Description of the activity including goals and objectives.
o Evaluation rationale/purpose/objectives, including a list of the main evaluation questions.
o Description of the evaluation design, methodology, and limitations.
● Findings – Describe the findings related to each of the evaluation questions, disaggregated 
geographically at the district level where applicable.
● Conclusions. 
● Recommendations – Based on the evaluation purpose and the findings, the evaluation team should 
present recommendations. Some examples of potential recommendations include (1) specifying 
what changes can be made in future program design or implementation to result in more effective 
and/or efficient execution and improved results, (2) identifying potential new solutions to 
problems the activity has faced, (3) identifying adjustments/corrections that need to be made to 
address any identified risks to the sustainability of improved resources, management practices and 
technologies, and market information, (4) outlining actions and/or decisions to be taken by 
management, and (5) communicating any additional lessons learned for program implementation, 
coordination, and beneficiary satisfaction.
● Annexes 
o Statement of Work
o Sources of information and data, identified and listed. Include a list of places and types of 
people and groups interviewed (e.g., “female beneficiaries in Village X,” “Senior Nguriza 
Nshore Staff Member,” etc.).
o Copies of all tools, survey instruments, and questionnaires used.
o Copies of background documents that were used by Contractor but which were not provided 
by USAID/Rwanda.
o Any unaddressed comments submitted by the NN activity team (if applicable).
o Summary information about evaluation team members—including qualifications, experience, 
and role on the team.
o Signed disclosure of COI forms for all evaluation team members, either attesting to a lack of 
COI or describing existing COIs.
o Statements of difference regarding significant unresolved differences of opinion by funders, 
implementers, and/or members of the evaluation team, if applicable.
(Note: The final report must not exceed 50 pages, excluding the cover page, table of contents and 
annexes.) 
See the Evaluation Toolkit for the How-To Note on Preparing Evaluation Reports and ADS 201mah, 
USAID Evaluation Report Requirements. An optional Evaluation Report Template is also available in the 
Evaluation Toolkit. 
35
Final Evaluation Report
The evaluation team will be asked to take no more than 15 business days (or as agreed upon in the work
plan) to respond to and incorporate final draft evaluation report comments from USAID. The evaluation 
team lead will then submit the final report to the activity manager. The final report must be ADS 508 
compliant.
Oral Presentation on Preliminary Results 
After the submission of the Final Report, the Evaluation Contractor will deliver a presentation (verbal and 
with PowerPoint) on results, key findings, and recommendations. The audience will include 
USAID/Rwanda and activity stakeholders. The presentation will:
● Review and assess NN performance based on the questions listed above. The core of the 
presentation should be aligned with the evaluation questions.
● Provide the methodology and approach used to evaluate the activity. 
● Provide analysis, findings, conclusions, and recommendations.
Submission of Dataset(s) to the Development Data Library
Per USAID’s Open Data policy (see ADS 579, USAID Development Data) the contractor must also submit 
to the COR and the Development Data Library (DDL), at www.usaid.gov/data, in a machine-readable, 
non-proprietary format, a copy of any dataset created or obtained in performance of this award, if 
applicable. The dataset should be organized, cleaned, and documented for use by those not fully familiar 
with the intervention or evaluation.
Please review ADS 579.3.2.2 Types of Data To Be Submitted to the DDL to determine applicability.
Submission of Final Evaluation Report to the Development Experience Clearinghouse
Per USAID policy (ADS 201.3.5.18), the contractor must submit the evaluation final report and its 
summary or summaries to the Development Experience Clearinghouse (DEC) within three months of 
final approval by USAID. 
B) Evaluation Design
The evaluation team must propose its own methodology, which must be approved by the USAID/Rwanda 
activity manager. At a minimum:
● The evaluation must follow the principles and guidelines for high-quality evaluations outlined in 
ADS 201.3.6.
● The evaluation team should consider a range of possible methods and approaches for collecting 
and analyzing information and data that are required to answer the evaluation questions.
● There must be a thorough review of key activity documents. USAID/Rwanda will provide a 
detailed document review matrix that lists relevant documents—including contracts, annual work 
plan, MEL plan, and progress reports; as well as other information available regarding previous 
similar activities and other initiatives—to provide a foundation for the evaluation work to be 
undertaken in the field. A notional document review matrix is provided in Annex A.
● The evaluation team will conduct key informant interviews with activity staff, sub-partner 
organizations, government officials, private sector, beneficiaries, and other stakeholders. As a 
result, it is expected that the evaluation team will work with a local organization to conduct 
beneficiary focus group discussions and other proposed surveys. We provide an initial list of key 
stakeholders for key informant interviews that should be considered in Annex B. 
● The evaluation team will conduct focus group discussions with GOR officials and program 
beneficiaries, which should include both men and women. 
● The evaluation team will analyze NN data generated by the DAI Spell Out (M&E) teams.
36
● The evaluation team will analyze relevant indicators and data from USAID’s annual Performance 
Plan and Report (PPR) and activity MEL plan.
● The reporting of findings should be broken down by district. (Note: There is no expectation that 
the findings will be statistically representative of the district.)
● The evaluation team will Integrate USAID’s Collaboration, Learning, and Adapting (CLA) approach 
to enhance the engagement of stakeholders and usefulness of the recommendations for current 
and future programming.
The evaluation team, in collaboration with USAID, will finalize the evaluation methods before fieldwork 
begins. The contractor will submit the preliminary evaluation design in response to the SOW for review 
by USAID/Rwanda. The evaluation COR will approve the finalized evaluation design two weeks or more 
before the team’s arrival in the country at the recommendation of the USAID/Rwanda activity manager. 
V. TEAM COMPOSITION
Describe the intended size of the evaluation team, the roles and responsibilities of team members, specific 
qualifications that team members are expected to possess, and the manner that their qualifications will 
contribute towards the evaluation effort.
At a minimum, the activity team shall include the key personnel positions indicated in the following chart. 
Given the centrality of NN’s jobs multiplier model to the evaluation scope of work, one or more team 
members should have experience evaluating similar large-scale economic tools. It is also expected that the 
evaluation team will work with a local consultant or organization to help translate, conduct, and analyze 
the quantitative and qualitative data collected. Proposed local support roles include M&E, access to finance, 
and logistics support. 
The USAID/Rwanda team reserves the right to approve or deny individual team members based on their 
CV and reference checks. The USAID/Rwanda team is also open to alternative key personnel proposals 
from respondents if they feel that adjustments to the list should be made.
Table 3. Caption
Key Position Responsibilities
Senior Evaluation 
Specialist
The Senior Evaluation Specialist will serve as the lead and will have overall 
responsibility for the evaluation’s quality and timeliness—including design, such as 
the protocol and questionnaire finalization; preparation, including various 
procedural, managerial, and training elements; direction; and oversight of the 
evaluation’s implementation, analysis, and report writing. The Senior Evaluation 
Specialist will serve as the primary Point of Contact (POC) with the USAID 
Mission, host country government, and the subcontractor, as applicable. The 
Senior Evaluation Specialist also will be the main POC on evaluation progress, 
quality, and adherence to budget. The Senior Evaluation Specialist will provide 
training to the survey subcontractor, if applicable, on translation, questionnaire 
pretesting, and listing procedures. The Senior Evaluation Specialist will also be 
responsible for the evaluation report, including writing many sections of the 
report. 
The Senior Evaluation Specialist should have a Master of Arts (MA) in social 
sciences with at least 12 years of practical experience conducting evaluations, 
preferably in international settings. S/he should have demonstrated skills in 
qualitative or quantitative evaluation methods in the agriculture or food security 
sector with preferable experience in market systems development. S/he must also 
have demonstrated experience managing evaluation teams.
37
Key Position Responsibilities
Evaluation Specialist
The Evaluation Specialist will provide high-level technical support for a broad 
range of evaluation activities—including evaluation design, such as the protocol 
questionnaire and finalization; implementation of fieldwork; cleaning and analysis 
of data; and report writing. The Evaluation Specialist will provide quality control 
for all survey deliverables, oversee data analysis, and help coordinate the 
development of the report. 
The Evaluation Specialist should have an MA in social sciences with at least eight 
years of practical experience conducting evaluations, preferably in international 
settings. S/he should have demonstrated skills in qualitative and quantitative 
evaluation methods. S/he must also have experience in the agriculture and/or 
economic growth sectors, with preferable experience working in market systems.
Research Assistant
The Research Assistant will support the Senior Evaluation Specialist and 
Evaluation Specialist with their respective tasks. The Research Assistant will 
report to the Senior Evaluation Specialist.
The Research Assistant should have an MA in social sciences with at least four 
years of practical experience conducting evaluations in international and 
agriculture settings, with preferable experience working in market systems. S/he 
should have demonstrated organizational and communication skills.
Financial Sector 
Specialist
In addition to the evaluation specialists, a financial sector specialist is required 
who can fully understand the NN activity mandate, approaches, and results. The 
Financial Sector Specialist will support the Evaluation Specialists in coming up with 
tools for KIIs, FGDs, and other methodologies that account for financial-sector 
terms and approaches. The Financial Sector Specialist will also be a key participant 
in all meetings with financial organizations and stakeholders.
The Financial Sector Specialist should have a Master of Business Administration 
(MBA) or related advanced degree with least eight years of practical experience
conducting and evaluating access to finance interventions in the international 
development context. 
VI. SUGGESTED LEVEL OF EFFORT (LOE)
The suggested LOE can be found in Table 4 below. USAID/Rwanda is open to proposed adjustments to 
this LOE as required by the proposed methods and approaches.
Table 4. Suggested Level of Effort (LOE)
Task/
Deliverable
Sr. 
Evaluation 
Specialist
Evaluation 
Specialist
Research 
Assistant
Financial 
Sector 
Specialist
Local 
M&E 
Specialist
Local 
A2F 
Specialist
Local 
Logistics 
Support
Conference 
Call/ Desk 
Review
5 5 5 5 0 0 0 
Evaluation 
Planning & 
Design
15 15 15 15 15 15 10
Data 
Collection 
(to include 
any training)
17 12 17 15 12 12 10
Data 
Cleaning & 
Analysis
0 10 10 5 10 5 0 
38
Task/
Deliverable
Sr. 
Evaluation 
Specialist
Evaluation 
Specialist
Research 
Assistant
Financial 
Sector 
Specialist
Local 
M&E 
Specialist
Local 
A2F 
Specialist
Local 
Logistics 
Support
Presentation 
of findings 3 3 3 1 1 1 0 
Draft Report 15 10 10 5 3 3 0 
Revisions of 
Final Report 10 8 8 1 1 1 0 
DEC & DDL 
Submissions 2 2 5 0 0 0 0 
39
ANNEX 2: BIBLIOGRAPHY
Gretton, P. 2013. On input-output tables: uses and abuses, Staff Research Note.
ILO. 2019. Sectoral employment multipliers in Rwanda: Comparing local multipliers and input-output 
analysis.
ILO. 2017. How to Measure and Model Social And Employment Outcomes Of Climate And Sustainable 
Development Policies.
ILO. 2015. The employment dimension of infrastructure investments: A guide for employment impact 
assessment.
Kahn, R.F. 1931. The Relation of Home Investment to Unemployment.
NN. 2021. Finance from Bank Report – EG 3.2-27. 
NN. 2021. Transaction Close Report – BiD Capital Partners (EG 3.1-14).
MicroSave. 2017. MSME Finance in Rwanda – Status and Opportunities for Financial Institutions. 
NN. 2020. NN Employment Micromodel v1.06.xlsx.
NN. 2020. NN_macro_multiplier_model.xlsx.
NN. 2020. FY2020 Nguriza Nshore Market Systems Approach and Initial Changes in the Financial Services 
Market System Summary.
UN. 2018, Handbook on Supply and Use Tables and Input Output-Tables with Extensions and 
Applications.
USAID. 2021. Rwanda Nguriza Nshore Quarterly Report FY 2021: Quarter 1.
USAID. 2020. Rwanda Nguriza Nshore activity – Annual Performance Report.
USAID. 2020. COVID-19 Financial Institution Survey.
USAID. 2019. Feed the Future Rwanda Nguriza Nshore activity Baseline Report.
USAID. 2019. Rwanda Nguriza Nshore activity – Annual Performance Report.
USAID. 2018. Rwanda Nguriza Nshore activity – Annual Performance Report.
USAID. 2018. Feed the Future Rwanda Nguriza Nshore activity Monitoring, Evaluation and Learning Plan 
Year 1.
USAID. 2018. Rwanda Banking and Investment Analysis.
USAID. 2017. Local Enterprise and Value Chain Enhancement (LEVE) Project Measuring Job Impacts in 
the Haitian Apparel Sector Technical Report.
40
ANNEX 3: LIST OF KEY INFORMANT INTERVIEWS (KIIS)
List of Key Informant Interviews (KIIs)
Type Business/Org. Person Name Email/Telephone
SACCO IJABO SACCO KICUKIRO Higiro Emmanuel higiroe@gmail.com
SACCO KANJONGO SACCO Mukarugira Annonciata Kanjongosacco@yahoo.fr
SACCO UBUSUGIRE SACCO 
KIYOMBE Nkurikiyimana Jean Pierre 250788826967
SACCO BADUKA SHYORONGI 
SACCO Mukayiranga Jeanne d'Arc mudarc2@gmail.com
SACCO COOPEC CODEMARU Furaha Theodore ftheos2013@gmail.com
SACCO SACCO ICYEREKEZO 
KIZIMA KIVUYE
Harerimana Jean 
Damascene hdamas1973@gmail.com
SACCO KIRA MUTETE Munyaneza Papias papimunya01@yahoo.fr
SACCO SACCO TWAMBUTSANYE 
MUHONDO Uwitonze Angelique uwangela3yahoo.fr
SACCO SACCO Uruyange Rukoma Muramira James saccorukoma@gmail.com
SACCO SACCO Mbonezisonga Sindikubwayo Jean Baptiste saccombonezisonga@yahoo.fr
SACCO UKURI NYABIMATA 
SACCO Nyirakubumba Basilisse nyabimatasaccoukuri@gmail.com
SACCO UBUMWE SACCO 
GIKUNDAMVURA Nsabimana Benjamin nsabimanaben@gmail.com
SACCO SACCO Kira Maraba Ndayisaba Delphine ndelphine05@gmail.com
SACCO SACCO UMURAVA 
KIGOMA Mujawimana Marie Aimee mujaime2010@gmail.com
SACCO CESEKA ( KATABAGEMU) Kampire Delphine kampide@gmail.com
SACCO INGENZI MURAMBI 
SACCO Munyaneza Lambert mulampard14@gmail.com
SACCO SACCO TERIMBERE 
KIBILIZI Nkundabumviye Patrick patricknkundabumviye@gmail.com
SACCO IJABO SACCO KIGARAMA Zabanyinshi Jean Baptiste ijabosaccokigaramaa@gmail.com
SACCO CPF INEZA Niyigaba Marie Benoite nbenoite3@gmail.com
SACCO KAGANO SACCO Micomyiza Pie micomyizapie66@gmail.com
SME Comp. 1 Ravi Pharmacy Vincent 250788532338
SME Comp. 1 DPMC Jean Pierre 250788303784
SME Comp. 1 General Supply Phillipe 250788777580
SME Comp. 1 Syneduc Isaac 250790648530
SME Comp. 1 Adememorwa John 250788515941
SME Comp. 1 Poultry Farmer MUYOBOKE Ettienne 788408583
SME Comp. 1 Poultry Farmer TUYISENGE Immacule 783338945
SME Comp. 1 Poultry Farmer KAYIJUKA Theogene 788797727
SME Comp. 1 Poultry Farmer HAKIZIMANA Cyprien 785267582
SME Comp. 1 Poultry Farmer MNAIRAGUHA Ellen 788582434
SME Comp. 1 Yego Innovision Ltd NA Online through FI
SME Comp. 1 MUNYAXECO Francine Munyaneza francine@munyaxeco.com
SME Comp. 2 HomeInn Jeanne d'Arc Nikuze niyonzimaraymond@gmail.com
SME Comp. 2 SINA Rwanda Gerard Sina owner@sinarwanda.com
SME Comp. 2 BIDUGWA Ltd Claudine Munganyinka claudine.munganyinka59@gmail.com
SME Comp. 2 SE & KA Company Ltd Odette Kantesi infosekacompany@gmail.com
SME Comp. 2 The HappyNess Aline Uwase jesuline1@gmail.com
SME Comp. 2 Adecor Damien Ndizeye ndizeyedamien2000@yahoo.fr
SME Comp. 2 IMARB group Ltd Martin Bunane bunanemartin@gmail.com
SME Comp. 2 ASAFFI Inc Ltd Gilbert Muvandimwe muv2020@gmail.com
41
Type Business/Org. Person Name Email/Telephone
SME Comp. 2 Chillington Rwanda Ltd Khaled Mohammad mohammad.khaled@msn.com
SME Comp. 2 Cooperative Dushishoze 
(COOPEDUSH) Emmanuel Mporwiki dushishe@yahoo.fr
SME Comp. 2 easyHATCH Themba Elias themba@easyhatch.rw
SME Comp. 2 GET IT Rwanda Sproston Mark mark@getitrwanda.com
SME Comp. 2 Hollanda FairFoods Ltd Martin Bunane bunanemartin@gmail.com
SME Comp. 2 Jali Finance Ltd Felix Nkundimana felix@jalipartners.com
SME Comp. 2 Kigali Farms Demuynck Laurent laurent@kigalifarms.com
SME Comp. 2 Masaka Creamery Julie Igiraneza julie@masakacreamery.com
SME Comp. 2 Seeds of Trust Rwanda George Ngarambe sotseedsoftrust@gmail.com
SME Comp. 2 Vieberg Tobias Reiter tobias@viebeg.com
SME Comp. 2 Zean Enode Nduwayezu nduwenodeo@yahoo.com
FI partner Benefactors Sara Oon sara@benefactorsltd.com
FI partner Cogebanque Polepole Kayumba P_Kayumba@cogebank.com
FI partner Goshen Ignace Musangamfura imusangamfura49@gmail.com
FI partner Inkomoko Sara Leedom sara@inkomoko.com
Invest. advisor BID Capital Partners Mark Mutaahi mark@bidcp.com
Invest. advisor Nisk Capital Mutahi Njogu mutahi.njogu@niskcapital.com
Invest. advisor Market Intelligence Africa Fadzayi Musanhu fadzayi@marketintelligenceafrica.com
Investor Iungo Capital Michael Shema michael@iungocapital.com
Investor Jochem In't Velt jochemitv@hotmail.com
NN staff DAI Robin_Padberg@dai.com
NN staff DAI Nadine Manzi Nadine_Manzi@dai.com
NN staff DAI Michael Baingana Michael_Baingana@dai.com
NN staff DAI Richard Karugarama richard_karugarama@dai.com
NN Consultant DAI Annie Alcid Annie_Alcid@dai.com
NN Consultant Vanguard Economics Mads Knudsen mads@vanguardeconomics.com
NN Partner AMIR Aimable Nkuranga ankuranga@gmail.com
Other Rwanda Chamber of
Women Entrepreneurs
JEANNE FRANCOISE 
Mubiligi jfmubiligi@gmail.com
Other American Chamber of 
Commerce in Rwanda Gabriel Seidman president@amchamrwanda.com
Other Rwandan Bankers’ 
Association Tony Francis NTORE tony.ntore@rba.rw
42
ANNEX 4: INTERVIEW GUIDES
SME Interviews - Component 1
Please read before the interview: Hello, we are conducting an assessment of the Nguriza Nshore project, 
funded by USAID. The project aims to improve access to finance for small and medium sized businesses 
to help grow businesses and increase employment. We want to learn from small businesses such as yours 
about your experiences with accessing funding and how it may have changed your business. Your 
participation is completely voluntary, and you can choose not to answer any question. All responses will 
be anonymous.
1. #1 - Do you agree to continue with the interview? *
Mark only one oval.
Yes
No Skip to question 36
2. # 2 - Is it OK if I record this interview so I can improve my notes later? *
Mark only one oval.
Yes
No
Not possible to record
Background
3. #3 - Interview Date *
4. #4 - Interviewer name *
Mark only one oval.
John 
Elie
Chris
5. #5 - Business name *
6. #6 - Business start year *
7. #7 - Business type *
8. #8 - Business sector *
9. #9 - Business district *
10. #10 - Business city or town *
11. #11 - Respondent position in business *
Mark only one oval.
43
Owner
Manager but not owner 
Other:
12. #12 - Respondent age *
Mark only one oval.
Over 30
30 and under
13. #13 - Respondent gender *
Mark only one oval.
Male
Female
Experience with Lender
14. #14 - Why do you choose to borrow from Cogebanque/Goshen/BeneFactor instead of other 
lenders? *
15. #15 - When did you first receive a loan with them? *
16. #16 - How many loans have you received from them? *
17. #17 - How easy or difficult was the loan application process for you? How does it compare with 
other lenders? *
18. #18 - How convenient are the terms and structure of the loan for you? How do they compare 
with other lenders? *
19. #19 - Overall, how is your experience with Cogebanque/Goshen/BeneFactor? Will you return to 
them next time you need a loan? Why or why not? *
Loan Details
20. #20 - How much credit did you receive through your largest loan from 
Cogebanque/Goshen/BeneFactor? (RWF) *
21. #21 - When did you receive this loan from Cogebanque/Goshen/BeneFactor? *
22. #22 - When did you apply for this loan? *
23. #23 - Did you supplement this loan with other financing? How much? From where? *
24. #24 - Can you describe how you're using the money? Is it for working capital or investment? 
Explain. *
25. #25 - [Additional responses]
44
26. #26 - Before you received the loan we've been discussing (the largest one), how many full time, 
part time and casual employees did you have? *
27. #27 - How many full time, part time and casual employees do you have now? *
28. #28 - Why has your number of employees changed over this time period? *
29. #29 - How much, if at all, has the loan changed your average monthly revenue from before you 
got it to now? (Ask for before and after revenue in USD/RWF (note which), or % change if they 
prefer)
30. #30 - Can you explain how the loan changed your average monthly revenue?
31. #31 - How much, if at all, has the loan changed your number of full time, part-time or casual 
employees? (Note: could be an increase or decrease) *
32. #32 - How do you know that these jobs changed because the loan and not something else? *
33. # 33 - If you did not get the loan, would you have had to let employees go? If yes, how many 
fulltime, part-time or casual employee jobs would you have had to let go? (be careful about 
double counting) *
34. #34 - [Only if they report saving jobs above] How did the loan prevent you from having to lay off 
workers?
Employment
and revenue
NOTE: For the rest of the interview we are asking only about the largest loan they
have received from Cogebanque/Goshen/BeneFactor
45
SME Interviews - Component 2
Please read before the interview: Hello, we are conducting an assessment of the Nguriza Nshore project, 
funded by USAID. The project aims to improve access to finance for small and medium sized businesses 
to help grow businesses and increase employment. We want to learn from small businesses such as yours 
about your experiences with accessing funding and how it may have changed your business. Your 
participation is completely voluntary, and you can choose not to answer any question. All responses will 
be anonymous.
1. #1 - Do you agree to continue with the interview? *
Mark only one oval.
Yes
No Skip to question 42
2. # 2 - Is it OK if I record this interview so I can improve my notes later? *
Mark only one oval.
Yes
No
Not possible to record
Background
3. #3 - Interview Date *
4. #4 - Interviewer name *
Mark only one oval.
John
Elie
Chris
5. #5 - Business name *
6. #6 - Business type *
7. #7 - Business sector *
8. #8 - Business start year *
9. #9 - Business district *
10. #10 - Business city or town *
11. #11 - Respondent position in business *
Mark only one oval.
46
Owner
Manager but not owner 
Other:
12. #12 - Respondent age *
Mark only one oval.
Over 30
30 and under
13. #13 - Respondent gender *
Mark only one oval.
Male
Female
Investment Pipeline
14. #14 - How did you hear about investment opportunities through Nguriza Nshore? *
15. #15 - When did you apply initially? *
16. #16 - What was the final stage in the investment pipeline that you reached? *
Mark only one oval.
4 Received investment from investor
3 Deal closure support from investment advisor
2 Investment readiness and deal making support from investment advisor
1 Selected to receive support from investment advisor
17. #17 - [Only if they did not receive investment] Why did you stop at this stage?
18. #18 - When did you either close a deal with the investor or leave the Nguriza
Nshore pipeline? *
19. #19 - What is your opinion about the initial selection process? *
20. #20 - [Only if they reached investment readiness and deal making support stage] What kind of 
help did you receive on investment readiness and deal making? How useful was it? *
21. #21 - [Only if they reached deal closure stage] What kind of help did you receive to close deals 
with investors? How useful was it? *
22. #22 - In general, what is your view of the investment pipeline experience? Which parts worked 
well? What could have been improved? *
47
23. #23 - Have you applied what you have learned in the investment pipeline to attract other 
investments? If yes, explain. *
24. #24 - [Additional responses]
25. #25 - [Did they receive investment through Nguriza Nshore?] *
Mark only one oval.
Yes
No (ends the interview) Skip to question 42
Investment Details
26. #26 - How many investments have you received through the initial investor you were introduced 
to through Nguriza Nshore AND through other investments you attracted through the skills you 
learned? (note each separately) *
NOTE for those with multiple investments: For the rest of the interview we want to talk about the largest 
investment you attracted only, either through Nguriza Nshore or with skills you learned from the 
experience in the pipeline
27. # 27 - How much funding did you receive for your (largest) investment? *
28. #28 - What did you use this investment for? Was it for working capital or investment? Explain.
29. #29 - When did you receive the funding? *
30. #30 - Did you supplement it with any other funding? If yes, from where? How much? *
31. #31 - [Additional responses]
Employment and Revenue
32. #32 - Before you received the investment we've been discussing, how many full time, part time 
and casual employees did you have? *
33. #33 - How many full time, part time and casual employees do you have now? *
34. #34 - Why has your number of employees changed over this time period?
35. #35 - How much, if at all, has the investment you received changed your average monthly 
revenue from before you got it to now? (Ask for before and after revenue in USD/RWF, or % 
change if they prefer) *
36. #36 - Can you explain how the investment changed your average monthly revenue? *
37. #37 - How much, if at all, has the investment changed your number of full time, part-time or 
casual employees? (note: could be an increase or decrease) *
38. #38 - How do you know that these jobs changed because the investment and not something 
else? *
48
39. # 39 - If you did not get the investment, would you have had to let employees go? If yes, how 
many fulltime, part-time, or casual employee jobs would you have had to let go? (be careful about 
double counting) *
40. #40 - [Only if they report saving jobs above] How did the investment prevent you from having to 
lay off workers?
49
SACCO Interview
Please read: Hello, we are conducting an assessment of the Nguriza Nshore project, funded by USAID. 
The project aims to improve access to finance and help grow businesses and increase employment. We 
hope to learn from financial institutions such as yours about your experience receiving training through 
Nguriza Nshore, conducted by AMIR. This interview should take about 40 minutes to complete. Your 
participation is completely voluntary, and you can choose not to answer any question. All responses will 
be confidential and will not influence future project participation.
1. #1 - Do you agree to continue with the interview? *
Mark only one oval.
Yes
No Skip to question 25
2. #2 - Do you agree if I record this interview so I can improve my notes later? *
Mark only one oval.
Yes
No
Not possible to record
Background
3. # 3 - Interview Date *
4. #4 - Interviewer name *
Mark only one oval.
John 
Elie
Chris
5. #5 - SACCO name *
6. #6 - SACCO district *
7. # 7 - SACCO city or town *
8. # 8 - Respondent position in SACCO * Mark only one oval.
Branch manager
Loan officer
Other
9. # 10 - Respondent gender *
Mark only one oval.
50
Male
Female
Trainings from AMIR/Nguriza Nshore
10. # 11 - When did your SACCO start receiving trainings? Were you involved in making the 
decision? What was your SACCO's motivation for signing up for these trainings? *
11. # 12 - How many trainings did you receive? On which topics did you receiving trainings? (probe 
topics: 1. Governance, 2. Loan management (especially new product and service development), 3. 
Business continuity planning) *
12. # 13 - What did you learn in governance training session? How useful was it? *
13. #15 - What did you learn in business continuity (mostly COVID related) training? How useful 
was it? *
14. #16 - Overall, what is your opinion on the trainings? How could they have been improved? (ask 
about each of the 3 training topics) *
15. #17 - [Additional responses]
Changes after Training
17. #18 - What changes, if any, have you made as a result of what you learned in Governance 
training? *
18. #19 - What changes, if any, have you made as a result of loan management training? *
19. #20 - What changes, if any, have you made as a result of business continuity training? *
20. #21 - Have these changes led to any improvements for your SACCO? Explain. *
21. #22 - Have these changes led to any improvements for your SACCO members?
Explain. *
22. #23 - Have you seen changes in the total value of savings and credit since the trainings? By how 
much? (RWF before and after or % change) *
23. # 24 - If yes, why? What are the causes of these changes?
24. #25 - [Additional responses]
51
ANNEX 5: ONLINE SURVEY QUESTIONNAIRE
Nguriza Nshore Online Survey
CONSENT: Hello, we are conducting an assessment of the Nguriza Nshore project, funded by USAID. 
The project aims to improve access to finance and help grow businesses and increase employment. We 
hope to learn from financial institutions such as yours about your experience receiving training through 
Nguriza Nshore, conducted by AMIR. This short questionnaire should take less than 15 minutes to 
complete.
Your participation is completely voluntary, and you can choose not to answer any question. All responses 
will be confidential and will not influence future project participation.
Do you agree to participate?
Background
1. Date
2. What is the name of your financial institution?
3. Location
3b. District
4. What is your gender?
5. Approximately what percentage of the value of your loans this year has been commercial?
Trainings Received and Changes Made
Note: All questions about trainings refer to the trainings conducted by AMIR for Nguriza Nshore (a USAID 
Feed the Future program).
6. In which year did you receive trainings through Nguriza Nshore/AMIR?
Please mark all that apply.
7. How many trainings did you receive?
8. Did you receive training on governance topics?
Yes
No
Don’t know
9. How would you rate your knowledge and capacity on governance topics before this training?
1-Very low 2 3 4 5-Very high
10. How do you rate your knowledge and capacity on governance topics now?
52
11. Has your SACCO implemented any changes to improve governance based on what you learned 
in training?
Yes
No
Don’t know
12. Did you receive training on loan management topics, including improving products and services 
for your clients?
Yes
No
Don’t know
13. How would you rate your knowledge and capacity on loan management topics, especially 
improving products and services before this training?
1-Very low 2 3 4 5-Very high
14. How do you rate your knowledge and capacity on loan management topics, especially improving 
products and services today?
1-Very low 2 3 4 5-Very high
15. Has your SACCO developed any new products or services for members as a result of what you 
learned in training?
Yes
No
Don’t know
16. Did you receive training on business continuity planning topics?
Yes
No
Don’t know
17. How would you rate your knowledge and capacity on business continuity planning topics before 
this training?
1-Very low 2 3 4 5-Very high
18. How would you rate your knowledge and capacity on business continuity planning topics today?
1-Very low 2 3 4 5-Very high
53
19. Has your SACCO implemented any improvements to your business continuity planning based on 
what you learned in training?
20. Overall, how would you rate the capacity of your SACCO before training?
1-Very low 2 3 4 5-Very high
21. Overall, how would you rate the capacity of your SACCO now?
1-Very low 2 3 4 5-Very high
22. Has the amount of your lending changed since training?
Yes
No
Don’t know
SME Financing
Please check your records if possible
23. What was the value of your loan portfolio last year before you started trainings with 
AMIR/Nguriza Nshore?
24. What is the source of information for the value of your loan portfolio?
25. What is the value of your loan portfolio now?
26. What is the source of information for the value of your loan portfolio now?
THANK YOU FOR YOUR TIME!
54
ANNEX 6: SIGNED CONFLICT OF INTEREST (COI) AND NON-DISCLOSURE 
AGREEMENT (NDA) FORMS
John Bamuturaki, Consultant – ME&A, Inc.
55
56
Elie Nsabimana, Consultant – ME&A, Inc.
57
58
Christopher Root, Consultant – ME&A, Inc.
59