Country:

Ethiopia

Project Name:

Women Development Initiatives

Project ID:

P050342

L/C/TF Number(s):

IDA-34070,IDA-3407A,ITAL-24926

ICR Date:

01/31/2007

ICR Type:

Core ICR

Lending Instrument:

LIL

Borrower:

FEDERAL DEMOCRATIC REPUBLIC OF ETHIOPIA

Original Total Commitment:

XDR 3.8M

Disbursed Amount:

XDR 3.7M

Environmental Category:C

Implementing Agencies

Ethiopia Women Affairs Office

Cofinanciers and Other External Partners

2. Key Dates

Process

Date

Process

Original Date

Revised / Actual Date(s)

Concept Review:

04/30/1998

Effectiveness:

03/22/2001

03/22/2001

Appraisal:

03/17/2000

Restructuring(s):

Approval:

07/27/2000

Mid-term Review:

10/31/2003

Closing:

06/30/2005

06/30/2006

3. Ratings Summary

3.1 Performance Rating by ICR

Outcomes:

   Satisfactory

Risk to Development Outcome:

   Substantial

Bank Performance:

   Satisfactory

Borrower Performance:

   Moderately Satisfactory

3.2 Quality at Entry and Implementation Performance Indicators

Implementation Performance

Indicators

QAG Assessments (if any)

Rating:

Potential Problem Project at any time (Yes/No):

No

Quality at Entry (QEA):

None

Problem Project at any time (Yes/No):

Yes

Quality of Supervision (QSA):

None

DO rating before Closing/Inactive status:

Satisfactory

4. Sector and Theme Codes

Original

Actual

Sector Code (as % of total Bank financing)

General public administration sector

25  

25  

Other social services

75  

75  

Original Priority

Actual Priority

Theme Code (Primary/Secondary)

Poverty strategy, analysis and monitoring

   Secondary

   Secondary

Participation and civic engagement

   Secondary

   Secondary

Gender

   Primary

   Primary

5. Bank Staff

Positions

At ICR

At Approval

Vice President:

Gobind T. Nankani

Callisto E. Madavo

Country Director:

Ishac Diwan

Oey Astra Meesook

Sector Manager:

Laura Frigenti

Arvil Van Adams

Project Team Leader:

Endashaw Tadesse Gossa

Christine Lao Pena

ICR Team Leader:

Adyline Waafas Ofosu-Amaah

ICR Primary Author:

Adyline Waafas Ofosu-Amaah

6. Project Context, Development Objectives and Design

(this section is descriptive, taken from other documents, e.g., PAD/ISR, not evaluative)

6.1 Context at Appraisal

(brief summary of country macroeconomic and structural/sector background, rationale for Bank assistance)

At the time of appraisal, the operative GoE development strategy for gender issues was the 1993 National Policy for Women in Ethiopia, which identified women's low participation in development initiatives, gender disparity in legal literacy, and weak institutional capacity on gender issues as key concerns. One of the main objectives of the Bank's strategy of assistance to Ethiopia in 2000 was to promote gender-balanced poverty reduction through job creation, agricultural inputs, and food security to assist the poorest groups. Due to a combination of social, cultural, and legal factors, women were disproportionately represented among the poorest groups, with relatively low human development indicators and limited access to economic resources. A joint World Bank/GoE study had identified institutional capacity to address gender issues, women's relatively lower participation in development interventions, and gender disparity in legal literacy as major constraints to successful implementation of gender policy.

6.2 Original Project Development Objectives (PDO) and Key Indicators (as approved)

This Learning and Innovation Loan (LIL) aimed to test methodologies to enhance the social and economic welfare of households in selected poor districts by improving and increasing women's opportunities through increased skills, productivity, and income. The key indicators consisted of:  (1) Input indicators (e.g., number of IEC campaigns launched); (2) Process indicators (e.g., administrative costs as percentage of project costs); and (3) Output/Outcome indicators (e.g., number of women's groups (WGs) strengthened). (See Annexes 1, 4, and 5).

6.3 Revised PDO and Key Indicators (as approved by original approving authority), and reasons/justification

There was no revision of the PDO.

6.4 Main Beneficiaries, original and revised

(briefly describe the "primary target group" identified in the PAD and as captured in the PDO, as well as any other individuals and organizations expected to benefit from the project)

The primary beneficiaries were poor women members of WGs.  The secondary beneficiaries were members of the communities in which WGs were formed, and the Intermediaries, Facilitators, and government institutions eligible to benefit from the institutional strengthening component.

6.5 Original Components (as approved)

• Component 1. Grassroots Initiative Fund (GIF): A demand-driven financing window to provide financial, supervisory, and capacity-building support to enable poor women to organize themselves around productive activities.  For income-generation activities, WGs were required to invest up to 10 percent of their own resources (5% cash, 5% in-kind).
• Component 2. Institutional Strengthening: A capacity-building window designed to raise awareness of both men and women about economic, social, and legal issues; promote IEC on gender and development, legal literacy, nutrition, harmful traditional practices and HIV/AIDS, family planning, and environment; and carry out participatory monitoring and evaluation (M&E).
• Component 3. Administrative/Management: This component covered administrative, equipment, and recurrent costs (salaries of non-civil service staff and operating costs).

6.6 Revised Components

None.

6.7 Other significant changes

(in design, scope and scale, implementation arrangements and schedule, and funding allocations)

As a LIL, the WDIP focused on experimentation and learning. As part of its design and during implementation, WDIP tested different methodologies for increasing women's socio-economic empowerment and addressing logistical and capacity challenges. This approach resulted in changes in business group types, group sizes, business modalities, sequencing of training, and IEC delivery mechanisms (see also Section 7.2). After experiencing a rapid turnover in performance following the start-up and first phase delays and accelerating implementation after the MTR, the project was extended by one year to allow it sufficient time to consolidate its implementation progress and apply the lessons of a South-South learning exposure visit and other experiences.

7. Key Factors Affecting Implementation and Outcomes

7.1 Project Preparation, Design and Quality at Entry

(including whether lessons of earlier operations were taken into account, risks and their mitigations identified, and adequacy of participatory processes, as applicable)

The WDIP's design incorporated lessons from Bank-financed projects within Ethiopia (e.g., ESRDF) and WID projects elsewhere in Africa (The Gambia and Côte d'Ivoire). These lessons include:

1. Capacity building, not only for the primary beneficiaries, but for other project stakeholders as well, to build their supervision and implementation capacity.
2. IEC on strategic topics for project beneficiaries and stakeholders, in recognition of the equal importance of socio-economic empowerment and the inclusion of community members and leaders to secure household and community support.
3. Emphasis on a participatory M&E system to track project outcomes and measure impacts.
4. Implementation arrangements drawn from experiences with community participation.
5. Community participation and contribution as a mechanism to engender commitment and project success.

The WDIP's components and sub-components reflect these lessons, and decisions taken at the project launch workshop (regarding WDIP staff training, annual work program planning, and preparation of training manuals) and incorporated in the project implementation manual (PIM) laid the foundation for building supervision and implementation capacity.

7.2 Implementation

(including any project changes/restructuring, mid-term review, Project at Risk status, and actions taken, as applicable)

The project was implemented in three phases. During each phase, the project's geographical coverage was expanded, and, in keeping with the LIL concept, lessons from previous phases were applied. Some of the methodologies tested and modified during implementation are described below:

• Phase 1 (Addis Ababa, Afar, Amhara, and Dire Dawa): One of the first modifications made, given the poverty levels and low socio-economic status of women in Ethiopia, was to change the 10% cash contribution to a flexible 5% cash contribution and 5% in-kind contribution, and to allow cash contributions to be paid in installments. The PAD and the PIM had specified maximum group size of 30, and due to high demand for the project and the GoE's desire to include as many beneficiaries as possible to promote equity, WGs consisting of 30 members were approved, creating problems such as drop outs and limited profitability (see Annex 5). Challenges faced during this phase included: staff turnover; intermittent and often weak support from the Women's Affairs Offices, partially because of lack of capacity to prepare GIF proposals according to established standards and lack of clarity about roles and responsibilities; and barriers to transmission and reception of IEC messages due to illiteracy. Actions taken included communication mechanisms to clarify roles, capacity building exercises, and recommendations to the Borrower to include a literacy module (this recommendation was not implemented).
• Phase 2 (Harari, Oromia, Somali, SNNPR, and Tigray). The project initiated regular intra-regional learning exchanges, and organized learning exposure visits by project staff to South Asia. The project also introduced savings and credit schemes and federations of WGs. Challenges faced included lack of skills to prepare viable business proposals; lack of diversity in the business models adopted, which reduced the profitability of businesses; group dynamics within WGs; geographic dispersion of projects due to GoE's desire to promote equity; and lack of capable intermediaries in remote areas. Actions undertaken by the project included providing grassroots management training (GMT) prior to preparation of business proposals; diversification of group modalities; vertical and horizontal linkages such as connecting WGs to institutional markets (schools, hospitals, and the garment industry); and tapping Intermediaries from NGOs, academic institutions, student volunteers, and government agencies in order to expand WG's access to this support mechanism.
• Phase 3 (Benshangul and Gambella). The main challenge included the possibility of business failure due to the late start in these regions. The actions taken included setting up a task force of RCU staff from earlier phase regions to assist the RCUs of these two regions.

Thus, implementation was characterized by learning and innovation.  The project began introducing modifications as early as project launch, and continuously sought advice and acted on recommendations, especially from the MTR stakeholder workshop and a Quality Enhancement Review (QER). A National Fair was organized in the first quarter of 2006 to showcase project achievements.

7.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization

The emphasis on a comprehensive M&E design came from previous experiences with inadequate M&E systems in The Gambia and Côte d'Ivoire projects. The WDIP's M&E system was designed to be participatory in order to develop the Borrower's institutional capacity and to promote civil society involvement. Most of the specified indicators were to be collected periodically by the Facilitators and Intermediaries as part of their support to the WGs. The extensive information collected helped to evaluate the 623 WGs' progress in terms of profitability, conversion into cooperatives, and readiness to graduate. Besides regular monitoring, the project carried out two rounds of rapid assessments (the first covering a sample of WGs in 6 regions, and the second covering all regions) and exit studies covering all regions. Given the scale of the operation, instead of covering all regions, data on a random sample would have been adequate. Three impact evaluations were conducted. The first covered 1,000 randomly-sampled women. The second and third, both undertaken in 2006, covered 136 and 1,220 randomly-sampled beneficiaries respectively.  The absence of a control group in the latter two studies poses difficulties for attributing the positive impacts recorded to the project.

7.4 Safeguard and Fiduciary Compliance

(focusing on issues and their resolution, as applicable)

The project was rated a category "c."  The financial management system produced timely and accurate reports. There were some delays in the GoE's provision of counterpart funds, resulting in financing gaps that were addressed following reminders from the Bank to MOFED, WAO, and the WDIP Board and Steering Committee.  Internal and external audits pointed to documentation gaps and the need to ensure internal controls.

7.5 Post-completion Operation/Next Phase

(including transition arrangement to post-completion operation of investments financed by present operation, Operation & Maintenance arrangements, sustaining reforms and institutional capacity, and next phase/follow-up operation, if applicable)

No post-completion operation was planned at the time of the ICR. Aide-Mémoires since 2004 (i.e., one year prior to the original closing date) indicated that the Bank and the Borrower discussed the need for transition arrangements, but no action was taken.  The existing WGs continue to operate their businesses, although there are some concerns about their viability without the project's continued support, raising questions sustainability of the GIF's outcomes (Annex 4). There is high demand and enthusiasm from communities and local government entities, as shown by continued formation of WGs and savings schemes in anticipation of WDIP-type support.

8. Assessment of Outcomes

8.1 Relevance of Objectives, Design and Implementation

(to current country and global priorities, and Bank assistance strategy)

Objectives: The continued relevance of the project's objectives is reflected in the fact that in May 2006, the GoE adopted a five-year Plan for Accelerated and Sustained Development to End Poverty (PASDEP) that identified gender as one of eight pillars. Preparation of a National Action Plan on Gender also highlights the priority need to focus on reducing risks and disadvantages that disproportionately affect women. The Bank's assistance to Ethiopia targets institution building and governance reform to accelerate pro-poor growth.

Design: Given this project's novelty in Ethiopia, the identified needs of the target beneficiaries for capacity, skills, and income, and the government's readiness to address constraints faced by women in participating in the development process, the project's design was highly relevant. This is because the project emphasized capacity building for individual beneficiaries through grassroots management training on business skills, and institutional capacity building for project implementation and monitoring at the grassroots level through skills development and on-the-job training for project staff and stakeholders (Annex 4).

Implementation:  As a LIL, the project's emphasis on experimenting and using project-specific lessons to make adjustments during implementation was also highly relevant. Annex 4 provides details on the following implementation issues and the methodologies tested:

1. of promoting women's economic and social advancement: Successful introduction of new group modalities, modification of group size, initiation of savings and credit schemes, establishment of unions, federations, and cooperatives, and facilitation of inter-group linkages.
2. Capacity building through grassroots management training: Revision of the sequencing of GMT to ensure WGs' effective application of acquired skills.
3. Use of Facilitators and Intermediaries:  Introduction of innovative ways to support Intermediaries and Facilitators through links with academic and government institutions.
4. IEC: Attempt at different promotion, dissemination, and delivery mechanisms that can be easily understood; and advocating the importance of a literacy module.

There were minor shortcomings such as the design of the final impact evaluation (Section 7.3) and the high administrative costs (Annex 3). Some of these shortcomings stemmed from the GoE's desire for nation-wide coverage so that as many poor women as possible could benefit.

8.2 Achievement of Project Development Objectives

(including brief discussion of causal linkages between outputs and outcomes, with details on outputs in Annex 4)

The project was highly successful in achieving its development objectives. The WDIP resulted in several positive social, economic, and livelihood outcomes for women. In particular, the project effectively:

1. for enhancing women's socio-economic welfare.  These methodologies include grassroots mobilization, group-based businesses, formation of economic activity-based linkages, promoting business relationships between the WGs and their Intermediaries and Facilitators, promoting microfinance and private sector linkages, cost-effective IEC delivery mechanisms, and promotion of peer learning (Section 7.2).
2. Reached 11,675 poor women in poor districts. About 60% of the members of the 627 WGs formed were illiterate, 22% were single heads of households, 40% had no income prior to joining the project, and 58% were located in rural areas. (Annex 4).
3. Built capacity among beneficiaries and stakeholders.  Through 36 skills development courses, the WDIP trained experts who in turn provided IEC courses to over 20,000 male and female beneficiaries, including 5,653 women who improved their skills and relied on this training to engage in WDIP-funded businesses.
4. Enhanced economic welfare through increases in incomes, food security, and asset accumulation (including land purchases and household assets). Internal rates of return on the WGs' businesses ranged from 20-25% with variations by location, business type, and group size (Annex 5). This increase in incomes brought about improvements in food and nutrition (for 82% of the women), access to health services (78%), and access to education (73%).

Improved social welfare through skill enhancement (85% of women), enhanced roles in household decision making (97%), community participation (83%), and self-reliance and confidence (97%). As a result of their incomes, 97% of women reported changes in the attitudes of community and family members towards them (Annex 4).

8.3 Efficiency

(Net Present Value/Economic Rate of Return, cost effectiveness, e.g., unit rate norms, least cost, and comparisons; and Financial Rate of Return)

Based on the information available for 574 WGs (92 percent), the overall IRR for this component was estimated to be 25 percent. While cost-analysis shows variations in the IRR based on group modality, group size, location, and business type, all cohorts show IRR of above the 10 percent minimum required for Bank projects (Annex 5, Table 5.1).

8.4 Justification of Overall Outcome Rating

(combining relevance, achievement of PDOs, and efficiency)

Rating: Satisfactory

The overall rating is satisfactory. The project recorded many socio-economic gains for women (Annex 4) and a 25 percent IRR for the GIF component (implying 10 percent IRR overall). These are positive outcomes for women's empowerment and a worthwhile investment for the Borrower and the Bank (Annex 5). Through trying different approaches, the project created and applied lessons that are also relevant at the country and institutional levels (Sections 7 and 11).

8.5 Overarching Themes, Other Outcomes and Impacts

(if any, where not previously covered or to amplify discussion above)

(a) Poverty Impacts, Gender Aspects, and Social Development

Evidence from the exit studies, rapid assessments, and impact studies indicate that the project's economic impacts had spillover effects on social and human development indicators, as described in Annex 4.  Health impacts include increased awareness of, and attitudinal changes about, family planning, HIV/AIDS, and harmful traditional practices (Annex 4, Box 4.1). Gender impacts include improvements in women's economic independence that translated into improved intra-household decision making and greater awareness of legal rights (Annex 4, Table 4.5).  Social impacts include improved social status as evidenced by community members' positive attitudes towards the women and appreciation of their economic contributions to the household and the community (Annex 4, Table 4.6).

(b) Institutional Change/Strengthening
(particularly with reference to impacts on longer-term capacity and institutional development)

The WDIP established a comprehensive institutional infrastructure at central and regional levels.  The institutional structure was a team effort comprising the CCU, eleven RCUs, WGs, Intermediaries, and Facilitators.  Capacity development achievements include the training of 193 Facilitators, 215 Intermediaries, and 322 experts on IEC topics (Table 4.3).  A women's development resources and documentation center (for electronic and hard copy WDIP documents as well as related materials from different organizations and the South-South Learning study tour reports) was also established, and continues to be independently run by the EWDF.  To the extent that these skills continue to be applied within Ethiopia on other projects, the impact on long-term capacity and institutional development is likely to be considerable.

(c) Other Unintended Outcomes and Impacts (positive or negative, if any)

None.

8.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops

(optional for Core ICR, required for ILI, details in annexes)

The team discussed the type of ICR to be conducted and it was agreed that because there were no plans for a follow-up operation, a Core ICR would be prepared.

9. Assessment of Risk to Development Outcome

Rating: Substantial

There are two major risks that the WDIP's development outcomes will not be sustained:

1. This is a significant risk. The likelihood that it will materialize depends on realization of the following risk mitigation mechanisms being set up (details in Annex 4, "sustainability of GIF outcomes"):

2.Scaling up and follow-up beyond project closing. This is a high risk.  The likelihood that it will materialize depends on the adoption in the near future of concrete plans to ensure that the project's economic and social impacts in the communities will continue to be sustained for a long time.  The withdrawal of funding implies limited institutional ownership and support. The lack of a follow-up project may lead to loss of momentum, loss of the institutional capacity built, and diminished confidence in projects of this nature.

10. Assessment of Bank and Borrower Performance

(relating to design, implementation and outcome issues)

10.1 Bank

(a) Bank Performance in Ensuring Quality at Entry
(i.e., performance through lending phase)

Rating: Satisfactory

The WDIP's objectives, approach, and design were strategically relevant, technically sound, and designed to address critical needs that were embedded in the country's overall development strategy.  The project's poverty, social, and gender aspects were sound and highlighted cross-sectoral linkages in health, education, agriculture, and private sector development.  The policy and institutional dimensions were comprehensive.  Fiduciary arrangements were standard practice that relied on annual work program budgets (AWPBs), a project implementation manual, and training of WDIP on financial management and procurement.  There were two minor shortcomings relating to complexity in the following areas:

• Design. The PAD adequately identified and rated as substantial six risks: (1) ability to find appropriate sustainable activities for WGs; (2) capacity of these groups to produce required savings; (3) availability of adequate Intermediaries; (4) willingness of communities to permit women to participate in the project; (5) inability of poor households to find time to participate in project activities; and (6) ability to attract adequately trained staff.  The design included detailed mitigation measures, such as incentives for group-based operation and project staff.  These measures effectively addressed the risks, but they also contributed to complex arrangements at Federal and Regional levels.
• The administrative component. Administrative costs amounted to 35 percent of overall project costs (Annex 3, Table 3.1).  This was possibly due to a combination of factors – the Bank's underestimation of the administrative costs, and the GoE's insistence on rolling out the project nation-wide (11 regions) to ensure equitable distribution of benefits. However, the Bank correctly recognized at appraisal that administrative costs of projects of this kind are usually higher than the norm.

At the beginning of the project, a 3-day project launch workshop was conducted to familiarize the project's stakeholders with the design concept.  The practical suggestions on implementation arrangements that emerged at this workshop were incorporated into the project's implementation plan. This is an indication of the inclusive process adopted to engage all project participants in order to ensure good quality at entry and minimize implementation challenges.

(b) Quality of Supervision
(including of fiduciary and safeguards policies)

Rating: Satisfactory

Bank supervision was regular, well organized, and participatory. Implementation bottlenecks were identified and successfully addressed. The MTR involved all stakeholders, thoroughly assessed implementation progress and challenges, and recommended a number of corrective measures. Supervision findings were expressed with candor and on a timely basis. Supervision resources were also adequate, with a good mix of expertise. The Bank team ensured that representatives of the co-financier (Government of Italy) participated in supervision missions and engaged in dialogue with Government. The Bank team created a learning environment, strengthened client institutional capacity through transfer of knowledge to and between the project's stakeholders, and fostered a learning-by-doing approach to enhance social capital through WGs.

There were three minor shortcomings relating to the following areas of importance for a LIL.  Monitoring of results and impacts:  The project's focus on results led to an emphasis on monitoring impacts.  However, the lack of a counterfactual in the impact study with the largest and most representative sample creates difficulties for directly attributing the very positive impacts to WDIP (Annex 4). Documenting lessons for wider dissemination: From the supervision reports and the MTR, it is clear that the project's many lessons were applied during implementation.  However, there was no institutional mechanism for documenting and disseminating these lessons beyond these project-specific reports so that they can have Bank-wide benefits. Transition arrangements: An important objective of a LIL is to experiment, learn, and pilot prior to a larger-scale operation.  Bank management was not proactive in conveying in clear and unequivocal terms the need to make concrete plans early enough to ensure scaling up into a larger-scale operation.

(c) Justification of Rating for Overall Bank Performance

Rating: Satisfactory

As detailed in Annex 7, Bank staff put a high level of effort into this project in terms of appraisal, design, and supervision. This level of effort was comparable, if not better than, a regular, large-scale operation, and ensured good quality design and supervision. Bank performance with regard to developing transition arrangements with the GoE did not indicate a similar level of commitment, and because this was a LIL, the Bank's overall performance is rated moderately satisfactory.

10.2 Borrower

(a) Government Performance

Rating: Moderately Satisfactory

The main elements of Government's performance, described in detail in Annex 7, include:

• High level of government ownership and commitment to the PDO.
• Satisfactory readiness for implementation and relations with the co-financier, IC.
• Moderately satisfactory performance on fiduciary arrangements and adequacy of M&E.
• Moderately unsatisfactory performance on contribution of counterpart funds, nation-wide dissemination of project achievements, and institutional structures.

Inadequate transition arrangements. The lack of government interest and commitment jeopardized the project's future. Staff attrition resulted from the delay in government's decision for a possible follow-up operation.  At the time of the ICR mission the lack of a budget to support the project staff beyond three months (subsequently extended to six months) of project closing was a matter of concern for the project's stakeholders.

(b) Implementing Agency or Agencies Performance

Rating: Satisfactory

Implementing Agency

Performance

Ethiopia Women Affairs Office

The main elements, also described in detail in Annex 7, include:

• High level of commitment and positive enabling environment, which resulted in the rapid execution of the MTR recommendations and brought about successful implementation. Confidence in the ability and commitment of the IA facilitated the decision to extend the closing date.
• Satisfactory stakeholder consultations and relations with donors.
• Proactive and effective resolution of implementation issues.
• Timely and adequate procurement practices and processes.
• Diligence in implementing M&E plans, despite capacity issues.

Lack of proactivity and effectiveness with transition arrangements. The IA should have tried to involve MoWA and MoFED earlier in the development of concepts and plans for project follow-up.

(c) Justification of Rating for Overall Borrower Performance

Rating: Moderately Satisfactory

Government performance was moderately satisfactory because of some shortcomings relating to timely infusion of counterpart funds and financial management issues. The funding delays resulted in financing gaps that required reminders from the Bank. The inadequate transition arrangements pose a significant threat that the positive outcomes and impacts of the operation will not be sustained (Annex 7). The lack of transition arrangements is especially worrisome in light of: 1) the project's successful outcomes; 2) the high demand for this type of project; and 3) the original intent to pilot this project with the objective of scaling up (Annex 4).

11. Lessons Learned

(both project-specific and of wide general application)

This experimental project set out to learn and innovate on ways to improve women's socio-economic status.  In principle, this is a sound and valid approach which generated many lessons that are relevant at the project, country, and institutional levels described below.

1. Project Level Lessons

• Projects initiating group-based businesses have to ensure: (i) group coordination; (ii) a sense of commitment to and ownership of the group business; (iii) flexibility to diversify operations; and (iv) adequate funding based on business type and group size.
• Effective project implementation necessitates linkages with both public and private sector services because of the limited scope of services and technical skills that can be provided by local implementing bodies and/or other government agencies.

• Sustained operation and profitability of WGs call for a more carefully considered plan consisting of: (i) an exit strategy for Facilitators in the form of an incentive-based fee system; (ii) phased cash contributions by WGs to cover technical support services; (iii) market and institutional linkages between the WGs and appropriate formal institutions; (iv) plans for WGs' graduation to cooperative or other legal status; and (v) diversification of WGs' business portfolios.
• Tracking of a project's strengths and weaknesses requires a simple but comprehensive and participatory monitoring system; and attribution of the impacts to a project requires a sound impact evaluation methodology.

2. Country-Level

• It is possible to empower poor, illiterate women socially and economically through small-scale and group-based micro-business initiatives. Of the different types of group-based initiatives that this project tested, individually-operated and hybrid group modalities give fast return and are more profitable, less conflict prone, and easy to manage, and therefore the most appropriate mechanism for empowerment.
• A prerequisite for women's empowerment is capacity enhancement through basic literacy, business skills development, legal literacy, and training on gender-relevant themes. Furthermore, the sequencing of delivery of capacity-enhancing topics is important for effective application of learning.
• Investing in and sharing lessons for a broad range of stakeholders are important internally (e.g., the intra-regional exchanges between RCUs); externally (e.g., the South-South exchange visits by WDIP staff to South Asia); and institutionally (e.g., through good practice/lessons learned notes for internal World Bank/WDIP staff consumption).

• A nation-wide pilot is not desirable from both financial and administrative perspectives, as it requires a complex and costly institutional set up.

3. Institutional Level Lessons

• LIL methodology: Learning-by-doing is the key methodology for a LIL. This methodology was used very effectively and was vital for the project's success.  However, the value of a LIL goes beyond the project's success with testing, innovating and learning – it lies in its scaling up. Such scaling up requires a long-term investment.

Transition arrangements are essential for a LIL: Differing development priorities on the part of both government and the Bank resulted in discontinuation of support for this project, threatening to undermine the sustainability of outcomes. The lack of transition arrangements also raises questions about the partners' responsiveness to the high expectations raised by this project, commitment to addressing women's economic empowerment issues, and appreciation of the value of the LIL as a learning tool.

12. Comments on Issues Raised by Borrower/Implementing Agencies/Partners

(a) Borrower/implementing agencies

In addition to factual and editorial comments, two issues were raised in the Borrower's comments (Annex 10), relating to the reasons for the challenges faced during the first phase and the delays in the provision of project funds. First, the comments confirmed that the challenges faced during the first phase were structural ones, which this ICR refers to in Section 7.2 as capacity issues and the lack of clarity about roles and responsibility. Second, the project documentation does not indicate any delays in the transfer of IDA funds.

(b) Cofinanciers

The Cofinancier's comments raised two sets of issues. First is the need for explanations (of terminology, methodology, composition of team members, TORs for the ICR mission, the significance of a "c" environmental category, the difference between a "core ICR" and an Intensive Learning ICR (ILI), data sources, etc.).  This ICR is prepared following a standard template, and, given space limitations it has provided details and clarifications on all relevant topics. There was no relevant environmental issue in this project – either at the design stage, or during implementation. An ILI covers the same areas as a core ICR, but, in addition, it includes more extensive analysis and text in order to help disseminate project results for application to new operations. No new operation is envisaged for this project. The data sources for this ICR are the MIS system and the commissioned studies that are listed in Annex 12.

Second is the suggestion that the PAD did not include an extensive social analysis by which it would have been easier to foresee the impact of the women's basic conditions (illiteracy, unemployment, position inside the household) on the project.  This ICR finds that the PAD did a good social analysis of the conditions and indeed set out to address all three issues in tandem, but it prioritized the issues of women's position inside the household and their unemployment (both of which improved for the majority of participating women) as more likely to have broader impact on families and communities.

(c) Other partners and stakeholders
(e.g. NGOs/private sector/civil society)

Annex 1. Results Framework Analysis

Project Development Objectives (from Project Appraisal Document)

This Learning and Innovation Loan (LIL) aimed to test methodologies to enhance the social and economic welfare of households in selected poor districts by improving and increasing women's opportunities through increased skills, productivity, and income. The key indicators consisted of:  (1) Input indicators (e.g., number of IEC campaigns launched); (2) Process indicators (e.g., administrative costs as percentage of project costs); and (3) Output/Outcome indicators (e.g., number of women's groups (WGs) strengthened). (See Annexes 1, 4, and 5).

Revised Project Development Objectives (as approved by original approving authority)

There was no revision of the PDO.

(a) PDO Indicator(s)

Indicator

Baseline Value

Original Target Values (from approval documents)

Formally Revised Target Values

Actual Value Achieved at Completion or Target Years

Indicator 1 :

Level of satisfaction of beneficiaries with group-level and community level outcomes

Value
(quantitative or
Qualitative)

none

none

none

High among the vast majority of beneficiaries

Date achieved

03/31/2001

03/31/2001

06/30/2006

06/30/2006

Comments
(incl. % achievement)

Among beneficiaries interviewed, 98% reported a change in attitudes of community members towards them, 97% report enhanced self awareness, and 94% report improved participation in household decision making (Table 4.6).

(b) Intermediate Outcome Indicator(s)

Indicator

Baseline Value

Original Target Values (from approval documents)

Formally Revised Target Values

Actual Value Achieved at Completion or Target Years

Indicator 1 :

No. of women's groups created or strengthened

Value
(quantitative or
Qualitative)

0

800 (original); revised in MTR to be 530

530

627

Date achieved

03/30/2001

06/30/2006

06/30/2006

06/30/2006

Comments
(incl. % achievement)

During the WDIP’s life, 1,658 WGs were formed and requested GIF funding. 627 of these groups received GIF funding.

Indicator 2 :

Number and % of projects completed

Value
(quantitative or
Qualitative)

0

800 original; revised in MTR to be 530

530

623

Date achieved

03/30/2001

06/30/2006

06/30/2006

06/30/2006

Comments
(incl. % achievement)

The number of WGs funded exceeded the revised MTR target by 17 percent.

Indicator 3 :

Proportion of target population that received training (grassroots management and skills development)

Value
(quantitative or
Qualitative)

0

All members of WGs

All members of WGs

100%

Date achieved

03/30/2001

06/30/2006

06/30/2006

06/30/2006

Comments
(incl. % achievement)

Fewer women (54%) received SDT, but all women received GMT.

Annex 2. Restructuring (if any)

Not Applicable

Annex 3. Project Costs and Financing

(a) Project Cost by Component (in USD Million equivalent)

Components

Appraisal Estimate (USD M)

Actual/Latest Estimate (USD M)

Percentage of Appraisal

GRASSROOTS INITIATIVE FUND

4.46  

3.27  

73.32  

INSTITUTIONAL STRENGTHENING (CAPACITY BUILDING, INFORMATION, EDUCATION, AND COMMUNICATION (IEC); AND MANAGEMENT INFORMATI ON SYSTEM

1.81  

1.61  

88.95  

PROJECT MANAGEMENT

1.73  

2.75  

158.96  

Total Baseline Cost  

8.00  

7.63  

Physical Contingencies

0.00  

Price Contingencies

0.00  

Total Project Costs 

8.00  

Front-end fee PPF

0.00  

0.00  

0.00  

Front-end fee IBRD

0.00  

0.00  

0.00  

Total Financing Required  

8.00  

7.63  

(b) Financing

Source of Funds

Type of Cofinancing

Appraisal Estimate (USD M)

Actual/Latest Estimate (USD M)

Percentage of Appraisal

   Borrower

0.70  

5.90  

842.86  

   ITALY, Govt. of (Except for Dev. Coop. Dept. - MOFA)

1.80  

1.80  

100.00  

   International Development Association (IDA)

5.00  

5.52  

110.40  

   MULTIPLE DONORS

0.50  

0.50  

100.00  

(c) Disbursement Profile

Annex 4. Outputs by Component

WDIP, an $8 million project (not including taxes and duties) financed by IDA ($5 million), a grant from the Government of Italy ($1.8 million), and the GoE (.59m), had three components:

Component 1. Grassroots Initiatives Fund (GIF), a financing window for demand-driven micro-subprojects owned and implemented by WGs who have organized themselves around productive activities (maximum grant of $4,000 and 10% contribution by WGs); at a total cost of $3.26 million. The GIF provided capacity building, training, and start-up funds to assist WGs to organize themselves around cost-effective, relatively small, and technically simple productive activities. The capacity building portion of the GIF was implemented through grassroots management training on topics such as general business, financial, and human resources management.  Intermediaries and Facilitators played instrumental roles in delivering this support.

• Intermediaries included government agency staff from relevant ministries (e.g., Ministry of Agriculture) and NGOs (e.g., ProPride, Christian Children's Fund, and Gewane Agricultural College).  They were contracted to provide training and guidance in identifying group businesses, monitor implementation, serve as liaisons between the WGs and other WDIP stakeholders, and supervise the Facilitators.  In a few cases, prospective Intermediaries refused to sign contracts because they considered them to be complex.
• Facilitators worked with WGs by providing support in preparing business and financial plans, market research, record and book keeping, and delivery of IEC messages. Facilitators' support was the most critical factor for a group's success. Because most members (about 60 percent) were illiterate, Facilitators performed financial recordkeeping tasks. The project's design unrealistically assumed that WGs would require Facilitators' support for only 18 months.  However, by the time the project closed, many WGs still needed this support. Exit studies were supposed to assess the percentage of groups ready to graduate. The full results of the Addis region exit study show that 30 percent of WGs were not ready to graduate. An indication of this dependency, but also a significant development, was the fact that some WGs were hiring their assigned Facilitators through their retained earnings.

Component Outcomes: In terms of outcomes, the project's MIS data show that as of June 30, 2006,  627 WGs had received WDIP support (of which 623 received GIF funding), benefiting 11,675 poor women and close to 60,000 of their family members (Table 4.1). 260 WGs (42%) were in rural areas, while 361(58%) were urban based.  621 WGs continued to operate at the time of project closing. While this number is lower than the original 800 WG set at project initiation, it is 17 percent higher than the 530 target set at the MTR in 2003.

  Table 4.1. Summary Process/Output Indicators

  Source: MIS Data

Through its emphasis on group operation, the GIF tested modalities such as peer support, group modality to maximize the number of beneficiaries, and business skills enhancement through training. Reaching out to build the capacity of the target group was slow and difficult because of high illiteracy levels; unemployment at the time of joining (44 percent); and the large number of women heads of households (22 percent). Therefore, in addition to the insufficient 18-month period to fully empower women to manage their businesses the phasing out strategy was inadequate.  A more realistic strategy would have involved a systematic and graduated cost sharing arrangement between the WDIP and the WG of Facilitators' services, spread out over a longer time frame, whereby the project could have covered 100% of Facilitators' fee during the first 12 months, decreasing to 50% for the next 6 months and 25% for the following 6 months. Furthermore, an incentive-based fee system should have been established for Facilitators, pegged to the WGs' performance, with the effective transfer of knowledge to WG members as a performance criterion.  Such provisions would have increased the Facilitators' sense of ownership of the success of the group businesses and provided a potential continuing source of employment for them independent of the project.

Benefits and risks of different group modalities: At the beginning of the project only one business group modality was open to the beneficiaries.  Later, the project allowed diversification into two additional modalities, primarily to address the insufficient profit generated from group-operated businesses and also as a risk minimization strategy. Each modality presented opportunities and challenges:

• Group-operated businesses: Under this modality, all WG members work on, and share the profits from one income-generating activity.  The potential limitation of this modality is that the business activities could engage only a few members at a time and members had to work in shifts. While working in shifts allowed members flexibility to manage household tasks, if the group size was large, the allocated time per shift could be inadequate to guarantee sufficient income. In fact many WGs operating group businesses incurred losses, and WGs with more than 22 members were the least profitable.  Some successful group businesses introduced an innovative and incentive-based business management system by appointing one group member who had business management skills as an executive manager.
• Individually-operated business: This modality allowed individual members to engage in their preferred income-generating activity by taking an interest-bearing loan (5-10 percent) from the WDIP grant. These loans were reviewed and approved by a Loan Committee set up to administer the grant.  Most WGs which use the individually-managed business modality have started saving schemes managed by the group.
• Hybrid business: This was a mix of the above two modalities, whereby the WDIP grant is divided into two portions.  One portion is allocated for the group business and the other for an interest-bearing loan to individual members, administered along the same lines as the individually-operated business modality. The group business profits are shared equally and individual business profits accrue to the individual member. For this type of operation, a large proportion of members' income comes from the individually-operated business.

The exit studies and economic analysis (Annex 5) confirm that the individually-operated and hybrid group modalities are more profitable, give fast return, are less conflict prone and easy to manage and, therefore, preferred by most WG over the group- operated businesses. Factors contributing to individual-based businesses' success include the fact that they involved family members (enhancing family members' commitment and learning), promoted business diversification and credit and savings, and encouraged rapid decision making to take advantage of market opportunities. Group businesses operated on the basis of consensus and decision making could sometimes not be timely enough to take advantage of market prices.  The final WDIP impact study observed that the group-operated businesses had little or no division of labor or economies of scale except in marketing, and use of traditional technologies. Transaction costs of keeping the group together were also high. Economic viability of group businesses was sometimes improved by reducing minimum group size (which the project did), introducing new technology and diversifying business activity.

Positive demonstration effects. The WDIP's positive effects on members' income and social welfare have spurred other women to replicate this success by forming their own groups and initiating savings in anticipation of WDIP support. Although the project has reached its absorptive capacity to support WGs, and the RCUs have stopped forming and supporting new WGs, they continue receiving requests from such groups, which is a clear indication that the potential for replication exists and is high.

Sustainability of GIF outcomes: The key challenge is to ensure the sustainability of GIF investments and the outcomes listed in Table 4.2. The implementing agencies, WGs, and other stakeholders have taken several measures to expand their abilities to operate in the future without WDIP support. Examples include:

1. Different linkages have been established to provide ready markets for WGs in certain industries.  Noteworthy examples of supply chain linkages fostered by WDIP are: dairy producers and milk processing plants (Amhara and Oromia); animal fattening operators and local slaughterhouses and live animals exporters (Oromia region); accessories producers and local leather factories producing for export in Addis Ababa; and spice and food producers to universities, hospitals, prisons etc. (Addis Ababa, Benshangul and Amhara).
2. Linking WGs to local Microfinance Institutions (MFIs) to facilitate access to credit to expand their businesses in Tigray and Dire Dawa. Access to MFIs in other regions is not easily available because of the limited coverage of MFIs. The WDIP's support was deliberately focused on areas where MFIs were not active to avoid competition with MFIs. As most WGs have now acquired legal status and are operating viable businesses, MFIs could be encouraged to expand their services to the areas where successful WGs are operating.
3. Attaining legal status as cooperatives. In August 2006, 263 WGs had secured cooperative licenses, while 310 were undergoing the process of securing licenses. Cooperative status is an important strategy that ensures business continuity and scaling up of WGs' activities, and secures the confidence of WGs' business partners. Other benefits of legal status include eligibility to bid in public tenders, tax exemption, and access to different government services (including annual audit services, training, credit, and product promotion) all of which enhance business growth and sustainability.
4. Creation of credit and saving schemes: Through the project's efforts, 310 WGs have started credit and saving schemes.  These schemes would facilitate an increase in group capital (through the interest generated from members' loans from the group capital) and thereby strengthen the sustainability of the groups.

Table 4.2. Outcome Indicators

Source: MIS Data

Component 2: Institutional Strengthening for (a) capacity building program ($0.63 million) for Intermediaries and Facilitators who will support WGs with organization, facilitation, training, project design, appraisal, monitoring and evaluation, and reporting; capacity building for the CCU, RCUs and other implementing units to ensure effective planning, implementation, management, and sustainability of micro-projects; and providing training to WGs; (b) IEC program ($0.72 million) for increasing awareness of both men and women on social and legal issues; and (c) participatory monitoring and evaluation program ($0.46 million) to track the project's indicators and assess the project's impact on beneficiaries' socio-economic status.

The World Bank's sourcebook on empowerment says, "information is power.…" In the communities of the developing world, poor people are usually denied information or, where no information is available, are unable to access it. Poor people are dependent on a few well-informed (or sometimes not so well-informed) local leaders for information. This information governs most of their actions and results in continuation of traditional practices which at times do not have any practical benefit and in some cases are harmful (e.g., female genital mutilation). Ensuring access to information is particularly important in the wake of rising HIV/AIDS prevalence, environmental degradation, and gender inequalities. For poor people to take effective action, it is important that they are provided relevant and easily understandable information.

Component outcomes: GIF support covered topics and themes for which information is important to empower poor people, and did so with several positive outcomes.  It trained 215 Intermediaries and 193 Facilitators (Table 4.3). It provided 36 skills development training courses and trained 322 experts, who in turn provided training on various business development and IEC courses on six themes.  There are no data to disaggregate the 623 IEC campaigns by theme. However, the campaigns were delivered to over 20,000 male and female beneficiaries. As a result of these trainings and exposure, 5,653 women improved their pre-existing skills, and over 95 percent of them actively engaged in businesses after such training. Because this component focused mainly on the IEC subcomponent, the review of the project's impact will focus on that aspect.

Table 4.3. Output Indicators

Source: MIS Data

Given the importance of information, the WDIP financed some new and innovative IEC programs through six modules aimed at increasing awareness of health and education issues. Each woman, her partner and local village leaders were provided information on one or more modules. By providing women with information on health and education issues, the project aimed not only to empower women to protect their families from harmful traditional practices and to use family planning methods, but also to enhance women's own awareness of, and understanding of, their rights and obligations. The project also empowered the women by improving their participation in community development, social, and political activities.

Component impacts:Did the WDIP's IEC program have an impact on the lives of the women? Did it empower the poor women? Three studies have been carried out to assess the WDIP's impact on women's empowerment. The conclusions in each are based on the analysis of data collected through structured household questionnaires.  Two studies (Legovini, 2004 and Ali, 2006) are based on only two regions. The third (B&M Development Study, 2006) covered 7 regions and has the most representative sample, but lacks a counterfactual, i.e., the questionnaire was fielded only to the women supported by the project and not to potential beneficiaries. So even though a high percentage of women report improved participation in and outside the house, increased awareness, improved self perception, etc., direct attribution to WDIP is difficult.

To assess the impact of the IEC program, a counterfactual was created from within the 1,220 beneficiary women surveyed by B&M Development Consultants. Six IEC modules were held, but only 41 percent of the women attended all of them. Sixty-eight percent attended four or more, and 35 women attended none. Table 4.4 provides a breakdown of attendance.

Table 4.4: Percent Women attending IEC Modules

Following the results chain, attendance should result in increased information, which, in turn, should influence behavior. Three variables were used from the questionnaire to assess beneficiaries' perception about enhanced knowledge and four variables to capture change in practice (Table 4.5). Attendees of IEC module(s) were far more likely to perceive enhanced knowledge as compared to non-attendees for all knowledge variables. This in no way should be interpreted to mean that there was no demonstration effect. All that the analysis reveals is that the direct impact from attending a module is higher than the indirect impact. Similar results were found for practice of knowledge acquired at IEC.