by
IMPACT EVALUATION REPORT
 January 2018
by
XIMIABWE:WORKS 
TABLE OF CONTENTS
TABLE OF CONTENTS ................................................................................................................ i
LIST OF FIGURES .......................................................................................................................... i
LIST OF TABLES ........................................................................................................................... ii
LIST OF ABBREVIATIONS AND ACRONYMS ....................................................................... iii
ACKNOWLEDGEMENTS ........................................................................................................... iv
EXECUTIVE SUMMARY ..............................................................................................................1
1. INTRODUCTION .................................................................................................................. 7
1.1 IMPACT EVALUATION OVERVIEW .............................................................................. 8
1.1.1 EVALUATION PURPOSE AND OBJECTIVES ........................................................................................................................8
1.1.2 EVALUATION QUESTIONS ..............................................................................................................................................8
2 EVALUATION METHODOLOGY ..................................................................................... 11
3 RESEARCH FINDINGS ....................................................................................................... 13
4 CONCLUSIONS AND RECOMMENDATIONS ............................................................... 50
5 ANNEXES ............................................................................................................................. 51
ANNEX 1: CASE STORIES ...............................................................................................................................................51
ANNEX 2: SCOPE OF WORK ................................................................................................................................................53
ANNEX 3: BASELINE QUESTIONNAIRE ...................................................................................................................................53
ANNEX 4: ENDLINE QUESTIONNAIRE ...................................................................................................................................53
LIST OF FIGURES
Figure 1: Personal incomes ...........................................................................................................................................13
Figure 2: Dwelling Index ................................................................................................................................................17
Figure 3: Access to Funding ..........................................................................................................................................21
Figure 4: Monthly revenues ..........................................................................................................................................21
Figure 5: Savings ..............................................................................................................................................................22
Figure 6: Economic Empowerment (Mediator) ........................................................................................................26
Figure 7: Self-efficacy (express opinion) ....................................................................................................................29
Figure 8: Resilience (adapt to changes) ......................................................................................................................35
Figure 9: Investment in risk games ..............................................................................................................................41
Figure 10: Compassion (relationships) .......................................................................................................................47
ANNEX 1: CASE STORIES ...............................................................................................................................................51
ANNEX 2: SCOPE OF WORK................................................................................................................................................53
ANNEX 3: BASELINE QUESTIONNAIRE...................................................................................................................................53
ANNEX 4: ENDLINE QUESTIONNAIRE ...................................................................................................................................53
LIST OF FIGURES
Fi re 1: Personal incomes Figure 1: Personal incomes...........................................................................................................................................13
Figure 2: Dwelling Index ................................................................................................................................................17
Figure 3: Access to Funding Figure 3: Access to Funding..........................................................................................................................................21
Figure 4: Monthly revenues..........................................................................................................................................21
Figure 5: Savings Figure 5: Savings..............................................................................................................................................................22
Figure 6: Economic Empowerment (Mediator)........................................................................................................26
Figure 7: Self Figure 7: Self efficacy -efficacy (express (express opinion) .................................................................................................................... .............................29
Figure 8: Resilience (adapt to changes)......................................................................................................................35
Figug re 9: Investment in risk games..............................................................................................................................41
Figure 10: Compassion (relationships).......................................................................................................................47
i
.............................................................................................. i
.............................................................................................. i
............................................................................................. ii
ACRONYMS ACRONYMS..................................... ....................................................................... .......................... iii
............................................................................................ iv
1
Table of Contents
 Acknowledgements List of Tables Ak ld c nowledgements
LIST OF TABLES
Table 1: Impact evaluation survey sample size and data collection summary ...................................................11
Table 2: Incomes .............................................................................................................................................................14
Table 3: Effect of the entrepreneurship modalities on the income index .........................................................15
Table 4: Effect of the employability modality on incomes .....................................................................................16
Table 5: Household assets ............................................................................................................................................17
Table 6: Effect of the entrepreneurship modalities on the assets .......................................................................19
Table 7: Effect of the employment modality on assets...........................................................................................20
Table 8: Entrepreneurship revenue, profits, savings and investments ................................................................22
Table 9: Effect of the entrepreneurship modalities on the entrepreneurship index .......................................23
Table 10: Employment ...................................................................................................................................................24
Table 11: Effect of the programme on the employability index ...........................................................................25
Table 12: Economic empowerment ............................................................................................................................26
Table 13: Effect of the entrepreneurship modalities on economic empowerment .........................................27
Table 14: Effect of the employability modality on economic empowerment ...................................................28
Table 15: Self-confidence ..............................................................................................................................................29
Table 16: Self-efficacy .....................................................................................................................................................29
Table 17: Effect of the entrepreneurship modalities on self-confidence ............................................................30
Table 18: Effect of the entrepreneurship modalities on self-efficacy ..................................................................31
Table 19: Effect of the employability modality on self-confidence ......................................................................32
Table 20: Effect of the employability modality on self-efficacy .............................................................................33
Table 21: Business growth rates for young entrepreneurs who received loans ..............................................34
Table 22: Wage employment job retention for at least 3 months ......................................................................34
Table 23: Resilience ........................................................................................................................................................36
Table 24: Effect of the entrepreneurship modalities on resilience ......................................................................38
Table 25: Effect of the employability modality on resilience.................................................................................40
Table 26: Risk attitudes .................................................................................................................................................40
Table 27: Time allocation ..............................................................................................................................................42
Table 28: Collective action ...........................................................................................................................................42
Table 29: Effect of the entrepreneurship modalities on risk attitudes ...............................................................43
Table 30: Effect of the entrepreneurship modalities on time allocation ............................................................44
Table 31: Effect of the employability modality on risk attitude ............................................................................45
Table 32: Effect of the employability modality on time allocation .......................................................................46
Table 33: Relationships ..................................................................................................................................................47
Table 34: Effect of the entrepreneurship modalities on relationships ................................................................48
Table 35: Effect of the employability modality on relationships ...........................................................................49
ii
LIST OF ABBREVIATIONS AND ACRONYMS
CD-RISC Connor–Davidson Resilience Scale
DFID Department for International Development
FGD focus group discussion
HH household
IT information technology
IYF International Youth Foundation
JAZ Junior Achievement Zimbabwe
KII key informant interview
MP Member of Parliament
NEDICO New Dimension Consulting
OLS ordinary least squares
PROWEB Professional Women, Women Executives and Business
PTS passport to success
RBCT Royal Business Consult Trust
RD Restless Development
Sida Swedish International Development Cooperation Agency
TV television
USAID United States Agency for International Development
USD United States dollar
YA Young Africa
Z:W Zimbabwe:Works Project Phase II
ND ACRONYMS
ence Scale
ional Development
ndation
babwe
List of Abbreviations and Acronyms
Z:W Zimbabwe:Works Project Phase II
iii
ACKNOWLEDGEMENTS
The Zimbabwe:Works (Z:W) project is appreciative of the 
partnership between the International Youth Foundation (IYF), 
United States Agency for International Development (USAID), 
Department for International Development (DFID) and Swedish 
International Development Cooperation Agency (Sida) that allowed 
for the success of this 66-month (June 2012-December 2017) 
initiative. Specific acknowledgements are extended to the Z:W 
partners (Junior Achievement Zimbabwe (JAZ), PROWEB, Royal 
Business Consult Trust (RBCT) and Restless Development (RD) 
who were part of the impact evaluation. Partners facilitated access to 
appropriate respondents in line with the impact evaluation inclusion 
criteria. Technical assistance provided by IYF Baltimore and Harare 
offices is also acknowledged. Most importantly, the impact evaluation 
was made possible by young people (20–35years) who took their time 
to be part of the data collection, and without whose participation this 
impact evaluation would not have been possible. New Dimension 
Consulting (NEDICO) sincerely hopes that this report will contribute 
towards improving the scope and quality of youth entrepreneurship 
and employability programmes in Zimbabwe and beyond.
AUTHORS
Victoria James
Natalie Carlson
Anselm Frieder Rink
Amos Milanzi
Innocent Yekeye
Tonderai Nyamwanza
 Acknowledgements
iv
1
EXECUTIVE SUMMARY
This report provides an overview of the impact evaluation findings of the Zimbabwe:Works Project Phase II (Z:W). The 
study sought to evaluate the project from the perspective of youth economic empowerment and income generation 
through entrepreneurship, financial inclusion and employability.
Operating environment
Zimbabwe experienced marked declines in socio-economic conditions between 2014, when the Z:W programme 
started, and 2016 when the impact evaluation was conducted. Deindustrialisation resulted in deepening economic 
informalisation. Structural economic degeneration catalysed a collapse in savings and investments rates. The programme 
was implemented in the context of declining macro-economic conditions. The evaluation was conducted in the context 
of an accelerated decline in the macro-economic and political environment. Progressive socio-economic degeneration 
between 2015 and 2017 resulted in the introduction of the bond notes, severe cash shortages and continued company 
closures. The Government also introduced Statutory Instrument 64 (SI64), which restricted the importation of specific 
goods. This affected the programme as some young people were running retail-related enterprises. Developments 
within the macro context affected programme implementation and were likely to exacerbate stress and negatively 
influence on risk taking, especially for youths intending to access loans. 
Design
The impact evaluation uses a randomised controlled trial to assess the effectiveness of the project. All individuals who 
were recruited and eligible to participate in a given modality were randomly assigned to a treatment or control group. 
Random assignment guarantees that participants in the project are, in statistical expectation, similar to non-participants.
Sample size and data collection
The impact evaluation targeted 2400 young people in both the treatment and the control groups. The impact 
evaluation comprised of baseline and endline surveys. The baseline survey was conducted from October to December 
2016 reaching 2,032 young people (1,147 the treatment and 885 the control groups). The endline survey was from 
September to October 2017, reaching 1,564 young people (869 the treatment and 695 the control groups). The 
overall baseline and endline response rates were 85% and 77% respectively, which are adequate to produce statistically 
significant results.
Data analysis, report writing and presentation
Impact evaluation data were analysed using the statistical programming software R. To assess the effectiveness of the 
modalities at the endline, the study team computed entrepreneurship and employability indices showing the overall 
effect of entrepreneurship and employability modalities on entrepreneurship and employment success respectively.
Measurement of effect
In order to be able to compare all the effects across the modalities, as well as the baseline and endline samples, this 
report converts all measures into standard scores. A standard score depicts the number of standard deviations by 
which a given estimate diverges from the mean. A positive standard score means a modality or treatment arm was 
higher than the mean in the whole sample. A positive score signifies that the value of an observation (in this case, in 
the treatment group) is above the mean value. In addition, we also report percentages and outcomes on scales (e.g. a 
scale ranging from 1 to 5).
Executive Summary
2
 Findings
FINDI NGS
ENTREPRENEURSHIP
Overview of entrepreneurship
Findings were computed into an entrepreneurship index computed from questions on income from the business, 
access to financing, availability of business savings and business investments. Assignment to treatment is associated 
with an average increase in monthly revenue of 125 USD. Relative to the control group, attending treatment (training 
only) was associated with an average increase of 99 USD in monthly revenue and 42 USD in monthly profits, while 
attending training and receiving a loan is associated with an average increase of 362 USD in revenue and 117 USD 
in profits. Assignment to treatment was associated with an increase in 13 USD in monthly business investment. The 
programme, combining modality 1a1
 and 1b2
 worked slightly better for women than for men. The evaluation estimated 
a 0.18 standard deviation increase as a result of the intervention. This estimate was substantively large and statistically 
significant. Modality 1a performed better than modality 1b. Specifically, modality 1a raised the income index by 0.18 
standard deviations (a statistically significant result), compared to 0.04 within modality 1b.
Household assets
A standardised household asset index was computed from questions regarding the ownership status for dwelling, type 
of roofing material, household assets owned and type of toilet used. Assignment to the treatment group was associated 
with approximately 0.05 of a standard deviation increase in household assets. This increase was not statistically 
significant, except among the subgroup that received a microfinance loan (about ƒ of the treatment participants), 
which was significant at the p=0.1 level. Access to loans provided young people with the financial potential to acquire 
household assets. Comparing modalities, modality 1a yields a higher treatment effect (0.08 standard deviations), 
compared to 0.04 standard deviations for modality 1b. 
Income
A standardised income index was computed from questions regarding individuals’ personal income, household income 
and the number of times individuals had gone without food, water, medicine or money in the past year. Personal 
monthly income increased by approximately 113 USD in the treatment group, while it remained constant in the control 
group. The programme increased the income index by 0.1 standard deviations. While this effect size was substantially 
meaningful, it was not statistically significant. Modality 1a performed better than modality 1b. Specifically, modality 1a 
raised the income index by 0.18 standard deviations (a statistically significant result), compared to 0.04 for modality 
1b. The effect of the programme on incomes was particularly promising for women as the evaluation estimated a 0.18 
standard deviation increase as a result of the intervention. The programme had the strongest treatment effect on 
incomes in Harare (an increase by 0.23 standard deviations). By contrast, in Bulawayo and Goromonzi the treatment 
effect was small and insignificant. In Goromonzi, the effect was even slightly negative (but small and insignificant).
Time allocation
The time allocation index was computed from questions regarding individuals’ self-reported free time in hours, 
individuals’ perceptions about their lives being too busy, and individuals’ likelihood to spend free time with leisure 
activities such as watching television, consuming alcohol or meeting up with friends. The programme had a negative 
effect on individuals’ free time as individuals assigned to the programme were 0.04 standard deviations less likely 
to have free time (broadly measured). The negative effect of the programme on free time was more pronounced 
for modality 1a. The reduction in free time was less pronounced and positive among women than among men. The 
effect (0.08 standard deviations), however, was not statistically significant. Regarding geographic differences, treatment 
effects were broadly similar across Harare (-0.07), Bulawayo (-0.02) and Goromonzi (-0.05). 
1 Modality 1a provided business development support and financial inclusion for aspiring young entrepreneurs, at least 6 months but less than 12 months in operation 
2 Analogous to 1a but targeting existing young entrepreneurs with at least 12 months in operation
3
Risk tolerance
The programme had a positive treatment effect on individuals’ willingness to take risks. The effect was more pronounced 
among those individuals that only attended the training, while it was slightly lower (and statistically insignificant) among 
those that also received funding. This may be owing to the fact that taking on credit makes individuals less likely to take 
risks than if they did not take on credit. Willingness to take risks was higher for modality 1b. Specifically, individuals in 
this modality scored 0.18 standard deviations higher on the risk tolerance index compared to their control group. In 
modality 1a, the increase was 0.08 standard deviations and not statistically significant. Overall, women also increased 
their risk tolerance, though the increase was slightly lower (0.12 standard deviations), compared to 0.14 standard 
deviations among men. The increase in risk tolerance due to the programme was largest in Goromonzi (0.21 standard 
deviations) and lowest in Bulawayo, where there was a negative, but insignificant effect.
Self-confidence
Self-confidence was raised by 0.02 standard deviations in modality 1a and reduced by 0.1 standard deviations in 
modality 1b. Both effects, however, were not statistically significant. The negligible effect on self-confidence was 
also confirmed among women. Regarding the geographic split, there was a significant reduction in self-confidence in 
Bulawayo (-0.86 standard deviations). In Harare, there was a positive effect on self-confidence, which, again, was not 
statistically significant. Taken together, then, the programme had a negligible effect on self-confidence.
Self-efficacy
The programme had a negative effect on self-efficacy. The reduction among individuals assigned to be treated was 0.17 
standard deviations. The effect was particularly pronounced among individuals who received funding (-0.34 standard 
deviations). It was insignificant and smaller among individuals who did not access funding (-0.08 standard deviations), 
than among individuals who did receive funding. Reductions in self-efficacy were particularly pronounced within 
modality 1a. The reduction was statistically significant (-0.21 standard deviations). Women experienced substantially 
lower self-efficacy rates, but it was statistically insignificant (-0.06 standard deviations). Reductions in self-efficacy were
particularly pronounced in Bulawayo and Goromonzi. 
Relationships
The programme reduced participants’ ability to maintain and develop social relationships. The reduction was particularly 
stark among the subset of beneficiaries that received funding (-0.47 standard deviations). It was not detectable among 
the subset of individuals that did not receive funding. This could be explained by qualitative accounts where young 
people within the treatment group reported investing more time in expanding or diversifying their ventures, which 
inevitably limited time for maintaining social relationships. Comparing modalities 1a and 1b, findings showed that the 
effect was largely comparable across the two with the treatment effect slightly more negative within modality 1b 
(-0.21) as compared to modality 1a (-0.18). However, these differences were not themselves statistically significant. 
With regard to gender, women showed a slightly lower negative treatment effect (-0.19), compared to men. Moreover, 
the treatment effect was not detectable in Harare, where it was even slightly positive. By contrast, the programme was 
associated with a statistically significant decrease in relationships in Bulawayo (-0.89) and Goromonzi (-0.32).
Economic empowerment
A standardised “Economic Empowerment” index was computed and it included all questions that measure individuals’ 
self-reported economic empowerment. The programme had a small and statistically significant negative effect on 
the economic empowerment index. The observed reduction in economic empowerment was roughly similar and 
statistically significant across modalities 1a (-0.18 standard deviations) and modality 1b (-0.14 standard deviations), as 
well as for women (column 5, -0.17 standard deviations). The reduction was particularly pronounced and statistically 
significant in Bulawayo (-0.51), followed by Goromonzi (-0.21), while the effect was close to 0 in Harare. 
Resilience
A standardised “Resilience” index was computed and it included all questions regarding individuals’ self-reported 
resilience. Resilience was slightly lower among treated youths, compared to those in the control group (-0.09 standard 
4
 Employability
deviations). The effect was statistically significant. The reduction was predominantly driven by individuals who received 
funding (-0.29 standard deviations, which was statistically significant). The effect was insubstantial and insignificant 
among beneficiaries who did not receive funding. The reduction was slightly larger among modality 1b (-0.12 standard 
deviations), compared to modality 1a (-0.05 standard deviations). Women showed slightly larger and statistically 
significant reductions in resilience (-0.17 standard deviations), compared to men. The effect was most negative and 
statistically significant within the Bulawayo sample. Respondents from Harare and Goromonzi did not witness a 
significant reduction in resilience.
EMPLOYABILITY3
Overview of employability
Findings were computed into an employability index based on questions on whether individuals had done an internship, 
whether they were paid during the internship, the length of the internship, as well as their current employment status. 
The programme had a positive effect as treated youths scored 0.11 standard deviations higher on the index compared 
to youths in the control group. Among young people who were exposed to Passport to Success (PTS), the programme 
increased the employability index by 0.22 standard deviations. This was statistically significant. This effect was large 
and reflected the success of the programme, particularly when coupled with training on PTS, a Life Skills training 
package used by IYF. The effect of the programme regarding incomes looked particularly promising for women with a 
statistically significant 0.24 standard deviation increase as a result of the intervention. 
Household assets
The programme had a negligible effect on household assets. The effect, -0.02 standard deviations, was very small and 
not statistically significant. In relation to gender, there were no discernible differences compared to the overall sample 
or to men. Women who participated in the programme were equally unlikely to see their household assets improve as 
a result of the programme. Geographical analysis showed that both Harare and Bulawayo yielded similar, inconclusive 
effects regarding assets, which were not statistically significant. However, most of the respondents stayed with their 
parents or guardians and they may not have focused on purchasing household assets. 
Income
The programme increased the income index by 0.13 standard deviations. This was statistically significant. Treated 
participants saw personal monthly incomes rise from 68 USD per month on average to 150 USD. Among the control 
group, monthly incomes rose from 88 USD to 144 USD on average. The difference in the effect sizes was substantially 
meaningful and statistically significant. The effect of the programme regarding incomes looked particularly promising 
for women with a 0.24 standard deviation increase as a result of the intervention. This estimate was substantively 
sizable and statistically significant. Geographic analysis showed that the programme had very similar effects in Harare 
and Bulawayo of around 0.12 standard deviations (not statistically significant).
Time allocation
The programme had a negative effect on individuals’ free time. The effect size, however, was small and insignificant. 
Young people who were in the treatment group were 0.08 standard deviations less likely to have free time (broadly 
measured). The reduction in free time was roughly similar among men and women, though the effect for women (-0.09 
standard deviations) however, was not statistically significant. Geographic analysis showed that the programme had 
stronger effects on time allocation in Bulawayo (-0.36), compared to Harare (0.07).
3 Modality 3=Work Readiness Training for New Job Entrants
5
Risk taking
The programme had positive and statistically insignificant treatment effects on individuals’ willingness to take risks. 
Overall, the effect was slightly lower among women (0.02 standard deviations). Finally, geographic analysis showed 
that the statistically insignificant increase in risk taking was more pronounced in Bulawayo (0.11), compared to Harare 
(0.02), though both estimates were insignificant.
Self-confidence
The programme had a strong, positive and statistically significant effect on self-confidence. Average scores among 
the treatment group on self-confidence measures rose from approximately 4.4 to 4.8 on a 1–5 scale. The positive 
effect on self-confidence was also confirmed among women. There was a strong and statistically significant increase 
in self-confidence in Harare (0.29 standard deviations), with in Bulawayo, the effect being 0. Insights from qualitative 
discussions showed that participating in the programme, (even without getting internship positions) enhanced young 
people’s confidence, especially in their own abilities and in their potential to secure employment.
Self-efficacy
The programme had a strongly positive and statistically significant effect on self-efficacy and the increase was pronounced 
among young people in the treatment group. The effect was particularly pronounced and statistically significant among 
individuals who did not participate in PTS (0.33 standard deviations). Women had highly similar estimates and thus 
an increase in self-efficacy compared to men. The increase was 0.30 standard deviations and statistically significant. 
Geographical analysis showed that the increase was particularly strong and statistically significant in Harare (0.47 
standard deviations) while the effect was small and insignificant in Bulawayo.
Relationships
The programme increased participants’ ability to maintain and develop social relationships (statistically significant. The 
increase was particularly stark among PTS participants (0.61 standard deviations). The ability to maintain relationships 
was slightly lower among the subset of individuals that did not participate (0.23). Regarding gender, women showed a 
slightly lower treatment effect (0.24), compared to men. However, the effect was still strong, positive and significant. 
The treatment effect was also particularly strong in Harare (0.38 standard deviations), while it was not detectable in 
Bulawayo. 
Economic empowerment
Empirical models showed that the programme had a statistically significant effect on economic empowerment. 
Individuals in the treatment group scored 0.25 standard deviations higher compared to those in the control group. 
Analysis of effects among actual beneficiaries (column 2) showed that the effect was particularly large and statistically 
significant among young people who participated in the PTS (0.38 standard deviations). The positive effect translated 
to women (0.23 standard deviations) and was particularly pronounced in the Harare. The effect was negligible and not 
statistically significant in Bulawayo.
Resilience
Resilience strongly increased with statistical significance as a result of the programme. Young people within the 
treatment group scored statistically significant 0.28 standard deviations higher on the resilience index. The increase 
was driven by individuals that participated in PTS (0.41 standard deviations) (statistically significant) although there 
were also effects observed among beneficiaries that did not participate in PTS. Women showed a slightly larger and 
statistically significant increase in resilience (0.31 standard deviations), compared to men. Finally, and in line with prior 
evidence, the effect was most pronounced and statistically significant in Harare (0.35 standard deviations).
6
 Recommodations
RECOMMENDATIONS
i. Future similar programmes should build on the evidence of multi-pronged entrepreneurship support anchored 
in needs analysis, needs driven capacity building, access to finance and mentorship.
ii. Future similar interventions should continue with an affirmative action approach to gender transformation, 
with a specific focus on deliberately targeting and supporting more females.
iii. Future similar projects should prioritise market linkages, as well as access to lucrative markets to ensure young 
entrepreneurs get the highest possible returns from their activities.
iv. There is need for further enquiry to explore reasons why entrepreneurial support, specifically access to 
microfinance loans, may potentially result in negative psychological outcomes.
v. Future similar projects should build on lessons learnt from the project with specific focus on supporting work 
readiness, facilitating access to internships as well as support towards accessing employment opportunities.
vi. Explore possibilities of scaling up the PTS programme towards more young people as a way of preparing them 
for employment.
vii. Future similar projects should focus on affirmative action for empowering young women, as well as basing 
employability interventions on robust labour market analyses.
viii. Future similar projects should focus on addressing the psychosocial needs of young people beyond supporting 
entrepreneurship and employability. 
7
Introduction
1 INTRODUCTION
The International Youth Foundation (IYF) invests in the extraordinary potential of young people. Founded in 1990, IYF 
builds and maintains a worldwide community of businesses, governments, and civil society organisations committed to 
empowering youth to be healthy, productive, and engaged citizens. IYF programmes are catalysts of change that help 
young people obtain a quality education, gain employability skills, make healthy choices, and improve their communities. 
IYF is currently implementing the second phase of the Zimbabwe:Works (Z:W) project, which is a USAID, DFID and 
Sida-funded 66-month (June 2012–December 2017) initiative of the IYF. The second phase of the project, which 
commenced in January 2015, sought to empower 22,000 Zimbabweans, particularly young women, economically and 
enable them to contribute towards and benefit from economic growth in Zimbabwe.
Context
Zimbabwe experienced marked declines in socio-economic conditions between 2014 when the Z:W programme 
started and 2017 when the impact evaluation was conducted. After a 52% decline in the gross domestic product 
(GDP) between 1999 and 2008, the economy experienced anaemic growth which averaged 2.9% from 2009 to 
2016. Significant structural degeneration was characterised by an increased dependence on primary commodities, 
deindustrialisation and informalisation of the economy. Negative saving, depressed investment levels, weakening public 
institutions, fiscal and debt mismanagement and corruption contributed towards infrastructure and public service 
degeneration (Kanyenze et al., 2016).
Deindustrialisation resulted in deepening economic informalisation. This was followed by reduced formal sector 
employment, the collapse of social protection systems and rising poverty levels. Studies have estimated that the 
share of manufacturing in the country’s GDP fell from 26.9% in 1992 to 11.7% in 2014. Declining capacity utilisation 
and reductions in the number of formal sector firms contributed towards downward trends. The share of informal 
employment in total employment increased from 80.0% in 2004 to 94.5% in 2014 (Kanyenze et al., 2016).
Structural degeneration of the economy resulted in a collapse of savings and investment rates, with savings estimated to 
have been negative since 2000 and, by 2015, were negative 11.0% of GDP. The economy was also affected by weakening 
public institutions characterised by fiscal and debt mismanagement, corruption accompanied by infrastructure and 
public service degeneration. Deepening fiscal deficits resulting from uncontrolled spending on government salaries and 
benefits contributed towards debt. Arrears in international debt further constrained the country’s ability to access 
external financing, and the low and falling levels of tax collection contributed towards the depleted tax base.
The programme was implemented in the context of declining macro-economic conditions. Specifically, the evaluation 
took place at the height of the economic crisis as there was a marked difference between end of 2014 when the 
project design was agreed upon and 2017. The accelerated decline in the macro-economic and political environments, 
progressively between 2015 and 2017, resulted in the introduction of the bond notes, severe cash shortages and 
continued company closures. These developments affected programme implementation and were likely to exacerbate 
stress and negatively influenced risk taking, especially for youths intending to access loans. The Government also 
introduced Statutory Instrument 64 (SI64), which restricted importation of specific goods. This affected the programme 
as some young people were running retail related enterprises.
Entrepreneurship modality (1a and 1b): Design and goals
The entrepreneurship modality consisted of a two-part intervention: first, 3–5 days of training, followed by a referral 
to a microloan organisation approximately a month after training. The targeted groups were young people aged 20 
to 35, including out-of-school youth, recent secondary school and university graduates, and university students, with 
businesses running for at least six months. The training focused heavily on developing business plans, advocating for a 
canvas model in which the entrepreneurs were asked to articulate their value proposition, cost structure, customer 
segments, and distribution channels. The entrepreneurs were also trained in financial literacy and encouraged to 
formalise both record keeping and business accounts. Other training modules focused on marketing, social media, 
customer relationships, and ethics.
The primary goals of the programme were as follows:
x to improve the economic well-being of young entrepreneurs, with specific goals of improving income, wealth, 
business revenues and profits, and business savings and investment;
x to promote the economic empowerment of participants; and
x to focus specifically on the empowerment and self-confidence of female entrepreneurs.
Employability modality (3): Design and goals
The employability modality was implemented in two different versions: one group received the basic partner customised 
work readiness training, while a second group received an additional component, the IYF flagship PTS programme. 
The standard training consisted of 3–5 days of training on work readiness, with a focus on workplace interpersonal 
skills, communication, and teamwork. Participants were also trained in self-promotion, including CV preparation and 
interview practice. The PTS programme lasted for three to six weeks and included sustained focus on life skills, with 
sessions on self-development, healthy behaviours, interpersonal skills, and workplace habits.
The primary goals of the programme were as follows:
x To improve the economic well-being of young people through employment, with the specific goal of securing 
internships that would lead to long-term gainful employment;
x To promote the economic empowerment of participants;
x To focus specifically on the employment access, empowerment and self-confidence of young women.
x
1.1 IMPACT EVALUATION OVERVIEW
1.1.1 Evaluation purpose and objectives
The impact statement of Z:W is: The broad empowerment of youth, particularly young women, to contribute 
towards and benefit from economic growth in Zimbabwe.
The objective of this study was to conduct a comprehensive impact evaluation of the Z:W Project Phase II. Specifically, 
the study sought to evaluate the project from the perspective of youth economic empowerment and income generation 
through entrepreneurship, financial inclusion and employability.
1.1.2 Evaluation questions
The evaluation objective was achieved through answering the following questions.
x Does the project improve the economic well-being of young women and men who participate? In particular, 
what is the impact on employment, business start-up and expansion, incomes, investment, savings, borrowing 
and lending? 
x Does the project promote the economic empowerment of participants, as measured by proxies such as 
physical mobility, aspirations for the future, control (or capacity to make decisions) over household resources 
(assets) or income and ability to organise with others? If so, which elements of the programme contributed 
to this?
x Does the intervention build young women’s self-confidence? If so, which elements of the programme contribute 
to building young women’s self-confidence: the employment/entrepreneurship training or the life-skills training? 
Or is it a combination of both elements?
x To what extent was the modality implemented in a gender-sensitive manner? 
x Is there a higher rate of enterprise start-up and growth rates, particularly amongst young women, for project 
participants than for non-participants?
8
x What are the business start-up and growth rates for young entrepreneurs who received loans, in 
comparison to those who did not receive loans?
x Have participants (or non-participants), particularly young women, who have accessed wage employment 
retained their jobs for at least three months?
x Which percentage of participants (or non-participants) who secured a job, retained the same job for three 
months? 
x Which percentage of participants (or non-participants), who secured a job, changed jobs (but were 
continuously working) for three months? 
x What were the outcomes of other participants (or non-participants) who secured a job, but did not 
continue to work and why?
x What is the influence of the integration of PTS life skills training in young women and men’s economic 
empowerment programming?
x Does PTS enhance impacts related to income, control over income, access to jobs, and personal resilience, 
particularly for young women?
x Does PTS trigger more transformative changes in participants, particularly young women, than other 
standard training-/capacity-building components of the project?
x Does the PTS curriculum improve young women’s and young men’s relationships with each other and with 
their families?
In addition to the evaluation questions above, the evaluation assesses the impact on the key empowerment dimensions 
of Power within, Power to, Power with, and Power over.
9
ZIMBAIWE:WORKS 
Evaluation Methodology
2 EVALUATION METHODOLOGY
This section provides an overview of the methodology for conducting the impact evaluation. The study was conducted 
using a randomised controlled trial. The section outlines the population, sample size selection as well as data collection 
and analysis processes.
Population
The target population for the intervention varied by modality. Modality 1a (Business Development Support and Financial 
Inclusion for Aspiring Young Entrepreneurs) targeted out-of-school youth, with business running for at least six, but less 
than 12 months. Modality 1b (analogous to 1a but targeting existing young entrepreneurs) targeted out of school youth 
with businesses running for 12 months and more. Modality 2 (Holistic vocational/technical training) targeted disadvantaged 
youth and those who have not completed secondary education. This modality was excluded from the study as the 
implementing partner, Young Africa, was no longer part of Z:W at the time of the evaluation. Modality 3 (Work 
Readiness Training for New Job Entrants) targeted unemployed youth with tertiary qualifications or new job entrants. The 
evaluation included participants in Harare, Goromonzi, and Bulawayo.
Design
The impact evaluation used a randomised controlled trial to assess the effectiveness of the three modalities, 1a, 1b, 
and 3. All individuals who were recruited and eligible to participate in a given modality were randomly assigned to 
either participate in the modality (treatment) or not (control). Random assignment guaranteed that participants were, 
in statistical expectation, similar to non-participants. The effect of the programme can, hence, be straightforwardly 
measured by comparing participants assigned to the control group with those assigned to a given treatment modality. 
The impact evaluation had two data collection phases, namely a baseline survey conducted in October to December 
2016 and the endline survey conducted in September to October 2017.
Overview of impact evaluation survey sample size and data collection
The impact evaluation initially targeted 2,600 young people; however, the figure was reduced to 2,400 young people in 
both intervention and control groups after Young Africa dropped out of the study.4
 Table 1 below shows the impact 
evaluation sample size and the number of participants reached during the baseline and endline phases of data collection. 
Table 1: Impact evaluation survey sample size and data collection summary
4 Young Africa was dropped due to governance and financial management challenges, which the organisation was facing at the time of evaluation.
11
0RGDOLW\ 7DUJHWHG
VDPSOH6L]H
3DUWLFLSDQWVUHDFKHG
GXULQJEDVHOLQH
VXUYH\
3DUWLFLSDQWV
UHDFKHGGXULQJ
HQGOLQHVXUYH\
D %HQHILFLDU\ ,QWHUYHQWLRQ ± 1R
376
  
&RQWURO   
7RWDO   
E %HQHILFLDU\ ,QWHUYHQWLRQ ± 1R
376
  
&RQWURO   
7RWDO   
 %HQHILFLDU\ ,QWHUYHQWLRQ ± 1R
376
  
%HQHILFLDU\ ,QWHUYHQWLRQ ± 376
,QWHJUDWLRQ
  
&RQWURO   
7RWDO   
$OO WKUHH
PRGDOLWLHV
%HQHILFLDU\,QWHUYHQWLRQ   
%HQHILFLDU\ ,QWHUYHQWLRQ ± 376
,QWHJUDWLRQ
  
&RQWURO   
7RWDO   
The overall baseline and endline response rates were 85% and 77% respectively, which were adequate to produce 
statistically significant results.
Qualitative data collection
The impact evaluation successfully conducted 10 focus group discussions (FGD) with young people, 11 key informant 
interviews (KII) with key stakeholders and four case stories with treatment group participants.
Data analysis, report writing and presentation
To assess the effectiveness of the modalities, NEDICO computed entrepreneurship and employability indices showing 
the overall effect of entrepreneurship and employability modalities on entrepreneurship and employment success 
respectively. A draft impact evaluation report was developed and shared with IYF and the donors. Feedback from the 
IYF and the donors was incorporated into the final impact evaluation report. 
Content and thematic analysis was used for qualitative data from FGDs and KIIs. The evaluation used Braun and 
Clarke’s5
 six phases of conducting thematic analysis where we: 
x familiarised ourselves with the notes; 
x generated initial codes; 
x searched for themes; 
x reviewed themes; 
x defined and named themes; and 
x produced the qualitative analysis results for each group. 
Impact evaluation strengths and limitations
This section presents the strengths and limitations of the impact evaluation.
Strengths
x the data exhibited almost no missingness;
x the measures were reliable and produced pronounced variation (i.e. standard deviation of variables);
x all measures scaled well, which was a result of careful measurement strategy;
x the data set was in a very polished shape, which aided the final report write up;
x all measures balanced very well across treatment and control groups.
Limitations, while few, were as follows:
x Cash shortages for respondents’ incentives.
Mitigation: NEDICO got the required cash at inflated rates.
x Low response rates at central meeting places despite adequate prior notice of the data collection dates.
Mitigation:
Below are the measures taken to address the response rate: -
o pre-mobilisation meeting between IYF Harare, Z:W partners and NEDICO to discuss the participants list and 
mobilisation strategies;
o NEDICO supported partners with community mobilisation support in terms of airtime and allowances for 
community mobilisers;
o NEDICO provided incentives of 2.00 USD to all participants (both treatment and control) to encourage 
participation in the endline;
o door-to-door visits of participants who could not come to mobilisation centres; and
o if the above all failed, in consultation with IYF Harare and partners, NEDICO conducted phone interviews
5 Braun, V. and Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3: 77–101. 
12
13
3 RESEARCH FINDINGS
This section presents research findings under each research question.
3.1 Research question 1: Does the project improve the economic well-being of young women and men who participate? In 
particular, what is the impact on employment, business start-up and expansion, incomes, investment, savings, borrowing 
and lending?
3.1.1 Incomes
The programme improved the economic 
well-being of young women and men who 
participated. Table 2 shows the overall effect of 
the programme on incomes using descriptive 
statistics. Comparisons between baseline and 
endline data show that the treatment group, 
across all modalities experienced income 
increases. This was particularly noteworthy 
given that income increases in the control 
group were less pronounced. The treatment 
group in modality 1a saw personal incomes 
increase from 160 to 307 USD (Figure 1). 
Meanwhile, the control group only saw incomes 
increase from 136 to 180 USD. The same 
trend was observed in relation to outcomes 
such as “has gone without food” or “has gone 
without water”. The items, which were scored 
on a scale, improved across most treatment 
conditions. This also held for modality three. 
The PTS component saw stronger increases, 
compared to the non-PTS components. This 
held across most outcomes, besides the very 
last.
Figure 1: Personal incomes
Research Findings
Table 2: Incomes6
Effec t of the entrepreneurship modalities on 
the income index 
The standardised income index which was 
computed using questions that measured 
an individuals’ personal income, their 
household income and the number of times 
individuals have gone without food, water, 
medicine or cash in the past year. The 
programme increased the income index by 
0.1 standard deviations. While this effect 
size was substantially meaningful, it was 
not statistically significant. Inputs from key 
informants showed that the extent to which 
the programme could influence income 
was constrained by the socio-economic 
conditions, which rapidly declined during 
the time when the evaluation was being 
conducted. Overall, the extent to which 
the young people succeeded or not, was 
influenced more by deteriorating macro￾level socio-economic conditions.
Access to funding contributed towards increases in income, which were shown more strongly for individuals that 
accessed funding. Among these individuals, the effect was 0.14 standard deviations, compared to 0.06 standard deviations 
among individuals that did not receive funding. These effect sizes were not statistically significant. Columns 3 and 4 in 
Table 3 below, compares the programme’s effectiveness across modalities 1a and 1b. Modality 1a performed better 
than modality 1b. Specifically, modality 1a raised the income index by 0.18 standard deviations (a statistically significant 
result), compared to 0.04 within modality 1b. Again, however, unobserved variables may explain this difference that 
has nothing to do with the two programmes. 
The effect of the programme regarding incomes was particularly promising for women (see column 5). The evaluation
estimated a 0.18 standard deviation increase as a result of the intervention. This estimate was substantively sizable and
6 BL=Baseline & EL=Endline
“My loan was approved just about the time when the 
cash crisis worsened. The microfinance institutions only 
paid out through bank accounts and in some instances, 
banks were giving out USD20 per day. I had a USD500 
loan, which meant that if I wanted to withdraw it I 
would need 25 days of travelling. Bus fare to the bank 
costs USD7 one way so that would mean USD14 to 
travel to get USD20” 
 (male FGD participant, Shumba Goromonzi).
14
Dependent variable
15
statistically significant. Females who were part of focus group discussions highlighted that prior to the programme their 
businesses were mostly not well planned but as a result of the programme they were starting to plan better, track 
their incomes and expenditure while focusing on growth. Finally, the programme, had the strongest treatment effect on 
incomes in Harare (an increase by 0.23 standard deviations). By contrast, in Bulawayo and Goromonzi, the treatment 
effect was small and insignificant. In Goromonzi, the effect was even slightly negative (but small and insignificant). 
Table 3: Effect of the entrepreneurship modalities on the income index
 Effect of the employability modality on incomes
Table 4 reports the overall effect of the employability programme on incomes. Column 1 reports the effect of being 
randomly assigned to the treatment (“Assigned to treat”) on a standardised income index. The index includes all 
questions that measure an individual’s income. Specifically, it includes individuals’ personal income, their household 
income and the number of times individuals have gone without food, water, medicine or cash in the past year. As 
before, the evaluation standardised these variables and averaged across them in order to build a comprehensive and 
comparable “Income Index”.
Column 1 shows the average treatment effect of the programme on the income index. T he evaluation found that the 
programme increased the income index by 0.13 standard deviations. This effect on size was substantially meaningful and 
statistically significant. When scrutinising the treatment effect among those individuals that attended the programme, 
comparing them to those that did not, the evaluation found that the income index, again, increased more strongly for 
individuals that participated in PTS. Among these individuals, the effect was 0.22 standard deviations, compared to 
0.08 standard deviations among individuals that did not participate. The effect of the programme regarding incomes 
looked particularly promising for women (see column 3). Here, the evaluation estimated a 0.24 standard deviation 
increase as a result of the intervention. This estimate is substantively sizable and statistically significant. Finally, the 
programme had very similar effects in Harare and Bulawayo of around 0.12 standard deviations.
Dependent variable
Dependent variable
Table 4: Effect of the employability modality on incomes
3.1.2 Household assets
Table 3 shows the overall effect of the programme on household assets using descriptive statistics. The table shows that 
most asset measurements are higher in the treatment group. The dwelling index (ranging from 1–5), for instance, is 4.43
points in the treatment group 1a and 1b in the endline, compared to 4.37 and 4.41 points in the baseline, respectively (see
Figure 2). 
A similar trend is detectable for most items. Specifically, focusing on the treatment group in modality 1a and comparing 
the endline data to the baseline, the roof index increases from 2.35 to 2.39. The proportion of individuals owning a 
radio increased from 0.69 to 0.76. The proportion of individuals owning a TV increased from 0.70 to 0.76. 
The evaluation documented a decrease in ownership of livestock from 0.47 to 0.42 in the rural area of Goromonzi. 
Overall, the same trend was detectable when scrutinising the treatment group in modality 1b. By contrast, outcomes 
did not increase within modality 37
 where there were reductions between the end- and baseline. Findings from modality 
3 could be attributed to the broader macro-economic challenges and the resulting high levels of unemployment 
which meant young people could not get employment which would have allowed them to transform their household 
asset holdings positively. Decreases in the overall household assets index could be attributed to retrenchments and 
unemployment, and could be attributed especially to these factors as evidence has shown that the selling of household 
assets is one of the coping mechanisms adopted by households in times of hardship. 
7 Work Readiness Training for New Job Entrants
Dependent variable
16
17
Figure 2: Dwelling Index
Table 5: Household assets
Effe ct of the entrepreneurship modalities on the assets
Table 6 shows the overall effect of the entrepreneurship programme on household assets. Column 1 reports the effect 
of being randomly assigned to the treatment (“Assigned to treat”) on the standardised household asset index. The 
index includes all questions that measure the quality of an individual’s household assets. Specifically, it includes whether 
individuals’ ownership status of their dwelling (e.g. rent or ownership), the material their roof is made of (e.g. metal 
vs. tiles), the household assets they own (e.g. radio or television), and the kind of toilet they use (e.g. flush toilet or pit
latrine). Again, the evaluation standardised these variables and averaged across them in order to build a comprehensive 
“Household Asset Index”.
As can be seen in column 1, the pro￾gramme (combining modality 1a and 
1b) had a weakly positive effect on 
household assets. The effect, 0.05 stan￾dard deviations, however, was small 
and not statistically significant. This 
could be attributed to the fact that the 
timeframe between baseline and endline 
was inadequate for young people to have 
transformed their incomes and positively change their household assets. It could also show that young people were 
focusing on reinvesting to grow their businesses before focusing on improving their household assets. 
Column 2 focuses on those individuals that attended the programme, comparing them to those that did not attend. 
The evaluation estimated the same small effect of 0.05 standard deviations. Again, however, the effect was more 
positive and statistically significant for those individuals that also obtained funding from the microfinance institutions 
in the programme. These individuals scored 0.12 standard deviations higher on the household asset index. Access to 
funding was a critical success factor as outlined by participants in qualitative discussions who raised two key themes in 
relation to it. The first was that access to funding enabled them to increase production in whatever business they were 
involved in. Respondents focusing on market gardening for example highlighted that big, well-paying clients often prefer 
to buy from suppliers who can guarantee constant supply. The second theme related to limited resources for branding 
and marketing which increased product appeal as well as facilitating access to more lucrative markets.
Columns 3 and 4 compare effect sizes across modalities 1a and 1b. Again, evidence shows that modality 1a yields a 
higher treatment effect (0.08 standard deviations), compared to 0.04 standard deviations for modality 1b. Both effects, 
however, are not statistically significant and are small. Comparisons between the two groups were also problematic, 
given that individuals were not randomly assigned to 1a or 1b. Rather, unobserved variables may explain these 
differences (although the evaluation controlled for salient variables, as discussed above).
Column 5 shows the effect sizes for women. The evaluation found no differences between the overall sample and men. 
Women who were part of the programme were equally likely to see their household assets improve. However, the 
effect is small and statistically insignificant. Finally, columns 6, 7 and 8 show the geographical analysis where Harare and 
Goromonzi were highly comparable. Here, the programme increased the asset index by 0.04 standard deviations. The 
effect was much higher in Bulawayo (0.17 standard deviations), though it was not statistically significant due to the low 
number of individuals (99).
“Since I got training and received funding I managed to increase the 
number of chickens I keep. Right now, I am focusing on repaying 
my loan and making sure my business grows so there are not many 
changes in what I own or the life I live” 
 (male FGD Participant, Goromonzi).
18
Table 6: Effect of the entrepreneurship modalities on the assets
E ffect of the employment modality on assets
Table 7 shows the overall effect of the employment programme on household assets. Column 1 shows the effect of 
being randomly assigned to the treatment (“Assigned to treat”) on the standardised household asset index. The index 
includes all questions that measure the quality of an individual’s household assets. Specifically, it includes individuals’ 
ownership status of their dwelling (e.g. rent or ownership), the material their roof is made of (e.g. metal vs. tiles), the 
household assets they own (e.g. radio or television), and the kind of toilet they use (e.g. flush toilet or pit latrine). Again, 
the evaluation standardised these variables and averaged across them in order to build a comprehensive “Household 
Asset Index”. 
Column 1 shows that the programme had a negligible effect on household assets. The effect, -0.02 standard deviations, 
was very small and not statistically significant. Column 2 focuses on those individuals that actually attended the 
programme, comparing them to those that did not attend. Here, the evaluation estimated that there is similarly small 
effect of -0.05 standard deviations. The effect is similar across those individuals that participated in the PTS and those 
that did not. 
Column 5 shows the effect sizes for women. Here, there were no discernible differences compared to the overall 
sample or men. Treated women were equally unlikely to see their household assets improve as a result of the 
programme. Finally, columns 4 and 5 repeated the aforementioned geographical analysis. Harare and Bulawayo yielded 
similar, inconclusive effects regarding assets, which were not statistically significant. 
Dependent variable
19
Table 7: Effect of the employment modality on assets
3.1.3 Entrepreneurship revenue, profits, savings and investments
Table 8 shows the overall effect of the programme on entrepreneurship using descriptive statistics. These outcomes 
only pertain to modality 1. Most outcomes in this category were positively affected by the programme. Specifically, 
when comparing the end- to the baseline in treatment group modality 1a, the evaluation found a 14-percentage point 
increase in individuals who accessed funding (see Figure 3). 
Dependent variable
20
Figure 3: Access to Funding
Figure 4 shows that monthly revenues increased from 432 USD at baseline to 637 USD at endline. This is in contrast 
to those in the control group which experienced a reduction from 386 USD to 301 USD. The trend was also visible 
within modality 1b.
Figure 4: Monthly revenues
In general, though, most outcomes, including savings per month, investments and overall savings, increased as a result 
of the programme. At the baseline, 19% of the control group and 21% of the treatment group had business savings. At 
the endline, 30% of the control group and 52% of the treatment group had savings. Assignment to treatment was also 
associated with an increase in 13 USD in monthly business investment.
21
Figure 5: Savings
Differences between receiving the training only and also receiving a microloan cannot be interpreted with the same 
causal certainty, as assignment to the loan was not random within the treatment group. However, relative to the 
control group, attending treatment (training only) was associated with an average increase of 99 USD in monthly 
revenue and 42 USD in monthly profits, while attending training and receiving a loan was associated with an average 
increase of 362 USD in revenue and 117 USD in profits.
Table 8: Entrepreneurship revenue, profits, savings and investments
Effect of the entrepreneurship modalities on the entrepreneurship index
Table 9, shows the overall effect of the entrepreneurship programme on entrepreneurial success. The entrepreneurship 
index presented was computed based on questions that measured the quality of an individual’s enterprise. Specifically, 
this included whether individuals obtained an income through their business, whether they have accessed funding, 
whether they have business savings and whether they had business investments. The evaluation standardised these four 
22
variables and averaged across them in order to build 
a comprehensive “Entrepreneurship Index”. 
The Z:W programme across modality 1a and 1b 
significantly increased the entrepreneurship index. 
Specifically, young people who were part of the 
treatment group scored 0.2 standard deviations 
higher on the index compared to individuals assigned 
to the control group. This marked a sizable and 
precisely measured improvement. Given that the 
treatment was randomly assigned, the effect can 
be interpreted as causal. The evaluation confirmed 
a strongly positive effect of the programme. This 
holds particularly true when scrutinising the subset 
of individuals that accessed funding. Here, the 
programme increased the entrepreneurship index 
by 0.8 standard deviations. This effect was large and 
speaks to the success of the programme, particularly 
when coupled with access to funding. 
Columns 2 and 3 compare the effect of the pro￾gramme on the entrepreneurship index across 
modality 1a and 1b. Both modalities significantly 
improved the entrepreneurship index. Interestingly, 
however, the evaluation found that modality 1a had 
a more positive effect than modality 1b. In particular, 
modality 1a improved the index by 0.3 standard deviations as compared to 0.2 standard deviations in modality 1b. This 
difference, however, needs to be interpreted with caution as individuals were not randomly assigned to the modalities. 
Column 4 focuses on the effect of the programme among women compared to men. Results showed that the programme 
– combining modality 1a and 1b – worked slightly better for women. Women, in particular, saw the entrepreneurship 
index improve by 0.26 standard deviations, compared to 0.23 in the overall sample and 0.20 among men. The finding 
thus speaks to the success of the programme in raising the economic prospects of female entrepreneurs. The difference, 
however, was small and in itself not statistically significant. 
Table 9: Effect of the entrepreneurship modalities on the entrepreneurship index
23
Table 10: Employment
Finally, columns 6, 7 and 8 compare the effectiveness of the programme across the three implementation areas, 
Harare, Bulawayo and Goromonzi. The programme improved the entrepreneurship index across all three areas. The 
strongest effects were found in Harare and Goromonzi, where, the programme raised the entrepreneurship index by 
0.25 standard deviations. The success of the programme was positive, but slightly less pronounced, in Bulawayo (0.13 
standard deviations). There, thus, does not seem to be a clearly interpretable urban/rural gap. 
3.1.4 Employment
Table 10 shows the effect of the programme on employment-related outcomes using descriptive statistics. The 
table shows a treatment effect on employment of approximately 15 percentage points for the PTS module, and of 8 
percentage points for the no-PTS modality. However, the control group also witnessed 8 percentage points, which 
showcases a rather mild effect of the overall intervention. Gross and net salaries, similarly rose as a result of the 
programme. The effect was most noticeable for the no-PTS component (429 USD for gross and 391 USD for net). 
Effect of the employability programme on employment
Table 11, shows the overall effect of the employability modality on employment success. Column 1 reports the effect 
of being randomly assigned to the treatment (“Assigned to treat”) on the standardised employability index. The index 
includes all questions that relate to an individual’s employment track records. Specifically, it includes information on 
whether individuals have done an internship, whether they were paid during the internship, the length of the internship, 
as well as current employment. The evaluation standardised all variables and averaged across them in order to build a 
comprehensive “Employability Index”. 
As can be seen in column 1, Z:W had a statistically insignificant effect on the employability index. Specifically, the 
individuals randomly assigned to participate scored 0.11 standard deviations higher on the index compared to individuals 
randomly assigned to the control group (statistically insignificant). However, given that the treatment was randomly 
assigned, the effect can be interpreted as causal. Still, all results henceforth presented control for the implementing 
partner (using fixed effects) and the area under study (using fixed effects), as well as individual-level pre-treatment 
covariates such as gender, age and education.
0RGDOLW\376 0RGDOLW\1R376 0RGDOLW\&RQWURO
(PSOR\HG(/   
*URVVVDODU\(/   
1HWVDODU\(/   
'RQHLQWHUQVKLS(/   
24
Table 11: Effect of the programme on the employability index
The Z:W programme did not result in differences in access to internships for young people. Key information outlined 
that macro-economic challenges, deindustrialisation, as well as high levels of unemployment contributed towards 
challenges in facilitating access to internships for young people. 
Column 2 shows the effect of the programme on those that attended the programme. The analysis compared individuals 
that attended to those that did not attend. Given that attendance may be plagued by self-selection, this estimate is to 
be interpreted with caution. Still, the analysis confirmed a positive effect of the programme. This held particularly true 
when scrutinising the subset of individuals that took part in the PTS component. Here, the programme increased the 
employability index by 0.22 standard deviations. This effect was large and speaks to the success of the programme, 
particularly when coupled with access the proprietary IYF PTS tool. 
Column 3 focuses on the effect of the programme on women compared to men. Women scored 0.08 standard 
deviations higher, though the effect was not statistically significant. Finally, columns 4 and 5 show comparisons of the 
programme’s effectiveness across the two implementation areas, Harare and Bulawayo. The programme improved 
the employability index across both areas. The strongest effect was found in Harare: here, the programme raised the 
Dependent variable
25
employability index by 0.20 standard deviations. The success of the programme was not positive, but not statistically 
significant, in Bulawayo (-0.09 standard deviations). 
3.2 Research question 2: Does the project promote the economic empowerment of participants, as measured by proxies such as 
physical mobility, aspirations for the future, control (or capacity to make decisions) over household resources (assets) or income 
and ability to organise with others? If so which elements of the programme contributed to this?
Table 12 shows the overall effect of the programme on economic empowerment using descriptive statistics. The first 
item is given in percentages, while the remaining items are scored on a five-point scale from “strongly disagree” to 
“strongly agree.” While the evaluation could not confirm a positive effect on the first outcome (“finds moving easy”), 
most other items were positively affected across the modalities. Being a good mediator increased from 4.12 to 4.43 
in the treatment group 1a from the base- to the endline (Figure 6). Similar increases were detectable across the other 
treatment groups and outcomes. However, the control group also experienced these increases. Regarding modality 3, 
the evaluation documented stronger increases within PTS when comparing treated individuals between the base- and 
the endline. For example, the mediator outcome increased from 4.11 to 4.65 within PTS, while the increase is slightly 
smaller from 4.19 to 4.6 in the no-PTS treatment group.
Figure 6: Economic Empowerment (Mediator)
Table 12: Economic empowerment
26
Eff ect of the entrepreneurship modalities on economic empowerment 
Table 13 shows the overall effect of the entrepreneurship programme on individuals’ empowerment. It provides a 
holistic econometric approach, which controls for background variables and differences across the treatment and 
control groups. Column 1 shows the effect of being randomly assigned to the treatment (“Assigned to treat”) on 
a standardised “Economic Empowerment Index”. The index includes questions regarding individuals’ self-reported 
economic empowerment. It includes questions such as “I can resolve conflicts between people while working in a 
group” or “I can work effectively with others in a team”. The evaluation standardised these variables and averaged 
across them in order to build a comprehensive and comparable “Economic Empowerment Index”.
In line with the findings on self-efficacy and self-confidence, column 1 demonstrates that the programme had a small 
and statistically significant effect on the economic empowerment index. Individuals assigned to be treated scored 0.15 
standard deviations lower on the index compared to individuals randomly assigned to the control group. The effect 
was slightly lower among individuals that attended the programme (treated: training only) and was not statistically 
significant. Once again, however, the evaluation found that individuals who received funding scored significantly lower 
on the index (-0.33 standard deviations). 
The observed reduction in economic empowerment was roughly similar across modalities 1a (-0.18 standard deviations) 
and modality 1b (-0.14 standard deviations), as well as for women (column 5, -0.17 standard deviations). As before, 
the reduction was particularly pronounced in Bulawayo (-0.51) followed by Goromonzi (-0.21), while the effect was 
close to 0 in Harare. 
Table 13: Effect of the entrepreneurship modalities on economic empowerment
Dependent variable
27
Effe ct of the employability modality on economic empowerment 
Table 14 shows the overall effect of the employability programme on individuals’ empowerment. Column 1 reports the 
effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Economic Empowerment”-
Index. The index included questions on individuals’ self-reported economic empowerment. It included questions such 
as “I can resolve conflicts between people while working in a group” or “I can work effectively with others in a 
team.” The evaluation standardised these variables and averaged across them in order to build a comprehensive and 
comparable “Economic Empowerment Index.”
The programme had a strongly significant effect on economic empowerment. Individuals randomly assigned to be 
treated scored 0.25 standard deviations higher compared to those assigned to the control group. When scrutinising 
the effect among actual beneficiaries (see column 2), evidence showed that the effect was particularly large among 
individuals that participated in the PTS (0.38 standard deviations). The positive effect translated to women (0.23 
standard deviations) and was particularly pronounced in the Harare sub-sample. The effect was negligible and not 
statistically significant in Bulawayo.
Table 14: Effect of the employability modality on economic empowerment
3.3 Research question 3: Does the intervention build young women’s self-confidence? If so, which elements of the programme 
contribute to building young women’s self-confidence: the employment/entrepreneurship training or the life-skills training? Or is 
it a combination of both elements?
Tables 15 and 16 show the overall effect of the programme on self-confidence and self-efficacy using descriptive 
statistics. All reported items were scored on the same five-point scale, which ranges from “strongly disagree” to 
Dependent variable
28
“strongly agree.” Here, too, the evaluation confirmed positive increases from the base- to the endline in the treatment 
groups. All outcomes increased by roughly 0.2 to 0.5 points. For example, Figure 15 shows that individuals in the 
baseline in treatment group 1a score a 4.1 on “I am able to express my opinions.” This number increases to 4.33 on 
the endline. Similarly, individuals scored 4.28 for the item “I stand up for things” during the baseline, which increases 
to 4.54 during the endline. Again, however, the control group, too, saw increases. These were not as pronounced, 
pointing to the positive effect of the programme. Once more, the tables 15 and 16 show that the PTS treatment group 
saw stronger increases compared to the no-PTS group.
Figure 7: Self-efficacy (express opinion)
The problem-solver outcome, for example, increase from 4.21 to 4.90 in the PTS treatment group, while the increase 
is slightly smaller for the no-PTS group (4.32 to 4.84).
Table 15: Self-confidence
Table 16: Self-effi cacy
29
Effect o f the entrepreneurship modalities on self-confidence 
Table 17 shows the overall effect of the entrepreneurship programme on individuals’ self-confidence. Column 1 reports 
the effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Self-Confidence Index”. 
The index included all questions that measure an individual’s self-assessed self-confidence. Specifically, it included 
questions such as “I am able to express my opinion and discuss sensitive issues” and “I am confident in my future.” The 
evaluation standardised these variables and averaged across them in order to build a comprehensive and comparable 
“Self-Confidence Index”. 
Column 1 of Table 17, demonstrates that the programme had a negligible effect on self-confidence. If anything, the 
programme slightly decreased self-confidence. However, the finding was not statistically significant and the standard 
errors were large. The same held for individuals without training. Across modality 1a and 1b, the evaluation found 
similarly inconclusive evidence. Self-confidence was modestly raised by 0.02 standard deviations in modality 1a and 
reduced by 0.1 standard deviations in modality 1b. Both effects, however, were not statistically significant. 
Column 5 demonstrates that the negligible effect on self-confidence was also confirmed among women. Regarding 
the geographic split, the evaluation found that there was a significant reduction in self-confidence in Bulawayo (-0.86 
standard deviations). This sample, however, was rather small. In Harare, there was a positive effect on self-confidence, 
which, again, was not statistically significant. Taken together, then, the programme had a negligible effect on self￾confidence.
Table 17: Effect of the entrepreneurship modalities on self-confidence
Dependent variable
30
Effect o f the entrepreneurship modalities on self-efficacy
Table 18 shows the overall effect of the entrepreneurship programme on individuals’ self-efficacy. Column 1 shows the 
effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Self-Efficacy Index”. The 
index included all questions that measure an individual’s self-assessed self-efficacy. Specifically, it included questions 
such as “I can always manage to solve difficult problems if I try hard enough” and “I am certain that I can accomplish 
my goals.” The evaluation standardised these variables and averaged across them in order to build a comprehensive 
and comparable “Self-Efficacy Index”. 
Column 1 shows that the programme had a negative effect on self-efficacy. The reduction among individuals assigned 
to be treated was 0.17 standard deviations. As column 2 shows, the effect was particularly pronounced among those 
individuals that also received funding (-0.34 standard deviations). Among those individuals that did not access funding, 
however, the effect was insignificant and smaller (-0.08 standard deviations). Young people outlined during focus group 
discussions that loan repayment modalities were unrealistic and put pressure on them. This was exacerbated by the 
fact that some had used their livestock and assets as collateral, and therefore were at risk of losing the assets if they 
failed to honour the repayment arrangements. Previous literature documents that microfinance may have adverse 
mental health effects (Fernald et al. 2008).
Comparing across modalities 1a and 1b, the evaluation found that the reduction in self-efficacy was particularly 
pronounced within modality 1a. Here, the reduction was both sizable and statistically significant (-0.21 standard 
deviations). In modality 1b, the reduction was 0.09 standard deviations, a small and insignificant effect. Women, as 
column 4 showcases, showed a substantially lower reduction in self-efficacy compared to men. The reduction, -0.06 
standard deviations, was also not statistically significant. 
Table 18: Effect of the entrepreneurship modalities on self-efficacy
Dependent variable
31
In a final step, the evaluation assessed geographic effect heterogeneity. As columns 6, 7 and 8 demonstrate, the 
reduction in self-efficacy was particularly pronounced in Bulawayo and Goromonzi. The reduction was substantively 
meaningful and statistically significant. The effect was not present in the Harare sample. 
Effect of the employability modality on self-confidence 
Table 19 shows the overall effect of the entrepreneurship programme on individuals’ self-confidence. Column 1 reports 
the effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Self-Confidence Index”. 
The index included all questions that measure an individual’s self-assessed self-confidence. Specifically, it included 
questions such as “I am able to express my opinion and discuss sensitive issues” and “I am confident in my future.” The 
evaluation standardised these variables and averaged across them in order to build a comprehensive and comparable 
“Self-Confidence Index”. 
Column 1 of Table 19, demonstrates that the programme had a strong, positive effect on self-confidence. The effect 
was 0.18 standard deviations and was detectable among individuals that attended as well. Here, it was particularly 
strong among those that did not participate in the PTS (0.24 standard deviations.)
Column 3 demonstrates a positive effect on self-confidence as also confirmed among women. Regarding the geographic 
split, the evaluation found that there was a strong and significant increase in self-confidence in Harare (0.29 standard 
deviations). In Bulawayo, the effect was essentially 0.
Table 19: Effect of the employability modality on self-confidence
Dependent variable
32
Effect of t he employability modality on self-efficacy
Table 20 shows the overall effect of the employability programme on individuals’ self-efficacy. Column 1 reports the 
effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Self-Efficacy Index”. The 
index included all questions that measure an individual’s self-assessed self-efficacy. Specifically, it included questions 
such as “I can always manage to solve difficult problems if I try hard enough” and “I am certain that I can accomplish 
my goals”. The evaluation standardised these variables and averaged across them in order to build a comprehensive 
and comparable “Self-Efficacy Index”. 
Column 1 found that the programme had a strongly positive effect on self-efficacy. The increase among individuals 
assigned to be treated was 0.30 standard deviations. As Column 2 shows, the effect was particularly pronounced 
among those individuals that did not participate in the PTS (0.33 standard deviations). Women, as column 3 showcases, 
yielded a highly similar estimate and thus increase in self-efficacy compared to men. The increase was 0.30 standard 
deviations and statistically significant. In a final step, we assessed geographic effect heterogeneity. In columns 4 and 5, 
the results show that the increase was particularly strong in Harare (0.47 standard deviations), while the effect was 
small and insignificant in Bulawayo.
Table 20: Effect of the employability modality on self-efficacy
Dependent variable
33
3.4 Research question 4: To what extent was the modality implemented in a gender-sensitive manner? 
The best evidence for how well the programmes were implemented in a gender-sensitive manner comes from the focus 
group discussions. Although the preponderance of the focus groups was implemented among the entrepreneurship 
participants, the one employment modality focus group indicated that the partner organisation, Restless Development 
(RD), made a substantive effort to recruit and focus on women participants. The entrepreneurship focus group 
discussions also concurred that all of the partner organisations worked hard to recruit female candidates. There was 
less discussion in the focus group discussions of the implementation of the training and the extent to which it was 
directed in a gender-sensitive manner, but female participants across the board described satisfaction in the way the 
programme appeared to be directed toward their needs.
3.5 Research question 5: Is there a higher rate of enterprise start-up and growth rates – particularly amongst young women – 
from project participants than non-participants?
All participants in the entrepreneurship modality had enterprises at the start of the evaluation. However, participants 
in the programme saw substantial growth relative to the control group. As already presented in 3.1.3 and Table 8 
above, all treatment participants saw growth from 594 USD to 624 USD (5%) in monthly revenues relative to the 
control group (declined from 402 USD to 311 USD). Modality 1a saw monthly revenues increasing by 48% (432 USD 
to 367 USD) relative to the control group (decline by 22% from 386 USD to 301 USD) and monthly profits increased 
by 68% (157 USD to 264 USD) compared to the control group (declined by 3% from 141 USD to 137 USD). 
3.6 Research question 6: What are the business start-up and growth rates for young entrepreneurs who received loans, in 
comparison to those who did not receive loans?
Table 21 shows that the subset of participants who received loans saw even more enterprise growth than the 
participants who did not receive loans. The loan recipients had an average growth of 84% in monthly revenues and 
55% in monthly profits, relative to the control group.
Table 21: Business growth rates for young entrepreneurs who received loans
Treated (training and loan) Control
Change in monthly revenue (USD) 384.15 -67.49
Change in monthly profi ts (USD) 85.72 -19.85
3.7 Research question 7: Have participants (or non-participants) – particularly young women – who have accessed wage 
employment retained their jobs for at least 3 months?
Participants in the employability component did not access wage employment at very high rates (11% of treated 
participants were employed, relative to approximately 8% of the control group). There was limited time available 
for young people to transition to full time employment after internship. Of those participants who were employed, 
28% had been employed for more than three months. The length of employment was affected by the project time 
limitations and the timing of the impact evaluation.
Table 22: Wage employment job retention for at least 3 months
Treated Control
Percentage employed 11.1% 8.5%
Average length of employment in months (if currently employed) 2.23 2.5
34
3.8 Research question 8: What were the outcomes of other participants (or non-participants) that secured a job, but did not 
continue to work and why?
Approximately 3% of participants secured an internship following the training. Of these, none were engaged in formal 
employment at the time of the endline survey (the 11% of participants employed at the endline were not placed in 
internships). There are no treated or control participants who got full time jobs and dropped out between baseline 
and endline. This was affected by the project time limitations and the timing of the impact evaluation.
3.9 Research question 9: What is the impact of the integration of Passport to Success (PTS) life skills training on young 
women and men’s economic empowerment programming?
As already presented in 3.2 and Table 12 above, participants in the PTS programme saw greater gains in economic 
empowerment relative to participants who received only the standard training programme. PTS participants saw 
a growth of 0.38 standard deviations in economic empowerment, relative to a gain of 0.12 standard deviations for 
standard training programmes. 
3.10 Research question 10: Does PTS enhance impacts related to income, control over income, access to jobs, and personal 
resiliency, particularly for young women?
PTS appeared to have an enhancing effect for both male and female participants in economic empowerment, income, and 
resilience. In the case of resilience and income, the enhancing effect of PTS was greater for women than for men. Female 
PTS participants saw an increase of 0.31 standard deviations in the income index, relative to a growth of only 0.04 standard 
deviations for male PTS participants. In the case of economic empowerment, PTS appeared to have conveyed greater 
benefits to men (0.40 standard deviations of growth) than women (0.27 standard deviations of growth). It is not clear why this 
discrepancy in gender patterns between different measures of impact was observed.
Resilience
Table 23 shows the effect of the programme on our resilience scale using descriptive statistics. All outcomes, of which 
there are many, were scored on a 0-4-point scale. When comparing the end- to the baseline, we confirmed increases 
across most variables and treatment groups. For example, respondents in the treatment group in modality 1a scored 
3.10 regarding “Adapt to changes” during the baseline (Figure 8). The score increased to 3.32 at the endline. Regarding 
PTS, we confirmed more pronounced increases when compared to the no-PTS group. Again, however, there is the 
caveat that the control group, too, saw these outcomes increase (see Table 24) for a comprehensive overview. 
Figure 8: Resilience (adapt to changes)
35
Table 23: Resilience
36
Modality 1a (Treatment) Modality 1a (Control) Modality 1b (Treatment) Modality 1b (Control) Modality 3 (PTS) Modality 3 (No PTS) Modality 3 (Control) 
Adapts to changes (BL) 3. 1 3.23 3.27 3.16 3.2 3. 17 3.26 
Adapts to changes (EL) 
Secure relationships (BL) 
Secure relationships (EL) 
Optimistic (BL) 
Optimistic (eL) 
Deal with anything (BL) 
Deal with anything (EL) 
Confidence (BL) 
Confidence (EL) 
Humorous (BL) 
Humorous (EL) 
Stronger through stress (BL) 
Stronger through stress (EL) 
Bounces back (BL) 
Bounces back (EL) 
Things happen for a reason (BL) 
Things happen for a reason (EL) 
Effort(BL) 
Effort(EL) 
Overcomes obstacles (BL) 
Overcomes obstacles (EL) 
Does not give up (BL) 
Does not give up (EL) 
Seeks help (BL) 
Seeks help (EL) 
Stays focussed (BL) 
Stays focus (EL) 
Leadership (BL) 
Leadership (EL) 
Not discourages (BL) 
Not discouraged (EL) 
Strength (BL) 
Strengh (EL) 
Makes unpopular decisions (BL) 
Makes unpopular decisions (EL) 
Handles bad feelings (BL) 
Handles bad feelings (EL) 
lntuffion (BL) 
Intuition (EL) 
Purpose (BL) 
Purpose (EL) 
Control over life (BL) 
Control over life (EL) 
Ukes challenges (BL) 
Ukes challenges (EL) 
Works to get goal (BL) 
Works to get goal (EL) 
Pn'des achievement (BL) 
Pride achievement (EL) 
ZIMBAIWE:WORKS 
3.32 
3.3 
3.37 
3.41 
3.63 
3.23 
3.29 
3.33 
3.47 
2.99 
3.15 
2.84 
3.14 
2.88 
3.22 
3.44 
3.61 
3.37 
3.61 
3.42 
3.59 
3.41 
3.51 
3.26 
3.5 
3.12 
3.07 
3.05 
3.28 
3.26 
3.52 
3.33 
3.54 
2.78 
3.01 
3.06 
3.37 
2.87 
2.78 
3.31 
3.66 
3.13 
3.51 
2.8 
2.79 
3.41 
3.63 
3.41 
3.79 
3.47 
3.37 
3.39 
3.5 
3.68 
3.25 
3.35 
3.46 
3.54 
3.04 
3.31 
3.18 
2.99 
3.13 
3.54 
3.45 
3.65 
3.61 
3.69 
3.58 
3.63 
3.41 
3.45 
3.41 
3.38 
3.22 
3.2 
3.33 
3.46 
3.41 
3.65 
3.47 
3.65 
2.82 
3.01 
3.1 
3.57 
2.91 
3.01 
3.48 
3.74 
3.32 
3.62 
2.79 
2.78 
3.53 
3.71 
3.54 
3.75 
3.26 
3.34 
3.34 
3.54 
3.64 
3.26 
3.3 
3.37 
3.44 
3.12 
3.09 
3.14 
3.16 
3.17 
3.3 
3.46 
3.52 
3.51 
3.57 
3.46 
3.58 
3.41 
3.42 
3.34 
3.41 
3.26 
3.1 
3.15 
3.22 
3.25 
3.52 
3.38 
3.53 
2.94 
2.91 
3.2 
3.29 
3.1 
2.9 
3.45 
3.59 
3.33 
3.54 
3.03 
2.77 
3.49 
3.58 
3.43 
3.63 
3.5 
3.25 
3.46 
3.44 
3.65 
3.13 
3.35 
3.42 
3.55 
3.09 
3.23 
3.09 
3.28 
3.13 
3.42 
3.4 
3.69 
3.53 
3.71 
3.46 
3.66 
3.42 
3.47 
3.44 
3.4 
3.2 
3.21 
3.12 
3.47 
3.27 
3.66 
3.32 
3.71 
2.71 
3.01 
303 
3.55 
2.97 
3.14 
3.5 
3.78 
3.41 
3.66 
2.86 
2.96 
3.45 
3.75 
3.43 
3.76 
3.67 
3.35 
3.79 
3.46 
3.9 
3.07 
3.63 
3.17 
3.83 
2.73 
3.63 
2.91 
3.55 
3.8 
3.35 
3.86 
3.36 
3.92 
3.33 
3.71 
3.2 
3.6 
3.27 
3.75 
2.98 
3.58 
2.83 
3.8 
3.09 
3.85 
3.24 
3.87 
2.61 
3.54 
2.9 
3.71 
2.74 
3.38 
3.29 
3.87 
3.2 
3.87 
2.92 
3.59 
3.2 
3.93 
3.35 
3.94 
3.76 
3.32 
3.62 
3.43 
3.88 
3.12 
3.58 
3.17 
3.82 
3.01 
3.51 
3.04 
3.52 
2.95 
3.74 
3.29 
3.83 
3.5 
3.92 
3.48 
3.83 
3.29 
3.59 
3.24 
3.65 
3.13 
3.49 
2.92 
3.74 
3.31 
3.88 
3.34 
3.84 
2.73 
3.4 
2.97 
3.71 
2.64 
3.26 
3.37 
3.89 
3.24 
3.89 
3.27 
3.57 
3.4 
3.89 
3.34 
3.88 
3.62 
3.35 
3.65 
3.52 
3.84 
3.27 
3.55 
3.31 
3.74 
3.08 
3.49 
3.16 
3.48 
3.22 
3.68 
3.37 
3.76 
3.43 
3.8 
3.48 
3.76 
3.47 
3.52 
3.31 
3.68 
3.22 
3.37 
3.13 
3.67 
3.32 
3.76 
3.37 
3.74 
2.87 
3.26 
3.15 
3.59 
2.8 
3.15 
3.41 
3.84 
3.29 
3.77 
3.21 
3.47 
3.46 
3.78 
3.34 
3.8 
Effect of the entrepreneurship modalities on resilience
Table 24 shows the overall effect of the entrepreneurship programme on individuals’ resilience. Column 1 reports the 
effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Resilience Index”. The 
index includes questions on individuals’ self-reported resilience including statements such as “I am able to adapt when 
changes occur”, “Past successes give me confidence in dealing with new challenges and difficulties” and “I feel in control 
of my life”. The evaluation standardised these variables and averaged across them in order to build a comprehensive 
and comparable “Resilience-Index”.
Column 1 shows that resilience, like the aforementioned psychological outcomes, was slightly lower among randomly 
assigned individuals compared to those relegated to the control group (-0.09 standard deviations). Th e effect was 
marginally significant. When examining the effect among participants (column 2), the evaluation found the reduction to 
be predominantly driven by individuals that received funding (-0.29 standard deviations). The effect was insubstantial 
and insignificant among beneficiaries that did not receive funding.
When comparing effects across the modalities, the evaluation found the reduction to be slightly larger among modality 
1b (-0.12 standard deviations), compared to modality 1a (-0.05 standard deviations). Women showed a slightly larger 
reduction in resilience (-0.17 standard deviations), compared to men. Finally, and in line with prior evidence, the effect 
was most negative within the Bulawayo sample. The Harare and Goromonzi samples did not witness a significant 
reduction in resilience.
37
Table 24: Effect of the entrepreneurship modalities on resilience
Effect of the employability modality on resilience
Table 25, shows the overall effect of the employability 
programme on individuals’ resilience. Column 1 reports 
the effect of being randomly assigned to the treatment 
(“Assigned to treat”) on a standardised “Resilience 
Index”. The index included questions on individuals’ 
self-reported resilience including statements such as “I 
am able to adapt when changes occur”, “Past successes 
give me confidence in dealing with new challenges 
and difficulties” and “I feel in control of my life”. The 
evaluation standardised these variables and averaged 
across them in order to build a comprehensive and 
comparable “Resilience-Index”.
Column 1 shows that resilience, like the aforementioned 
psychological outcomes, strongly increased as a result of the programme. Randomly assigned individuals, compared to 
those relegated to the control group, scored 0.28 standard deviations higher on the resilience index. When scrutinising 
the effect among participants (column 2), evidence shows the increase to be driven by individuals that participated in 
PTS (0.41 standard deviations). The effect was also detectable among beneficiaries that did not participate in PTS.
Women showed a slightly larger increase in resilience (0.31 standard deviations), compared to men. Finally, and in line 
with prior evidence, the effect was most pronounced in Harare (0.35 standard deviations).
Dependent variable
“Just participating in the programme helped 
me to be more optimistic about my future. 
I graduated in 2013 and since then I had never 
been invited for an interview. After I received 
training on CV writing and also PTS I have been 
invited to three interviews. Although I did not 
manage to get the posts, I am confident that 
soon I will get my opportunity” 
 (female FGD participant. Bulawayo).
38
, 39
Table 25: Effect of the employability modality on resilience
3.11 Research question 11: Does PTS trigger more transformative changes on participants – particularly young women – than 
other standard training / capacity building components of the project?
PTS appears to create more transformative changes than the standard programme in the subjective measures of 
well-being, particularly on resilience, economic empowerment, risk attitudes, time allocation, and collective action. 
Participants who received PTS also had greater gains in income versus those who received only the standard curriculum, 
but this pattern was not reflected in asset or employment outcomes.
Risk attitudes
Table 26 reports the programme’s effect on risk attitudes using descriptive statistics. Here, however, the evidence 
was mixed. While respondents report greater optimism (see the first two rows), their investment in a risk game, by 
and large, decreases. For example, while 66% invested during the baseline in the treatment group for modality 1a, this 
number was reduced to 52% during the endline. Similar decreases were found in the other treatment groups. 
Dependent variable
40
Figure 9: Investment in risk games
Risk attitudes – scored on an eleven-point scale – painted an unclear picture as well. Individuals were more likely 
to have a taste for risk in investing and careers, but reduced risk behaviour when it came to driving or sport. These 
findings, then, confirmed a healthy re-alignment of risks toward employment and entrepreneurship-related outcomes.
Within the employability modality, there was no significant difference between PTS and non-PTS participants in terms 
of the effect of the programme on risk tolerance attitudes. 
Table 26: Risk attitudes
Time allocation
In Table 27, we report the effect of the programme on time allocation using descriptive statistics. Results here, in 
essence, were negligible. Moreover, there were only two outcomes. When taken at face-value, free time in hours 
mostly decreased in the treatment group (except modality 1b). But, decreases were also detectable within the control 
group. Evidence regarding a statement “life is too busy”, too, was mixed. The effect on time allocation was thus 
unclear. Again, however, we caution that a more realistic and comprehensive analysis was found under econometric 
techniques below. Within the employability modality, there were some slight differences to be observed between PTS 
and non-PTS participants. PTS participants reported having slightly more free time than non-PTS participants, although 
they were also more likely to agree with the statement “life is too busy”.
41
Table 27: Time allocation
Collective action
Table 28 reports the effect of the programme on collective 
action-related outcomes using descriptive statistics. The 
findings here, comparing the end- to the baseline, were 
somewhat mixed. Individuals, by and large, were less 
likely to contact councillors, MPs or agencies. Qualitative 
discussions showed that there was a general apathy among 
young people towards actions deemed to be political. By 
contrast, they were more likely to participate in community 
affairs. Differences (both negative and positive) were more 
pronounced when comparing PTS to no-PTS, attesting to 
the pronounced effects of PTS, which were also confirmed 
above.
Table 28: Collective action
E ffect of the entrepreneurship modalities on risk attitudes
Table 29 shows the overall effect of the entrepreneurship programme on individuals’ willingness to take risks. Column 1 
reports the effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Risk Tolerance 
Index”. The index included all questions that measure an individual’s propensity to take risks. Specifically, it included 
individuals’ self-reported optimism about the future, their willingness to invest money in a hypothetical risky mental 
game, their generalised self-reported risk willingness, and their willingness to take risks in specific situations such as 
driving a car, investing money and playing sports. The evaluation standardised these variables and averaged across them 
in order to build a comprehensive and comparable “Risk Tolerance Index”. 
Column 1, found that the programme had a positive treatment effect on individuals’ willingness to take risks. The 
model compared individuals randomly assigned to the programme to those that were not randomly assigned. The 
difference of 0.13 standard deviations – which is significant and substantively meaningful – was thus the causal effect 
of Z:W. The estimate was robust to the inclusion of control variables and fixed effects. Column 2 shows that the 
“It’s always difficult to talk to politicians or even 
the city council or other agencies which serve 
the public. The main reason is that speaking to 
them will not change anything and in the case of 
politicians you can become a target of violence” 
(female FGD participant, Harare).
42
effect was more pronounced among those individuals that only attended the training, while it was slightly lower (and 
statistically insignificant) among those that also received funding. This may be owing to the fact that taking on a credit 
makes individuals less likely to take risks. 
Columns 3 and 4 show the sample split into modality 1a and 1b and evidence shows that the willingness to take risks 
was higher within modality 1b. Specifically, individuals in this modality scored 0.18 standard deviations higher on the 
risk tolerance index compared to their control group. In modality 1a, the increase was 0.08 standard deviations and 
not statistically significant. Overall, women (column 5) also increased their risk willingness, though the increase was 
slightly lower (0.12 standard deviations), compared to 0.14 standard deviations among men. Columns 6, 7 and 8 show 
geographic differences where an increase in risk tolerance due to the programme were largest in Goromonzi (0.21 
standard deviations) and lowest in Bulawayo, where the evaluation estimated a negative, but insignificant effect. 
Table 29: Effect of the entrepreneurship modalities on risk attitudes
Ef fect of the entrepreneurship modalities on time allocation
Table 30, shows the overall effect of the entrepreneurship programme 
on individuals’ time allocation. Column 1 reports the effect of being 
randomly assigned to the treatment (“Assigned to treat”) on a 
standardised “Free time index”. The index included all questions that 
measured an individual’s free time. Specifically, it included individuals’ 
self-reported free time in hours, individuals’ perceptions about their 
lives being too busy, and individuals’ likelihood to spend free time 
with leisure activities such as watching television, consuming alcohol 
or meeting up with friends. As before, the evaluation standardised 
these variables and averaged across them in order to build a 
comprehensive and comparable “Free Time Index”. 
Dependent variable
“After training we realised that 
there is a lot more we can do with 
our time. Instead of spending time 
drinking beer or talking to friends, 
we now spend more time working 
and trying to engage in activities 
which bring in some money” 
(male FGD participant, Harare).
43
As was to be expected, column 1 estimated that the programme had a negative effect on individuals’ free time. 
The effect on size, however, was small and insignificant. Individuals assigned to the programme were 0.04 standard 
deviations less likely to have free time (broadly measured). This effect translated into the sample of attendees (column 
2). Here, however, the evaluation found that individuals that accessed funding, were more likely to have free time. 
Columns 3 and 4 show that the negative treatment effect was more pronounced within modality 1a. This effect was to 
be expected as there were more pronounced treatment effects across modalities 1a and 1b in Table 30. Specifically, 
modality 1a reduced free time by 0.18 standard deviations (a significant reduction) compared to 0.6 standard deviations 
in modality 1b.
Column 5 shows that the reduction in free time was less pronounced and, indeed, positive among women. The effect 
(0.08 standard deviations), however, was not statistically significant. Regarding geographic differences, the evaluation 
found in columns 6, 7 and 8 that the treatment effects were broadly similar across Harare (-0.07), Bulawayo (-0.02) 
and Goromonzi (-0.05). None of these effects, however, were statistically significant. 
Table 30: Effect of the entrepreneurship modalities on time allocation
 Effect of the employability modality on risk attitude
Table 31 shows the overall effect of the employability modality on individuals’ willingness to take risks. Column 1 
reports the effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Risk Tolerance 
Index”. The index included all questions that measured an individual’s propensity to take risks. Specifically, it included 
individuals’ self-reported optimism about the future, their willingness to invest money in a hypothetical risky mental 
game, their generalised self-reported risk willingness, and their willingness to take risks in specific situations such as 
Dependent variable
44
driving a car, investing money and playing sport. The evaluation standardised these variables and averaged across them 
in order to build a comprehensive and comparable “Risk Tolerance Index”. 
Column 1 shows that the programme had a positive treatment effect on individuals’ willingness to take risks. The 
model compared individuals randomly assigned to the programme or treatment and to the control. The difference of 
0.04 standard deviations, however, was small, though it did represent the causal effect of Z:W. Interestingly, in column 
2 the evaluation found that the effect was less pronounced among those individuals that only attended the training, 
while it was slightly lower (and statistically insignificant) among those that also took part in PTS. Overall, the effect was 
slightly lower among women (column 5: 0.02 standard deviations). Finally, the evaluation found that the increase was 
more pronounced in Bulawayo (0.11), compared to Harare (0.02), though both estimates were insignificant. 
Table 31: Effect of the employability modality on risk attitude
Ef fect of the employability modality on time allocation
Table 32 shows the overall effect of the employability programme on individuals’ time allocation. Column 1 shows 
the effect of being randomly assigned to the treatment (“Assigned to treat”) on a standardised “Free time index”. The 
index included all questions that measured an individual’s free time. Specifically, it included individuals’ self-reported 
free time in hours, individuals’ perceptions about their lives being too busy, and individuals’ likelihood to spend free 
time with leisure activities such as watching television, consuming alcohol or meeting up with friends. As before, the 
evaluation standardised these variables and averaged across them in order to build a comprehensive and comparable 
“Free Time Index”. 
Dependent variable
45
As was to be expected, column 1 estimates that the programme had a negative effect on individuals’ free time. The 
effect size, however, was small and insignificant. Individuals assigned to the programme were 0.08 standard deviations 
less likely to have free time (broadly measured). This effect translated into the sample of attendees (column 2). Here, 
however, the evaluation found that individuals that participated in PTS were significantly less likely to have free time 
(-0.15 standard deviations).
Column 5 demonstrates that the reduction in free time was roughly similar among men and women. The effect for 
women (-0.09 standard deviations), however, was not statistically significant. Regarding geographic differences, the 
evaluation found in columns 4 and 5 that the treatment effects were particularly pronounced in Bulawayo (-0.36), 
compared to Harare (0.07), where the effect was positive.
Table 32: Effect of the employability modality on time allocation
3.12 Research question 12: Does the Passport to Success curriculum improve the relationships of young women and young 
men with each other and with their families?
Table 33 shows the overall effect of the programme on relationships using descriptive statistics. These items, again, 
were scored on a five-point scale. Like in the preceding section, we confirmed a positive increase between the base￾and the endline across most treatment groups. Respondents scored 4.33 points in the treatment group 1a regarding 
compassion during the baseline and 4.61 in the endline (see Figure 10). Similar increases were found across all other 
variables, as well as treatment groups. PTS, once more, outperformed no-PTS. Again, however, the control group, 
too, saw increases. 
Dependent variable
46
Figure 10: Compassion (relationships)
Table 33: Relationships
Effec t of the entrepreneurship modalities on relationships
Table 34 shows the overall effect of the entrepreneurship programme on individuals’ success at maintaining social 
relationships. Column 1 reports the effect of being randomly assigned to the treatment (“Assigned to treat”) on a 
standardised “Relationship Index”. The index included questions on individuals’ self-reported compassion for others, 
their ability to get along with others, their reaction to criticism, and their relation to authorities. The evaluation 
standardised these variables and averaged across them in order to build a comprehensive and comparable “Relationship 
Index”.
Findings confirm that the programme reduced participants’ ability to maintain and develop social relationships. Column 
1 shows estimates that being randomly assigned to the programme reduced the index by 0.2 standard deviations, 
which was a sizable and statistically significant reduction. The reduction was particularly stark among the subset of 
beneficiaries that received funding (-0.47 standard deviations). It was not detectable among the subset of individuals that 
did not receive funding. Qualitative interviews showed that as young people received support from the programme, 
they focused more on growing their businesses, as well as worrying about repaying loans. This meant they were less 
likely to have time to nurture social relationships. In addition, previous literature also documents that microfinance 
may have adverse mental health effects (Fernald et al. 2008).
Comparing modalities 1a and 1b, the evaluation found that the effect was largely comparable across the two. If anything, 
the treatment effect was slightly more negative within modality 1b (-0.21) as compared to modality 1a (-0.18). But, 
these differences were not themselves statistically significant. In relation to gender, the evaluation showed that women 
showed a slightly lower negative treatment effect (-0.19), compared to men. Moreover, the treatment effect was not 
detectable in Harare, where it was even slightly positive. By contrast, as columns 7 and 8 show, the programme was 
associated with a strong decrease in relationships in Bulawayo (-0.89) and Goromonzi (-0.32).
47
Table 34: Effect of the entrepreneurship modalities on relationships
Effect of the employability modality on relationships
Table 35 shows the overall effect of the employability programme on individuals’ success at maintaining social 
relationships. Column 1 reports the effect of being randomly assigned to the treatment (“Assigned to treat”) on 
a standardised “Relationship Index”. The index included questions regarding individuals’ self-reported compassion 
for others, their ability to get along with others, their reaction to criticism, and their relation to authorities. The 
evaluation standardised these variables and averaged across them in order to build a comprehensive and comparable 
“Relationship Index”.
The evaluation found that the programme increased participants’ ability to maintain and develop social relationships. 
In column 1, the evaluation estimated that being randomly assigned to the programme increased the index by 0.29 
standard deviations, which was a sizable and statistically significant effect. The increase was particularly stark among 
the subset of beneficiaries that took part in the PTS (0.61 standard deviations). It was slightly lower among the subset 
of individuals that did not participate (0.23).
Regarding gender, the evaluation noted that women showed a slightly lower treatment effect (0.24), compared to men. 
But, the effect was still strong, positive and significant. The treatment effect was also particularly strong in Harare (0.38 
standard deviations), while it was not detectable in Bulawayo. 
Dependent variable
48
Table 35: Effect of the employability modality on relationships
Dependent variable
49
50
4 CONCLUSIONS AND RECOMMENDATIONS
DOMAIN CONCLUSION RECOMMENDATION
Entrepreneurship i. The programme resulted in individuals 
enhancing their income through their 
business, facilitated access to funding, 
catalysed business savings and resulted in 
business investments. 
i. Future similar programmes should 
build on the evidence of multi-pronged 
entrepreneurship support anchored on 
needs analysis, needs driven capacity 
building, access to finance and mentorship.
ii. The programme combining modality 1a 
and 1b worked slightly better for women, 
and hence has strong prospects. The 
evaluation documents the success of 
the programme in raising the economic 
prospects of female entrepreneurs.
ii. Future similar interventions should 
continue with an affirmative action 
approach1
 to gender transformation2
 with 
specific focus on deliberately targeting and 
supporting more females. 
iii. Increases in income as a result of the 
programme were highest in Harare, 
showing that the city has potential to 
have the highest paying entrepreneurial 
opportunities.
iii. Future similar projects should prioritise 
market linkages as well as access to 
lucrative markets to ensure young 
entrepreneurs get the highest possible 
returns from their activities.
iv. The programme had negative effects on 
individuals’ free time, a negative effect 
on self-efficacy, negligible effects on 
self-confidence or slightly decreased self￾confidence. It further resulted in reduced 
participants’ ability to maintain and 
develop social relationships while having 
marginally negative effects on resilience.
iv. There is a need for further enquiry to 
explore reasons why entrepreneurial 
support may potentially result in negative 
psychological outcomes.
Employability v. The programme had positive effects on 
young people’s access to longer, paid 
internship as well as employment. 
v. Future similar projects should build on 
lessons learnt from the project with specific 
focus on supporting work readiness, 
facilitating access to internships as well as 
support towards accessing employment 
opportunities.
vi. The PTS programme increases young 
people’s employability. 
vi. Explore possibilities of scaling up the PTS 
programme towards more young people as 
a way of preparing them for employment.
vii. The Z:W project had strong positive 
effects on the employability of young 
women and employability opportunities 
are much higher in Harare, compared to 
Bulawayo.
vii. Future similar interventions should 
continue with an affirmative action 
approach to gender transformation with 
specific focus on deliberately targeting and 
supporting more females.
viii. The project had negative effects on 
individuals’ free time, increased young 
people’s willingness to take risks, 
increased young people’s self-confidence 
(especially women), had positive effects 
on self-efficacy, increased participants’ 
ability to maintain and develop social 
relationships while increasing young 
people’s resilience.
viii. Future similar projects should focus on 
addressing the psychosocial needs of young 
people beyond supporting employability. 
4 CONCLUSIONS AND RECOMMENDATIONS
DOMAIN CONCLUSION RECOMMEN
Entrepreneurship Entrepreneurship i. The programme resulted in individuals resulted in individuals i Future sim
enhancing their income through their
bi f ili d f di
i. Future sim
build on t
business, facilitated access to funding,
catalysed business savings and resulted in 
entrepren
needs ana
business investments. building, a
ii. The programme combi i d li 1 ning modality 1a ii Fi
and 1b worked slightly better for women, 
ii. Future sim
continue w
and hence has strong prospects. The
evaluation documents the success of
approach1
evaluation documents the success of specific fo
the programme in raising the economic 
ff l
specific fo
supporting
prospects of female entrepreneurs.
iii. Increases in income as a result of the Increases in income as a result of the iii. Future sim Future sim
 Conclusions and Recommendations
51
Annexes
5 ANNEXES
Annex 1: Case stories
Tichaona’s story (RBCT: Goromonzi)
My name is Tichaona Rusike (male). I am a 27-year-old man who resides in Goromonzi. I have always been 
into market gardening and mostly focused on tomatoes, beans, leafy vegetables and cucumber. Because I am a 
farmer, I did not have business skills to enable me to get the most value from produce and to also ensure that I 
increase production and earn more. A friend of mine from my village introduced me to the Z:W programme and 
encouraged me to register for their trainings as he felt they would assist me with the work I was doing. I was initially 
reluctant to join but later on I decided to try it and see what they were offering. I have never regretted that decision 
to register and join up. Once I signed up, I was initially asked questions about business. We were invited for training 
and the training focused on how to run a business legally, business ethics, financial literacy, record keeping and how 
to construct a business plan. We were also referred to people who provide loans.
After training, I applied for a loan of 1000 USD. The microfinance institution came and assessed me and my 
business. Unfortunately, they approved USD500 USD based on their assessments. With the money I received I 
bought seed and fertilisers to put up irrigation pipes within my garden. Access to water has always been a challenge 
and it made my work more labour intensive. With my water engine and irrigation system functioning, I expanded 
the area planted for cucumber and strawberry. Instead of spending more time watering the garden, I have more 
time to focus on marketing. I have also started the process of registering for tax because whenever I supply my 
produce to big supermarkets like Pick n Pay, they withhold money for tax.
I have managed to repay three quarters of the money I borrowed and I am on course to complete repaying. I 
am hoping that with expanded production and access to lucrative markets in Harare, I will be able to further 
increase the production scale and even move to a bigger plot. Although I am still focusing on repaying my loan and 
expanding production, I am optimistic that my involvement with the programme will change my life. I now know 
of branding and packaging which I am investing in. I also think that I am doing well. My peers and others in the 
community respect me and my opinions more.
While I am very grateful for all the support that was provided, I think there is need to ensure that the conditions 
for getting loans should be relaxed a bit especially the need for collateral and also that there should be technical 
support for specific business. I am in the market gardening business and I think market gardening technical support 
will be more welcome to compliment business management training.
Ottilia’s story (PROWEB: Goromonzi)
My name is Ottilia Chinake (female). I am a 30-year-old mother of two who resides in Goromonzi. I am divorced 
and I have a tuckshop which sells groceries as well as stationeries. I have always known Z:W but did not have 
in-depth information regarding what they do in supporting young people. When I heard by our youth chairperson 
that they were offering training for young people with businesses, I decided to join.
Although my business was doing fairly well by my own standards, I was not keeping proper records and not 
calculating profits or losses. Sometimes I also mixed business money with my personal funds which made it difficult 
to measure the business’ performance. After registering for training PROWEB trained us on record keeping, 
marketing and expanding my business. These are all skills that were very necessary for my business. After training, 
we were referred to a microfinance institution, which provides loans.
I applied for a loan of 400 USD and it was approved. In order to ensure that the money for the loan would work 
in a way I could track, I started keeping records for my business. I also started giving myself a salary as opposed 
to when I never used to. At the end of each month, I now do reconciliations and I can now see the products that 
bring more income and also business performance. 
I am repaying the money that I received and my business is growing. As a woman, it is important that I can be able 
to provide for my children and also contribute towards household subsistence. Already there is some stigma in being 
a single mother, but that is worse if you do not earn a living. I think that I have earned some respect at home and 
within the community. The programme was very helpful and gave me skills to be better in doing the things I enjoy 
doing. I think it will be important if the programme tries to involve more women, not only during trainings, but in 
ensuring they receive loans to enhance their activities.
Fisiwe’s story (RBCT: Bulawayo)
My name is Fisiwe Juba (female). I am a 26-year-old lady from Luveve (Bulawayo). I am into buying and selling. 
I usually sell clothes, kitchenware, groceries and other products from South Africa and from Botswana. I usually 
have big orders and did not struggle to sell my stuff. However, I always had challenges with people paying since I 
sometimes sold my stuff on credit. Sometimes I did not record the number of people who owed me money nor the 
amounts they would owe. A friend of mine told me about the Z:W programme and I decided to attend especially 
as she said they would be linking us to microfinance institutions after training. 
I registered for the training and attended all the sessions. I had been running my business for seven years and 
thought I knew everything. During the training I discovered that I knew very little about running a business 
successfully. I did not keep records and there were times I had to put my personal money into the business. There 
were also other very important aspects of the training especially marketing. I usually focused on my personal 
networks and people from Church. However, after training I realised that there was a lot of potential especially 
with social media. Another important thing relates to linkages with microfinance institutions. Previously I did not 
know where to access loans and neither did I know the prerequisites for accessing a loan.
I did not apply for a loan as my business was not structured in ways that would make me eligible for a loan. 
However, I now know what’s required for me to get a loan and have set out to try and do so. I am now keeping 
records and know people who owe me. This has made it easy for me to make follow ups and remind people to 
pay. I am no longer struggling to raise funds to buy stuff to resell and prospects of business growth are good. I also 
got a stall at Unity Village and I now have a permanent physical space where I sell my products.
The prospects for enhancing my business are good. However, I struggled a bit when the Government banned 
importation of groceries including cooking oil. These are always fast moving and the ban resulted in challenges 
within the business. It was unfair from the Government and sometimes it may be helpful for the Government to 
engage us before making these decisions. It is frustrating that after being trained and having the motivation to work 
hard, you are then restricted by Government policy.
Freddy’s story (PROWEB, Harare)
My name is Freddy Chigodora (male). I am a 30-year-old male based in Harare. I am a carpenter by trade and 
am into furniture manufacturing. Business has been good, but it has also had some challenges which means there 
is a lot of room for things to be better. As I am a carpenter by trade, my strengths have always been around 
manufacturing, but I also realised that there is a lot more to a successful business. I have always worked alone 
to try to do the manufacturing, buying of raw materials, marketing and following up on debtors. The people from 
Z:W came here to our complex because they had a programme for entrepreneurs. What they said convinced me 
because the training they were offering was very necessary for my business.
I attended the training and what I remember very well was the guidance on how to structure your business in terms 
of what paperwork should be in place, what you should do if you want to recruit employees, how to keep financial 
records, how to market your products and how to follow up on people who owe you money without straining 
relations.
52
These things were very important because, as I said, I am a carpenter and my skill is in making furniture. However, 
instead of getting people to assist me, I tried to do everything by myself and this affected my business. Getting 
raw materials would take me an average of two days per week and that meant I had limited time to actually 
manufacture furniture. I also did not do any marketing besides selling to people who come here because they know 
this is where furniture is found. 
After the training I set out to ensure my business had some structure. I now engage some people who assist me on 
short term basis. I have also started discussing with Teecherz Furnitures with a view to supply them with furniture. 
I also now have business cards and flyers which I have engaged a flyer distribution company to assist in distributing. 
I now have a bank account so it’s very easy to keep records of my finances. 
I did not apply for a loan because I felt what my business needed was organisation not more finances. Maybe with 
time I will realise that I need more capital and when that time comes I now know the prerequisites for accessing it. 
The training was very important, but there are also other things that Z:W can try to assist with. Here our biggest 
challenge is workspace. Council does not want to engage us and there are other informal groups that harass and 
extort money from us. It will be helpful if we can get support in terms of ensuring we feel secure in our workspaces.
Annex 2: Scope of work
Annex 3: Baseline questionnaire
Annex 4: Endline questionnaire
(Footnotes)
1 Making deliberate efforts to ensure that more women participate and benefit;
2 Explicitly seeks to redefine and transform gender norms and relationships to redress existing inequalities. 
Approaches will include engaging men, households and communities.
53
(Footnotes)
1 Making deliberate efforts to ensure that more women participate and benefit;
2 Explicitly seeks to redefine and transform gender norms and relationships to redress existing inequalities.
Approaches will include engaging men, households and communities.
53