Document of The World Bank Report No: ICR00003568 IMPLEMENTATION COMPLETION AND RESULTS REPORT (TF-11069) ON A TRUST FUND IN THE AMOUNT OF US MILLION (US$ 5 MILLION EQUIVALENT) TO THE REPUBLIC OF TUNISIA FOR A STATE AND PEACEBUILDING FUND (SPF) PARTICIPATORY SERVICE DELIVERY FOR REINTEGRATION PROJECT September 30, 2015 Social Protection and Labor Global Practice Middle East and North Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective August 21, 2015) Currency Unit = Tunisian Dinar (TND) TND 1.00 = US$ 0.52 US$ 1.00 = TND 1.93 FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS ALMP Labor Market Programs CBT Central Bank of Tunisia CSO Civil Service Organization DPL Development Policy Loan FM Financial Management G2P Government to People Payment GOT Government of Tunisia IA Implementation Agency ISR Implementation Status & Results report M&E Monitoring and Evaluation MENA Middle East and North Africa MEVT Ministry of Employment and Vocational Training MoSA Ministry of Social Affairs PAD Project Appraisal Document PDO Project Development Objective PIU Project Implementation Unit PNAFN National Assistance for Needy Families Program SPF State and Peace-Building Fund TND Tunisian Dinar UTSS Union Tunisienne de la Solidarité Sociale ii Vice President: Hafez Ghanem Country Director: Marie Francoise Marie-Nelly Sr. Global Practice Director: Arup Banerji Practice Manager: Yasser El-Gammal Project Team Leader: Heba Elgazzar ICR Team Leaders: Amr Moubarak and Suzanne Essama iii TUNISIA State and Peacebuilding Fund (SPF) Participatory Service Delivery for Reintegration Project CONTENTS Data Sheet A. Basic Information....................................................................................................... v B. Key Dates ................................................................................................................... v C. Ratings Summary ....................................................................................................... v D. Sector and Theme Codes .......................................................................................... vi E. Bank Staff .................................................................................................................. vi F. Results Framework Analysis .................................................................................... vii G. Ratings of Project Performance in ISRs .................................................................... x H. Restructuring (if any) ................................................................................................. x I. Disbursement Profile ................................................................................................. xi 1. Project Context, Development Objectives and Design ............................................... 1 2. Key Factors Affecting Implementation and Outcomes .............................................. 6 3. Assessment of Outcomes .......................................................................................... 11 4. Assessment of Risk to Development Outcome......................................................... 17 5. Assessment of Bank and Borrower Performance ..................................................... 17 6. Lessons Learned ....................................................................................................... 19 7. Comments on Issues Raised by Grantee/Implementing Agencies/Donors .............. 20 Annex 1. Project Costs and Financing .......................................................................... 21 Annex 2. Outputs by Component ................................................................................. 24 Annex 3. Economic and Financial Analysis ................................................................. 25 Annex 4. Grant Preparation and Implementation Support/Supervision Processes ....... 27 Annex 5. Beneficiary Survey Results ........................................................................... 28 Annex 6. Stakeholder Workshop Report and Results................................................... 30 Annex 7. Summary of Grantee's ICR and/or Comments on Draft ICR ........................ 31 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ....................... 33 Annex 9. List of Supporting Documents ...................................................................... 41 Annex 10. Country Map ............................................................................................... 41 iv TUNISIA State and Peacebuilding Fund (SPF) Participatory Service Delivery for Reintegration Project DATA SHEET A. Basic Information Participatory Service Country: Tunisia Project Name: Delivery for Reintegration SPF Project ID: P127212 L/C/TF Number(s): TF-11069 ICR Date: 08/18/2015 ICR Type: Core ICR REPUBLIC OF Lending Instrument: SIL Grantee: TUNISIA Original Total USD 5.00M Disbursed Amount: USD 4.76 M Commitment: Revised Amount: USD 5.00M Environmental Category: C Implementing Agencies: Union Tunisienne de la Solidarité Sociale Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 05/23/2011 Effectiveness: 10/03/2012 Appraisal: 06/23/2011 Restructuring(s): 02/27/2014 Approval: 11/02/2011 Mid-term Review: 12/01/2013 12/01/2013 Closing: 03/31/2014 03/31/2015 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: Substantial Bank Performance: Moderately Satisfactory Grantee Performance: Moderately Satisfactory v C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Satisfactory Government: Moderately Satisfactory Implementing Quality of Supervision: Satisfactory Satisfactory Agency/Agencies: Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance: Performance: C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Quality at Entry Project at any time No None (QEA): (Yes/No): Problem Project at any Quality of No None time (Yes/No): Supervision (QSA): DO rating before Satisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Health 20 20 Other social services 30 30 Vocational training 50 50 Theme Code (as % of total Bank financing) Other human development 100 100 E. Bank Staff Positions At ICR At Approval Vice President: Hafez M. H. Ghanem Inger Andersen Country Director: Marie Francoise Marie-Nelly Neil Simon M. Gray Practice Yasser Aabdel-Aleem Awny El- Enis Baris Manager/Manager: Gammal Project Team Leader: Heba Elgazzar Heba Elgazzar Suzanne Essama/ Amr S. ICR Team Leader: Moubarak ICR Primary Author: Suzanne Essama Amr S. Moubarak vi F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The Project Development Objective of the Tunisia Participatory Service Development for Reintegration Project (PSD, the Project) is to pilot participatory approaches to employment generation through a cash-for-service program for vulnerable Tunisian households. The Project supports the financing of Subprojects focusing on meeting social needs by improving service delivery at the local level. The Project contributes to: (i) fostering social cohesion and stabilization among disadvantaged populations following the January 2011 revolution, and (ii) mitigating the socioeconomic risks faced by Tunisians returnees fleeing Libyan civil strife and by Tunisian communities near the border with Libya affected by depressed cross-border trade and commerce. Revised Project Development Objectives (as approved by original approving authority) N/A vii (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Beneficiaries that have participated in implementing Sub-projects (Number) Indicator 1: Value quantitative or 0.00 4000.00 6180.00 Qualitative) Date achieved 11/02/2011 06/30/2014 03/31/2015 Comments At project completion, the total number of direct beneficiaries participating in (incl. % Sub-projects exceeded the target number by 55%. achievement) This indicator was Achieved. Proportion of beneficiaries who perceived an improvement in job skills as result Indicator 2: of the training acquired through the implementation of Sub-projects (%) Value quantitative or 0.00 60 54 Qualitative) Date achieved 11/02/2011 03/31/2015 03/31/2015 Comments 75% of beneficiaries indicated that "work experience", was the biggest reason (incl. % for an improvement in their job skills. achievement) This indicator was Partially Achieved. Indicator 3: Proportion of women among beneficiaries employed (%) Value quantitative or 0.00 30.00 70 Qualitative) Date achieved 11/02/2011 03/31/2015 03/31/2015 Comments The proportion of women exceeded the target by over 200% (incl. % This indicator was Achieved. achievement) Members of communities benefiting from Sub-projects who receive services Indicator 4: through local Sub-projects (Number). Value quantitative or 0.00 12000.00 62669.00 Qualitative) Date achieved 11/02/2011 06/30/2014 10/17/2014 Comments By project completion, the number of community members exceeded the initial (incl. % target by more than 500%. achievement) This indicator was Achieved. Proportion of communities members benefiting from Sub-projects who perceive Indicator 5: services received respond to local needs (%) Value quantitative or 0.00 60.00 67.00 Qualitative) Date achieved 11/02/2011 06/30/2014 03/31/2015 viii Comments (incl. % This indicator was Achieved achievement) Proportion of beneficiaries and community members benefiting from sub- Indicator 6: projects who perceive greater involve in local decision-making regarding service delivery (%) Value quantitative or 0.00 60.00 56.00 Qualitative) Date achieved 11/02/2011 03/31/2015 03/31/2015 Comments This indicator was Partially Achieved (93% of target). (incl. % achievement) (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Employment opportunities are generated and access to key basic services has Indicator 1: increased (%) Value (quantitative 0.00 60.00 67.00 or Qualitative) Date achieved 11/02/2011 03/31/2015 03/31/2015 Comments Percent of community members that reported “additional services provided (incl. % beyond those in the public sector” in the beneficiary survey assessment. achievement) This indicator was Achieved. Local community participation and job skills in social service delivery have Indicator 2 : increased (Workers perceive service delivery training responds to job skills development needs, at least 60 percent of workers, %) Value (quantitative 0.00 60.00 72.00 or Qualitative) Date achieved 11/02/2011 03/31/2015 03/31/2015 Comments (incl. % Achieved target. Exceeded target (based on beneficiary survey). achievement) Central and local capacity for project management, monitoring and evaluation has increased [Bi-annual project management (results matrix) targets have been Indicator 3 : met by PIU UTSS central and regional staff, at least 70 percent of bi-annual results met, %] Value (quantitative 0.00 70.00 100.00 or Qualitative) Date achieved 11/02/2011 03/31/2015 03/31/2015 ix Comments Achieved target. “Targets” refers to data collection and implementation targets (incl. % as per beneficiary enrolment and sub-project calls for proposal calendar. achievement) G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 09/26/2012 Satisfactory Satisfactory 0.00 2 05/21/2013 Satisfactory Satisfactory 0.50 3 02/15/2014 Satisfactory Satisfactory 2.21 4 11/12/2014 Satisfactory Satisfactory 4.23 H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions To extend the closing by 12 months from March 31, 2014 to March 31, 2015. To allow completion of some sub- 02/27/2014 S S 2.45 projects, completion of cash transfer disbursements to 4000 project beneficiaries, and to prepare beneficiary assessment. x I. Disbursement Profile xi TUNISIA State and Peacebuilding Fund (SPF) Participatory Service Delivery for Reintegration Project 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal Country Background 1. In January 2011, Tunisia experienced an unprecedented popular uprising, driven largely by underlying social disparities, allegations of corruption against the government, and broad frustration with lack of employment and economic opportunities. Protests swept Tunisia in December 2010 and January 2011, which culminated in the ousting of the sitting president, Ben Ali, and his regime, after 23 years in power. An interim government was established and tasked with organizing the elections for the Constituent Assembly, which was elected in October 2011. Within its capacity as a caretaker government, the interim Government of Tunisia (GoT) adopted short-term measures to prepare the way for greater social and economic inclusion. Following elections in November and December 2014, a permanent Government with a five-year mandate was formed in 2015. 2. Since the 2011 uprising, Tunisia’s macroeconomic outlook has been considered uncertain due to reduced tourism, trade and foreign investment following the revolution and the crisis, at the same time, in neighboring Libya. Before the revolution, Tunisia’s GDP growth was steady at 5 percent, relatively high for the Middle East and North Africa region. Using a poverty line of US$2 per day, poverty in Tunisia was estimated to be 15 percent as of 2010, the latest available data (new household survey data is expected by late 2016 to update the poverty estimates). Following the revolution, unemployment was estimated to have grown from 13 percent (500,000 people) in 2010 to approximately 18 percent (700,000 people) in early 2011. An estimated 39 percent of the unemployed have a secondary education and 32 percent have a university education. Sector Background 3. As of 2011, Tunisia had made considerable progress in social outcomes for over two decades, yet wide regional disparities in poverty, unemployment and human development outcomes were evident. Human development indicators in poorer, western and southern regions lagged behind the levels seen in other middle-income countries. With a population of approximately 10 million (2011) and an average GDP per capita of US$3,072 (current dollars), life expectancy at 74 years was considered relatively average for a middle-income country. The adult literacy rate was estimated at 77 percent. The population was and remains largely urban at 65 percent. As low as 40 percent of the population in rural regions had access to modern sanitation, compared to 95 percent in urban areas. Unemployment was estimated at 15 percent of the population, exacerbated by a young population where 23 percent were below the age of 15. Youth unemployment was estimated to be as high as 43 percent (ILO/World Development Indicators). The infant mortality rate (IMR) was relatively low at 18 per 1,000 live births as of 2008, yet higher than in middle-income countries such as Ukraine and Thailand. The rate of use of prenatal services (4 or more visits) was also poorer in Tunisia than in most other middle-income countries. 1 4. Social policy in Tunisia has been based on principles of solidarity according to Tunisia’s series of five-year National Development Plans, with social sector expenditures accounting for over 18 percent of GDP. Increased investment in rural roads, housing programs, food subsidies, cash transfers and free health cards had supported improvements of living conditions of vulnerable populations. These programs, however, have not been adequately designed given persistent regional disparities. Social sector expenditures in 2011 comprised approximately: (i) 4 percent of GDP on pensions; (ii) 5 percent of GDP on health (7 percent of total public expenditure); (iii) 6 percent of GDP on education (20 percent of total public expenditure); (iv) 1.0 percent of GDP on active labor market programs; (v) over 3 percent on food and fuel subsidies, and (vi) approximately 0.4 percent on social assistance programs (including cash transfers and disability benefits). 5. In terms of social safety nets and economic inclusion, since 2003, the GoT had been providing social assistance to vulnerable households through three main programs administered by the Ministry of Social Affairs (MoSA), including a cash transfer program, a free health card program. In 2011, the GoT had been providing monthly cash transfer allowances to 185,000 households (7.2 percent of the population) and subsidized health insurance cards to 752,900 individuals (7.4 percent of the population). Beneficiary households received a cash transfer allocation of 70 TND per month with an additional supplement of 10 TND per school-age child, up to a limit of three children (average household size is 4). Prior to the 2011 revolution, the number of household beneficiaries of income support was 135,000, which was increased by 50,000 after the January 2011 revolution. As of 2009, cash transfer beneficiaries were distributed as follows: 52 percent of beneficiaries were female; 68 percent were older than 60 years; 49.5 percent lived in rural areas; and 18 percent had a disability. The number of free health insurance card beneficiaries also increased after the revolution from 175,000 to 195,000 (an increase of 11 percent), with another 557,900 receiving subsidized insurance cards. In terms of disability benefits, approximately 2 percent of the population was estimated to have had a long-standing disability; MoSA provided disability benefits to 180,000 individuals as of 2011. Table 1 : Overview of social assistance programs administered by MoSA, 2011 Program Expenditure As No. Percent of (million TND) percent beneficiaries population of covered MoSA (%) budget (%) Cash transfers: National Assistance 146 87.4% 740,000 7.2% Program to Needy Families (PNAFN) (185,000 households) Health Insurance Card Program n/a n/a 752,900 7.4% Disability Assistance 12 7.1% 180,606 1.8% Social Counseling, Child Care and 9 5.5% n/a n/a Adult Education Programs TOTAL 158 100.0% n/a n/a Source: Ministry of Social Affairs, 2011. Reflects government estimates as of April 2011. 2 6. In terms of governance and transparency in service delivery, the period leading up to the 2011 revolution witnessed serious issues of governance, transparency and accountability that weakened citizens' voice and undermined economic prospects. Government’s agenda did not include international technical assistance and collaboration on the topic. Following the revolution, the Interim Government established an independent governance commission to investigate corruption and embezzlement allegations against the Ben Ali regime (2011). Although the Government adopted a reform to implement participatory monitoring of public services, this approach had not yet been implemented in the employment arena at the time of the project design. Participation of the private sector and civil society in general regarding the identification, selection and implementation for employment programs was lacking. A lesson learned for government strategy and the Bank’s Interim Strategy Note has been to identify areas of investment that respond to the demands for good governance, transparency, accountability and better service delivery as part of the shared and sustainable growth goals for the country. Government Strategy 7. At the time of project approval in 2011, Tunisia’s interim Government had garnered the support of the May 2011 G8 Summit for its Social and Economic Program. This program focused on two key pillars for the transition period: (i) improving transparency and accountability in a quick and visible way to respond to the aspirations of the population and to signal that Tunisia is creating a level playing field for private sector-led employment growth, and (ii) taking immediate measures to relieve the plight of the unemployed and the most vulnerable of families and accelerate the development of lagging regions. Rationale for Bank Assistance 8. The social impact of the Libyan conflict was acute in Tunisia’s regions that bordered Libya. These regions had historically relied heavily on trade and remittances from Libya as important sources of revenue. Prior to the conflict, Libya was host to 87,177 Tunisian workers. Following the Libyan conflict, Tunisian migrants, largely low- and semi-skilled workers, returned mainly to poorer governorates. They were predominately male, aged 30-50 years, with dependents. MoSA requested the World Bank’s support in piloting social assistance programs in fragile communities, with an influx of refugees and returning Tunisian migrants facing social unrest, in regions bordering Libya. This request came under the auspices of creating a pilot that could generate rapid employment opportunities for the disenfranchised communities most affected by the Libyan crisis, with an emphasis on also improving service delivery to areas that were economically inactive. The selection of these service delivery activities (or sub-projects) involved community participation (see section 2.1). 9. World Bank assistance for the Participatory Service Delivery for Reintegration Project (the Project), financed by the State and Peace-building Fund (SPF), was mobilized to support state building by mitigating the negative economic and social effects on households associated with conflict-affected regions bordering Libya. The Project was to support the SPF objectives of state and peace building through the following: (i) employment or income- generating activities through inclusive development sub-projects which can increase access to basic services in conflict-affected areas; (ii) peace-building through reintegrating returning migrant workers and building trust in emphasizing a participatory nature to the selection and monitoring of these service delivery activities to these communities affected socially and 3 economically by the ongoing crisis. It would complement earlier World Bank support for Tunisia’s Social and Economic Recovery Program provided through the Governance and Opportunity Development Policy Loan (DPL), approved in June 2011 which included measures of improving transparency in the public sector, increased participation and citizen voice in provision of services. This DPL also supported Tunisia’s policies to expand civil society involvement in monitoring service delivery in the social sectors, and ramping up employment generating activities. 10. The Project targeted the regions of Tataouine and Medenine that had among the poorest economic and social outcomes in Tunisia. In 2011, unemployment was estimated to be 23 percent in these regions (as compared to 18 percent nationally). Local livelihoods depended heavily on trade, commerce and agriculture. The region of Medenine included several districts with high poverty rates, refugees (including foreign nationals) and a high proportion of vulnerable families whose livelihoods had previously been dependent upon trade with Libya. The Project supported the GoT’s broader vision for improving livelihoods and stability in these regions. 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) The Project Development Objective was “to pilot participatory approaches to employment generation through a cash-for-service program for vulnerable Tunisian households.” 11. The Project’s Results Framework (RF) presented six PDO Indicators against which progress towards the PDO would be measured, and five intermediate results indicators. The PDO indicators did not, however align fully with either the Project’s PDO as defined, or with the project design (Section 2.1). 12. The PDO Indicators were: 1. At least 4,000 workers are employed to implement projects 2. At least 60 percent of workers perceive improvement in job skills as a result of sub-project training 3. At least 30 percent of workers employed are women 4. At least 12,000 community members receive services through local sub-projects 5. At least 60 percent of community members perceive services received respond to local needs 6. At least 60 percent of workers and community members perceive greater involvement in local decision-making regarding service delivery 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification There were no revisions to the either the original PDO or to its key indicators. 1.4 Main Beneficiaries 13. As identified in project documents, the Project’s primary beneficiaries were semi-skilled, unemployed workers from vulnerable households in the poorer governorates of Tataouine and Medenine bordering Libya. Local bureaus of the Union Tunisienne de la Solidarité Sociale (UTSS) would be responsible for selecting beneficiaries. Specific household expenditure criteria and other relevant details were described in the Project’s Operational Manual (OM). As employment in social 4 services tended to attract female labor force participation, the Project specifically included a target indicator for female employment. The number of Tunisian migrants as of the end of May 2011 in the governorates of Tataouine and Medenine was estimated to be 4.3 percent of the total number of returnees, or 2,021 potential workers. Eligibility criteria was used to target beneficiaries in an equitable manner given that the wage to be offered (minimum wage) was likely not low enough to allow for efficient self-targeting (whereby lower-skilled and poorer workers would have been the majority of applicants). To that end, a number of proxy indicators of vulnerability were used as eligibility criteria as detailed in the Project Document. Eligibility criteria for beneficiaries were similar to those used by MoSA for its National Program for Assistance to Needy Families (PNAFN). 14. Indirect beneficiaries included local community members who would receive services provided by workers employed by the Project, such as other vulnerable Tunisian and non- Tunisian refugees from Libya. It was estimated that the Project would provide basic social services to 12,000 community members, assuming 3 in habitants would benefit from every worker enrolled in the program. 1.5 Original Components 15. The Project consisted of three components: (i) Pilot Local Service Delivery Sub-Projects (estimated cost: US$4.3 million); (ii) Local Service Delivery Sub-Project Training (estimated cost: US$0.43 million); and (iii) Project Management, Monitoring and Evaluation (estimated cost: US$0.27 million). Each of the components are described below. (i) Component A: Pilot Local Service Delivery Sub-Projects (US$4.3 million). The component was to finance wages paid to beneficiaries and contracted services aimed at providing basic service delivery through employment generating sub-projects. The sub- projects were awarded and supervised by the Implementing Agency (IA), the Union Tunisienne de la Solidarité Sociale (UTSS). All project finances were to be managed by the IA through a Project Implementation Unit (PIU) established within the UTSS. Sub- projects would be demand-driven and selected from proposals submitted by local NGOs (legally established as “associations”) based on community and local government consultations. (ii) Component B: Local Service Delivery Sub-Project Training (US$0.43 million). This component was to finance consultants’ services and goods for training and skills development for workers and technical assistance to sub-project teams and local government in support of the local community sub-projects awarded to select NGOs (Component A). Training activities on planning and management of service delivery sub- projects, including administrative and financial management, social needs assessment, and human resource management would be organized and managed by UTSS and provided to local government and civil society. (iii) Component C: Project Management, Monitoring and Evaluation (US$0.27 million). This component was to finance consultants’ services, goods and office equipment and supplies for the PIU of the UTSS for day-to-day management and monitoring of the Project. It would support evaluating the performance of sub-projects, including periodic interim and final evaluations. The PIU was to be staffed by UTSS staff as well as consultants hired through the Project for additional support. 1.6 Revised Components 5 The Project’s components were not revised. 1.7 Other significant changes 16. Following a request from the Government in 2014, the Project was restructured on February 27, 2014 to extend the closing date by an additional 12 months, i.e., to March 31, 2015. This was justified to allow for the completion of the following activities: (i) finalization of the sub-projects that had not yet been completed; (ii) completion of cash transfer disbursements to 4,000 project beneficiaries, which had turned out to take longer than had originally been anticipated; and (iii) preparation of the project beneficiary assessment, which would help develop recommendations for possibly scaling up this type of project in the future. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 17. Project Development Objective and Intended Outcomes: As stated in section 1.2, the wording of the PDO did not align fully with the Project’s design and the key indicators by which progress would be measured. Further, the PDO was output oriented, with no outcome oriented objectives. In view of this, this Implementation Completion and Results Report (ICR) interprets the Project’s intended outcomes as understood from a review of the project design, key indicators, and implementation support and supervision documents. The intended outcomes were to have designed and piloted a project that: (i) built participatory approaches for the target population, and (ii) generated opportunities for employment (including self-employment) for the target population. A secondary objective would be to improve service delivery in the communities. Key PDO indicators have been aligned with these intended outcomes to measure progress, as follows: Indicator Key Indicator Intended Outcome Number 1 At least 4,000 workers are employed to Employment and income- implement projects generating opportunities 2 At least 60 percent of workers perceive Employment and income- improvement in job skills as a result of sub- generating opportunities project training 3 At least 30 percent of workers employed are Employment and income- women generating opportunities 4 At least 12,000 community members receive Improving service delivery services through local sub-projects 5 At least 60 percent of community members Building participatory perceive services received respond to local approaches AND improving needs service delivery 6 At least 60 percent of workers and community Building participatory members perceive greater involvement in local approaches decision-making regarding service delivery 6 18. Project Design. The Project’s overall design supports the intended development outcomes above. The contracted sub-projects were implemented by adopting three approaches: (a) identification of potential local sub-projects through open calls for proposals at the local level to NGOs and the private sector to solicit designs for local workfare projects targeting community services were the projects selected were expected to be labor intensive (specifically demanding low-skill labor) (b) selection of projects through consultations with civil society groups against specific criteria (as described in the PAD) to ensure that citizen participation plays a role in the selection of sub-projects that addressed community needs, and (c) involvement of beneficiaries (through employment in sub-projects or related income-generating activities) in delivering basic community services promised through awarded sub-projects. 19. Participatory Processes-Theory of Change: As the border communities’ economic potential was very limited for private sector led job growth (Section 1.1), the service delivery sub- projects chosen were considered some of the limited options available to generate employment or income opportunities for low-skilled and semi-skilled beneficiaries in affected areas. The selection of these sub-projects was participatory by bringing the beneficiaries and end users of infrastructure into local decision-making. These participatory processes included: capacity building for community facilitators; local communication strategies; call for proposals; service delivery skills training; and, participatory monitoring assessments among others. Participation helped build trust between communities and government officials, thereby helping to meet secondary SPF objectives of peace-building and social cohesion. Also, as the sub-projects were designed in a participatory fashion, the relevance of their design to local conditions was strong (in terms of income generating opportunities, service needs). This in turn resulted in services and employment and income- generating opportunities more responsive to beneficiary needs, thus improving service delivery and job creation and placement. 20. Bank Preparation and Project Design. World Bank missions were held between February and May 2011 to prepare the proposed project. During the preparation period, the stated PDO and theory of change were less clearly articulated although the intent is reflected in elements of project design. Specifically, clear definitions were prepared to guide participatory design, the types of services expected to be delivered, eligibility criteria, and implementation arrangements that show elements of comprehensive design. This included completed operational manuals and PIU staffing arrangements which were defined prior to effectiveness. 21. Institutional Arrangements. The Project’s institutional arrangements were designed to take into account the institutional capacity of MoSA and UTSS. MoSA provided overall strategic direction and supervision of the Project’s effectiveness, while UTSS built on its current capacity and prior experience working with development partners to develop a project for this unique pilot. 22. Technical and Background Analysis: Project preparation reflected international lessons learned in using participatory approaches for employment generation through the provision of community services and infrastructure support. Preparation and project design also considered lessons in the design and implementation of social protection operations in Tunisia. Employment opportunities and ensuring access to basic services at the local level can help to support social reintegration in fragile and post-revolutionary contexts. The project design as described in the PAD reflected lessons learned from similar approaches in Former Yugoslav Republic of 7 Macedonia1, Thailand2, Iraq3, Bulgaria4 and other Eastern and Central European countries5. In Tunisia, participatory approaches in public service delivery in Tunisia had been limited, but nonetheless the Project took into also account lessons learned from small-scale NGO projects funded through private donations and the Government. 23. Country and Sector Risks: Country and sectoral risks were assessed during preparation through extensive consultations with Governmental and civil society groups. Risks were considered to be high due to the regional context at the time, and due to the fact that the pilot approach proposed by the Project was to be the first of its kind in Tunisia. Stakeholder consultations were conducted to allow for a participatory process, although they remained limited in the Project’s target regions given the political volatility at the time. Previous operations and analytic work showed that opportunities and ensuring access to basic services at the local level can help to support social reintegration in fragile contexts. The World Bank team appropriately anticipated these risks to achieving the PDO. 2.2 Implementation 24. The US$5M State and Peace-building Fund (SPF) Grant Agreement was signed on April 20, 2012. The four effectiveness conditions were met and the grant was declared effective on October 3, 2012. 25. Initial Implementation Delays. The Project showed little progress during its first year of implementation due largely to the need to mobilize and train new PIU staff (on World Bank fiduciary requirements), and also to comply with effectiveness conditions (due to Government staffing changes). By September 2012, the Project had shown no progress towards targets for key indicators (most of which reflected outputs). Implementation began in earnest in 2013. By May 2013, 485 beneficiaries (out of a target of 4000) had enrolled in the program. Project implementation accelerated significantly thereafter, and most activities were completed on time thereafter. 26. Implementation of Activities Aligned with Project Design. The Project was implemented in accordance with its design, as detailed in the Project Proposal, and the Operational Manual The first component, employment opportunities through community service projects, granted. Sub-projects that were supervised by the local associations that received the funds. The second component provided training to beneficiaries who were hired under the subprojects. These trainings included skill development (e.g., fishing, welding), financial literacy, among others. 27. Institutional Arrangements during Implementation: The Project’s institutional arrangements remained as originally designed throughout implementation. Coordination improved 1 World Bank (2007). The FYR Macedonia Community Development Project: Empowerment through Targeting and Institution Building. Washington DC: World Bank. 2 “Piloting Community Approaches in Conflict Situation in Three Southernmost Provinces in Thailand”, State and Peacebuilding Fund Project Document, 2009. World Bank. 3 “Iraq Consultative Service Delivery Program”, State and Peacebuilding Fund Project Document, 2009. World Bank. 4 “Social Assistance for New Employment: from pilot to policy: Introducing community -based social services in Bulgaria”, United Nations Development Program (UNDP) project review, 2009, UNDP. 5 World Bank (2000). Moving from Residential Institutions to Community-Based Social Services in Central and Eastern Europe and the Former Soviet Union. Washington DC: World Bank. 8 between the Ministry of Vocational Training and MoSA, whereby verification of work status and vulnerability were systematically provided by local MoSA offices. However, there was little documentation provided on the role and inputs of the inter-ministerial Steering Committee to allow for improvements as needed. UTSS through MoSA was the IA throughout project implementation which was responsible for launch of all sub-projects during implementation until project completion. According to ISR reviews and an interview with the project director, the project did not experience any implementation challenges at the implementing agency level. All project finances were managed by the Project Implementation Unit (PIU), in close coordination with UTSS regional offices, which provided on the ground capacity building support. For the most part, PIU staff remained the same working through a design model as introduced during project inception throughout project implementation. 28. In-Country Supervision and Implementation Support: The World Bank’s Task Team Leader (TTL) and fiduciary team were based in Tunisia. Furthermore, the same team was responsible for implementation support until completion. This allowed the TTL and team to sustain on-the-ground assistance and implementation support on any issues that arose. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 29. Selection of PDO Key Indicators. The Project’s key indicators reflected the Project’s intended outcomes in its three main areas (Section 2.1). Although the PDO could have been better defined, the PDO Indicators mostly reflect intended outcomes and design. Two indicators were output-oriented and linked directly to project implementation results. For the participatory approaches objective and secondary objective of service-delivery, the Project would have benefited from indicators to monitor how well both the selection and implementation process was participatory to deliver services to target communities. Specific questions at baseline and end of project on delivery of select services could have showed progress towards objective of improved service delivery. 30. Project Monitoring. The PIU (with significant implementation support by the World Bank team) established a project administrative database which was maintained and monitored by a dedicated statistical M&E specialist. Data was collected by the local project coordinator and by NGOs on approximately all 6,100 beneficiary demographics, sub-project details, NGO partner organizations, indirect beneficiaries, beneficiary perceptions of sub-projects, and follow-up employment data following project exit (for a sub-sample of approximately 1,450 beneficiaries) in each sub-region. 31. For project evaluation, a survey firm was hired to conduct a final beneficiary assessment survey during October-November 2014, near completion (Section 3.6). This survey provided information on a sample of approximately 850 beneficiaries and a sample of NGO partners related to demographics, beneficiary perceptions of training and work experience through the sub-projects, and NGO perceptions of project implementation. The results of the project beneficiary survey helped: (i) confirm the appropriateness of the project design model used in targeted communities (i.e., labor intensive service-delivery sub-projects which are chosen by local communities); (ii) confirm that interventions can lead to ‘increase in skill development’ and ‘satisfying local needs’ (as perceived by beneficiaries); and, (iii) provide insightful disaggregation to the effect of the project in generating employment opportunities. These results will likely inform the future roll-out of the program if expanded. 32. Robustness in Measurement of Outcomes. Based on the PIU’s January 2015 Implementation Report and information from the project M&E database, the number of indirect 9 beneficiaries for instance, is based on estimates of catchment areas rather than direct surveying or monitoring. This was due to budget and feasibility in implementation constraints, which may have inflated the number of indirect beneficiaries by accounting for all community members as having benefitted from the public services provided through the community sub-projects carried out. The Project would have benefited from tools to help improve monitoring such as the use of mobile phones and digital technology. 33. Implementation and Utilization. The PIU compiled information on the progress of each component by using information submitted in the context of disbursement processing and of site- visits to monitor intermediate outcome indicators and through tracking MIS data on progress of implementation and service delivery to beneficiaries. In an interview, the Project Director confirmed that the PIU used this information to provide implementation and capacity building support to NGOs experiencing delays in implementation of selected sub-projects.6 2.4 Safeguard and Fiduciary Compliance 34. There were no safeguard and fiduciary compliance issues during the implementation. The project was tagged as Category “C” (no safeguard triggers), and remained as such until completion. There were no safeguards which applied to the project. 35. As for fiduciary compliance, during implementation support missions by the World Bank, regular audits and interim financial reports were prepared and delivered as required. However, some delays were experienced due to a lack of familiarity among PIU staff with Bank operational procedures and technical requirements. Financial Management and Disbursement 36. Financial management (FM) supervision was carried out throughout implementation, including meetings with PIU staff and desk reviews in accordance with OP/BP 10. The main purpose was to review the adequacy of the project’s financial management arrangements at different phases of implementation. Overall FM arrangements in terms of staffing, accounting, reporting, budgeting, internal controls, flow of funds and audit arrangements were reviewed regularly. 37. The Project’s financial management, including budgeting, accounting, internal controls, funds flow, financial reporting and auditing, was deemed adequate and provided: (i) accurate financial information, and (ii) reasonable assurance that project funds were being used for the purpose intended. However, there were initially a few challenges that were addressed in order to fully comply with World Bank’s fiduciary requirements (Annex 9). Procurement 38. An assessment of procurement capacity carried out during project preparation confirmed that UTSS had the capacity, at both central and local levels, to manage and administer the project’s procurement. Two grant effectiveness conditions were related to procurement and they were satisfactorily met in October 2012. These included: (i) the establishment of the PIU which included 6 ICR author telephone interview with Project Director, Latifa Rahmani, UTSS, Tunisia. September 3, 2015 10 an officer in charge of procurement; and (ii) the adoption of the project Operational Manual, including the procurement procedures to be followed. 39. Two major procurement risks were identified and rated to be Substantial: (i) the lack of specific expertise on Bank’s procurement procedures, as this was the first Bank’s financed project implemented by UTSS; and (ii) the potential lack of competition that could have affected the implementation of the small sub-grants assigned to local NGOs. Adequate mitigation measures were identified and agreed with UTSS, and were adequately implemented, particularly the Guidelines for the Selection and Implementation of Project Sub-Grants (which were used for Component 1) which were well prepared and adhered to, allowing the use of “Community Participatory” procurement procedures as enlisted in the Grant Agreement. 40. Compliance with the procurement requirements was good as were control procedures. As a result, procurement was rated Satisfactory throughout implementation. The administrative structure was well organized and it was able to keep record of all appropriate sub-project documentation, including those on procurement. The PIU adequately supervised procurement procedures and ensured that they met the minimum number of bidders required under shopping for goods and civil works contracts to guarantee a sufficient grade of competition. Over time the capacity of the PIU to plan and execute relatively simple procurement in accordance with the World Bank’s procedures increased. The procurement plan for components 2 and 3 were regularly updated and received the Bank’s non-objection. 2.5 Post-completion Operation/Next Phase 41. Based on the performance and results of the Project, the GoT has expressed plans to sustain the model piloted as part of its vision to enhance its social protection and labor system through the following three pillars which are: (i) Strengthening economic inclusion through reform of the National Assistance for Needy Families Program (PNAFN) to support program exit strategies; (ii) Strengthening and modernizing the information and monitoring system at the regional level of social programs and the digitalization of social transfers; and, (iii) Improving financial inclusion through digitizing social transfers and government-to-people payments (G2P) through financial service providers and banking. 42. The GoT has begun to prepare for continuing investments needed to scale-up and integrate the model piloted under the Project within the framework of existing social assistance programs. MoSA has sent a written request to the World Bank for a follow-on project scaling-up the model it piloted and introducing reforms to its existing national cash transfer program for graduation of beneficiaries into the labor market, in coordination with the Ministry of Employment and Vocational Training. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Rating: Substantial 43. Relevance of Objective. The project’s objectives are relevant for the Tunisian context today as they were at the time of the inception of the project. Specifically, Tunisia continues to face significant challenges in high unemployment (specifically for youth and low-skilled workers), and lack of equitable access to services in rural areas. The country also continues to witness challenges in improving transparency and participatory approaches to service delivery where currently few 11 government programs are implemented on the ground that respond to these various objectives. The Interim Strategy Note (ISN) for Tunisia (2013 – 2014), in place at the time the Project was approved, has been extended through 2015. The Project, if it’s objectives had been stated clearly, would remain highly relevant to the ISN and to Tunisia in terms of job creation (ISN Area 1), improving service delivery (ISN Area 2, Social and Economic Inclusion), and building participatory approaches (ISN Area 3, Strengthening transparency and accountability). However, given that the PDO statement did not clearly reflect the intended outcomes and was set at the level of outputs, rather than the intended-outcomes as demonstrated in project design and implementation, the relevance of objectives is rated Modest. 44. Relevance of Design. Given the intended objectives and theory of change as assessed in this ICR (See Section 2.1), the design of the Project was and continues to be very relevant. The overall scope of the project remains relevant where it focuses on two regions bordering Libya were local services and local private-sector job opportunities lacked which has given way for community-driven sub-projects to be piloted. These areas continue to have significant demand for such an approach given little changes in the market demand and employment situation. The intervention chosen (cash-for-work) continues to also be relevant as other possible solutions that would have involved temporary employment were public works projects that faced community skepticism and did not allow for a participatory approach in selection of projects. The pilot operation, which was implemented through an NGO with a dedicated PIU team for the project, also showed a break with the negative legacy and poor implementation record of previous government- implemented public works projects. Finally, the flexibility in the implementation (which was part of the application based/ community awarded) basis of the project, allowed for the participation of several types of entities depending on context. The operators of the SPF project included CSOs, firms, training centers and local professional trade associations. To that end, the project’s relevance of design is rated Substantial. 45. Relevance of Implementation. Although the Project met some implementation challenges during the first year, its implementation was both participatory and demand-driven by local communities and NGOs which resulted in the development and implementation of relevant pilot sub-projects. The implementation of these sub-projects addressed local needs in target areas and contributed to achieving the intended outcomes. Relevance of implementation was especially apparent in (i) addressing the contextual challenges which the project faced at the beginning of the implementation; and (ii) through IA responsiveness to hiccups faced at the project onset. This relevance was measured namely through the beneficiary surveys administered by a third-party firm in the last year of the project. To that end, the relevance of implementation is rated Substantial. 46. Specifically, key operational tools were developed and tested which further supports the rating of ‘Substantial’ for implementation. These tools can be scaled-up and applied to other employment and training programs including: (i) Participatory approaches to launching calls for proposals and defining local development priorities and designing tailored solutions; (ii) Results-based contracting with firms and training providers to improve mid-term employability, which can be improved through gradually strengthening the terms of the contract (iii) On the ground supervision ensures consistency in dialogue and just-in-time response to implementation challenges and issues before they are exacerbated. (iv) Continuity in leadership of the Project (one TTL from inception until completion of the Project) provided continuity and fulfilled the intended outcomes of the Project. 12 Similarly, PIU implementation team and leadership did not experience much reshuffling which encouraged continuity in implementation the project design. (v) Monitoring and evaluation tools to target and ensure accountability of training for improving employability, which can be improved through increased investment and more robust information systems. (vi) Using financial literacy trainings to reinforce employability by promoting financial inclusion and preparing beneficiaries for longer-term self-employment opportunities 3.2 Achievement of Project Development Objectives Rating: Substantial 47. The achievement of the PDO is assessed based on the three intended outcomes presented in Section 2.1. Using these to measure achievement, the Project has, achieved its intended outcomes and PDO based on the key indicators. 48. The project managed to meet most of the PDO indicators generally satisfactorily and combined social safety nets (or cash transfers targeted to poor, low-skilled unemployed) with on- the-job training, making them workfare plus sub-projects. In most cases, small firms and trade associations took an active part in designing and delivering the on-the-job training, often reflecting the skills demanded most by the local labor market. Specifically, a “tripartite” model of implementation at the local level was created for the Tunisian context involving Government, civil society and the private sector, but more can be done in the future to refine and improve local participation in decision-making. 49. Overall, targets for four of six key indicators were achieved or exceeded, and for two were partially achieved. This suggests a rating for Achievement of Objectives as Substantial. 50. Intended Outcome 1: Pilot sub-projects to increase employment and income- generating opportunities. There were three key indicators, which monitored the achievement of the above intended objective.  Indicator 1: At least 4,000 workers are employed to implement projects. The Project resulted in 6,182 beneficiaries who were enrolled in 70 sub-projects. The total number of direct beneficiaries therefore exceeded the target number by 55%. The key indicator was achieved.  Indicator 2: At least 60 percent of workers perceive improvement in job skills as a result of sub-project training. Based on the PIU January 2015 Implementation Report and information from the project M&E database, 54 percent of workers perceive an improvement in jobs skills as a result of sub-project training. The key indicator was partially achieved.  Indicator 3: At least 30 percent of workers employed are women. According to PIU reports, 70 percent of workers employed were women. The sectors with the highest shares of women were education and social assistance, where women accounted for 77 and 75 percent of all beneficiaries employed, respectively. The key indicator was achieved exceeding target by over 200%. 51. Intended Outcome 2: Pilot sub-projects deliver services to communities in select regions.  Indicator 4: At least 12,000 community members receive services through local sub- projects. The target number of indirect beneficiaries has also largely been surpassed 13 (62,669 achieved versus 12,000 targeted). The number of indirect beneficiaries is based on estimates of catchment areas rather than direct surveying or monitoring due to budget and feasibility in implementation constraints, which may have inflated the number of indirect beneficiaries. The key indicator was achieved.  Indicator 5: At least 60 percent of community members perceive services received respond to local needs. This indicator appears to measure (i) whether the participatory process in selection of community sub-project was successful; as well as (ii) whether needed services are delivered. A total of 67 percent of community members reported perceiving that services responded to local needs. Qualitatively, communities, included neighboring households that benefited by services provided by project beneficiaries, and authorities alike found the Project opened up career opportunities in sectors and services with unmet needs, either because public services were unable to reach certain households, or because the services had never been created. The key indicator is achieved. 52. Intended Outcome 3: Pilot sub-projects build participatory approaches in community decision making  Indicator 5 falls under this objective as well. The key indicator was achieved.  Indicator 6. At least 60 percent of workers and community members perceive greater involvement in local decision-making regarding service delivery. By project completion, 56 percent of workers and 47 percent of community members perceive greater involvement in local decision-making regarding service delivery. 7 The key indicator is partially achieved. 53. Detailed description of Achievement of Objectives by key indicator is available in Annex 9.B. 3.3 Efficiency Rating: Substantial 54. The project’s investments yielded relatively high returns as compared to traditional, non-cash for work based social safety nets (see Annex 3 for the detailed economic and financial analysis). The value of the Project lies in its capacity to increase productivity of social safety net beneficiaries and improve local community development incrementally to a greater extent than do traditional social safety nets. The expectation was that within 1 year of program exit, at least a share of these beneficiaries would be expected to have found income-generating activities, as a result of the skills, work and collaborative experience gained. These activities would provide at least the equivalent of Tunisia’s minimum wage (286 TND per month, which is also the average for self-employed, low-skilled earners). The efficiency rating for the project was based namely on the economic allocation of resources to achieve PDO indicators. A more rigorous impact evaluation in the future may help collect the detailed data required for a full analysis in the future. 55. In terms of the cost-effectiveness of the Project, the comparator for purposes of this analysis is the current social safety net program in Tunisia (PNAFN). PNAFN provides benefits to 235,000 heads of households, or on average TND 130 per month per household over the course of 2012-2015 when this project was implemented. Overall, during the period of 2012-2015, the GoT spent approximately 420 million TND per year on PNAFN, or the equivalent of 0.4 percent 7 Source: PIU (UTSS), in partnership with local associations. 14 of GDP. In terms of generosity, these cash transfers amount to nearly 21 percent of average household expenditure. Furthermore, among the sample of 1,430 beneficiaries surveyed in the project within 1 year of program exit, 13 percent had been employed for at least six months. 56. Were the Project to be scaled-up to working-age beneficiaries of PNAFN, this would equate to approximately 30 percent of current beneficiaries, or 70,500 heads of households (leaving 164,500 who would continue receiving unconditional cash transfers). On average, the workfare program, assuming the same design, would cost 94 million TND per year in cash transfers and an additional 14 percent in training and operating costs, or a total of approximately 107 million TND. For the 107 million TND investment, this yields a rate of return (ROR) of approximately 15 percent after accounting for the additional earnings after program exit. 57. These additional benefits are expected to accrue at a financial cost that is nearly equivalent to the current PNAFN budget and in line with average global best practice, making the Project relatively cost-effective. Adding the cost of the productive social safety net (107 million) with PNAFN only among the 70 percent remaining (294 million), gives a total of budget of 401 million TND (or just under 0.4 percent of GDP). This total amount is lower than the current social safety net spending of 420 million TND, and is expected to have a higher return on investment. This level of public spending on a scaled-up program, the coverage and transfer amount would be in line with that of well-performing programs productive social safety net programs elsewhere, such as Brazil’s Bolsa Familia. 3.4 Justification of Overall Outcome Rating Rating: Satisfactory 58. The overall outcome rating for the project is Satisfactory, reflecting an overall rating of substantial for relevance of objectives, design, and implementation (Section 3.1), a substantial rating for achievement of PDO (Section 3.2), and a rating of substantial for efficiency (Section 3.3). 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 59. Longer-term Employment and Income-generation Impact. Continued employment was found at six months following the Project for some beneficiaries. Among 1,360 beneficiaries who had completed and exited the Project surveyed by the PIU as of November 2014, approximately 13 percent had found employment within six months after the completion of workfare projects (Figure 5 Annex 9.C). Most of these beneficiaries have become salaried employees in small firms (65 percent), with the remainder becoming self-employed (35 percent) (Figure 6 Annex 9.C.). As a rough benchmark, among public employment programs in Tunisia targeting the highly skilled unemployed (which also include additional job placement services), the job placement rate was estimated to be approximately 20 percent within one year as of 2012. 60. Social and Economic Inclusion. The Project had a positive impact on income generation and training particularly for the poor and women not benefiting from public employment services and social safety net programs, as well as on social cohesion by promoting involvement in local decision-making. 61. Gender Impact: The project’s key indicators included an outcome (which the PIU has set partially as a goal) of having a proportion of 30% among the direct beneficiaries (i.e., persons hired for service delivery sub-projects). The proportion of women exceeded the 15 target by over 200%, beyond program meeting its prescribed output. In part this was achieved because of the design of the project which allowed women to take up opportunities as direct beneficiaries through sub-projects that did not necessarily include manual labor that is traditionally associated with males in public work projects. Although the evaluation of the project lacked information beyond participation, the expected effects on women empowerment are most likely aligned with similar projects. An impact evaluation of the project is expected to provide more information on the impact of gender. 62. Beyond the impact of the project on direct beneficiaries, there has been recorded effect of the project on the financial inclusion of females. Specifically, through the PIU partnership with the Center for Arab Women for Training and Research (CAWTAR) in delivering three-day basic financial literacy trainings for female beneficiaries focused on bank services, budgeting, debt management, savings, and financial negotiations. The objective of these trainings was to reinforce employability through enhancing financial capability and better equipping beneficiaries with tools for effective self-employment over the long-term. CAWTAR has since 2013 been delivering financial literacy trainings for low-income women and youth across the MENA region, under the Enhancing Microfinance Amongst Women and Youth Project (P144655), financed by the MENA Multi-Donor Trust Fund (MDTF). Beneficiaries who attended the training were found to be more likely to access formal financial services as well employ budgeting and savings to meet financial goals. (b) Institutional Change/Strengthening. 63. The Project helped to develop a model for community-driven graduation mechanisms and livelihoods support programs for the low-skilled, poor households not benefiting from public assistance and for social safety net program beneficiaries. The model is institutionalized through MoSA aim to scale-up and institutionalize the employment support through community sub- projects. An assessment is ongoing in light of the lessons learned from this pilot looking at which aspects of the community-driven participatory approach to selection and implementation of sub- projects to include. Adoption of these elements will demonstrate institutional change to design and implementation of a category of social safety nets programs. (c) Other Unintended Outcomes and Impacts (positive or negative) 64. Financial Inclusion and Financial Literacy. Positive development outcomes were realized that were not initially envisioned in the Project’s design. The Project piloted interventions which sought to reinforce employment skills through promoting financial inclusion amongst beneficiaries. The Government has requested that financial inclusion be included under a future World Bank-financed operation focused on promoting employability in lagging regions. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 65. An external beneficiary survey was conducted after the mid-term evaluation by a survey firm recruited by the PIU in line with terms of reference approved by the World Bank (see Annex 5 for details). The survey was conducted to assess beneficiary, community and partner organizations’ perceptions of process and implementation of the project and, among beneficiaries, the relevance of the workfare sub-project and training modules. Overall, the main benefits of the Project were perceived to be work experience gained by beneficiaries and the services delivered to communities. A total of 75 percent of beneficiaries reported that they benefited by gaining new work experience. However, only 66 percent were satisfied by the content of the training module 16 they received, mainly due to the lack of formal certification offered by the primarily informal training they received from individual instructors. An important negative effect reported by NGOs and beneficiaries was related to delays in receiving payments and cash transfers. These delays were due to lags in financial management and disbursement processing, largely due to familiarity with forms required and Central Bank processing timelines. 4. Assessment of Risk to Development Outcome Rating: Substantial 66. Although the overall high risk posed by the political and institutional context as identified in the PAD was accurate and mitigation measures as described were put in place, the effectiveness of those mitigation measures remained hampered by institutional and implementation challenges. Regardless, the risk for development outcomes remains substantial given the country context (which continues to pose similar political, economic and security challenges as on entry) and institutional arrangement challenges to ensure adequate capacity and inter-governmental coordination for implementation. Specifically, the government intention on using the pilot approach to scale-up cash-for-work programs that are driven by community participation shows a high-level of commitment. However, institutional level coordination between MoSA and UTSS is still not clarified even now that the operation is closed. Furthermore, general delays in implementation of public programs initiated after the revolution hamper the sustainability or scale- up of new programs. Finally, Tunisia’s growth outlook (including decline in public revenues in recent years) and continuous government reshuffles may make it difficult to finance larger programs which justifies a rating of Substantial. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Satisfactory 67. The Project had a strong, appropriate and relevant design at Entry. It was prepared to respond to a request by the Government of Tunisia in less than one year after the January 2011 revolution. That said, there were two limitations to the Quality at Entry. First, the PDO, as articulated, did not reflect the extent of the intended outcomes (both primary and secondary), which were reflected in the Project’s design and implementation. Two of the six indicators chosen to measure progress were output-oriented and did not fully reflect type of robust indicators that the Project could have adopted at entry to measure progress towards achieving the objectives. Second, readiness to implement by the Borrower was delayed in part due to inadequate preparation and training of the IA on World Bank on fiduciary procedures. Although these risks were highlighted in the Project Proposal and described, more attention could have been paid to: (i) support provided to developing and adopting the OM in advance of effectiveness, including guidance on disbursement procedures and timelines required for the PIU and the Central Bank of Tunisia; (ii) support provided to preparing and recruiting TORs for key PIU staff, despite budget constraints in advance of approval; and (iii) launch of pilot activities to test implementation activities. Despite these limitations, Ensuring Quality at Entry is rated Moderately Satisfactory. (b) Quality of Supervision Rating: Satisfactory 17 68. Bank implementation support missions were undertaken routinely and included day-to- day support provided by both World Bank headquarters- and the Tunisia-based task teams, including the TTL, financial management and procurement specialists. Throughout implementation, support missions included a mix of skill sets, including (i) operational manual and implementation support provided by the Operations Officer, Operations Analyst; (ii) financial management training and support provided by two Financial Management Specialists over the course of the project; (iii) procurement training and support provided by two Procurement Specialists over the course of the project; (iv) local coordination and day-to-day implementation support particularly over the first half of the project provided by a Project Management and Social Development Consultant; (v) monitoring and evaluation training and support provide by a Social Protection and Labor Economist; and (vi) additional World Bank support provided through participation by the PIU and GoT project staff in World Bank-financed capacity-building and knowledge-sharing events, including the Middle East and North Africa (MNA) Employment and Social Safety Nets Community of Practice (videoconferences and materials) during 2012-2104 and the Brazil South-South Learning Forum in 2014. Quality of Supervision is rated Satisfactory. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Satisfactory 69. Overall Bank performance is rated Moderately Satisfactory based on the ratings of Moderately Satisfactory for Quality at Entry and Satisfactory for Quality of Supervision. 5.2 Borrower Performance (a) Government Performance Rating: Moderately Satisfactory 70. Government performance (Prime Ministry, and Cabinet at the central level; and, local Governors at the local level) is rated as Moderately Satisfactory given the positive results which the Project achieved despite early delays undertaking the necessary legal steps to making the project effective and supporting its launch of activities. To some extent, these delays were attributed to numerous Government reshuffles and lagging capacity, as Tunisia grappled with broader political transition just after the January 2011 revolution, which relatively stabilized by 2013. The close involvement and guidance by local Governors in coordination with appointed regional ministerial representatives suggests that coordination was generally adequate. (b) Implementing Agency or Agencies Performance Rating: Satisfactory 71. Implementing Agency (MoSA implementing through UTSS) performance is rated Satisfactory. The Project, being a pilot, innovative initiative in several respects, witnessed a number of implementation challenges at the onset due to having to develop and build capacity among a new PIU. However, UTSS as the implementation agency, played a critical role in building capacity to overcome these challenges. These challenges were largely overcome after the first year of implementation, with capacity now strengthened for future operations. Also important to note is that based on the mid-term evaluation, of the 58 NGOs enlisted in the Project, 77 percent of them were new NGOs that were created after the 2011 revolution. This signified that most NGOs that participated in the Project, as in most of Tunisia, have little or no experience in project management (as revealed in on-going sub-project evaluations). Low capacity delayed project implementation somewhat, a constraint that will gradually recede as NGOs gain more experience over time. (c) Justification of Rating for Overall Borrower Performance 18 Rating: Moderately Satisfactory 72. Overall Borrower performance is rated Moderately Satisfactory based on the ratings of Moderately Satisfactory for Borrower Performance and Satisfactory for Implementing Agency Performance. 6. Lessons Learned Overall Main findings 73. The Project was implemented effectively in a tri-partite approach involving local government, civil society and the private sector. It successfully funded “demand -driven” paid apprenticeships, chosen and designed by community, local civil society groups, training centers and firms and which focused on delivering basic community services but also generating livelihoods in local markets Specific Lessons Learned from this Operation 74. Through testing new approaches, the Project highlighted certain key lessons learned in terms of: (i) implementation arrangements, (ii) mechanisms to boost employability through delivering basic services, and (iii) participatory approaches to build social cohesion and address accountability of service delivery in fragile contexts to various degrees. 75. Implementation through public-private partnerships is feasible in new contexts, assuming contracting and monitoring are sound. In terms of implementation arrangements, the Project showed that public-private partnerships can support the implementation of employment- generating, cash-for-service programs and projects. Despite early delays in setting up the legal institutional arrangements of the Project and training a new PIU on Bank procedures, this Project exceeded targets and nearly fully disbursed all grant funds. Moving forward, monitoring and results-based contracting can be gradually strengthened to boost the quality and effectiveness of implementation further. 76. Employability through delivering basic services can help boost short-term employment opportunities in specific sectors. In terms of employability, the Project shows the importance of adapting designs to the profile of the unemployed and local markets. The gaps in local service delivery provided short-term employment opportunities to semi-skilled beneficiaries. As a result of the training and experience provided, the preliminary follow-up survey at six months after program exit of a sample beneficiaries also showed that this experience in transferable, as a share of beneficiaries found wage employment or became self-employed following program exit. Moving forward, these efforts need to be coupled with boosting private investment more strategically and more rigorous training for ensuring sustainability in those sectors that led to employment. Broadening this kind of coordination help links beneficiaries to new employment opportunities. 77. Participatory approaches can help foster social cohesion and accountability in service delivery to varying degrees. Finally, in terms of participatory approaches to build social cohesion, the Project showed the importance of involving local actors in selecting and targeting local development needs and solutions. Giving greater choice and involvement to key stakeholders at the local level, including public authorities, the private sector and civil society groups helped match demand and supply in identifying the most viable, needed opportunities. However, greater local capacity in terms of technical know-how and financial resources and more systematic coordination between stakeholders are needed moving forward. 19 78. The lessons learned from the Project are relevant to the design of future social protection and labor programs and reforms, notably in terms of: (i) introducing productive social safety net (SSN) programs as part of Tunisia’s existing SSN framework, (ii) consolidating productive SSN and active labor market programs (ALMP) through consolidating targeting for social protection and labor (SPL) programs between the Ministry of Social Affairs (MOSA) and the Ministry of Employment and Vocational Training (MEVT), and (iii) public-private partnerships at the local level for SPL programs, notably productive SSN and active labor market programs (ALMPs). 7. Comments on Issues Raised by Grantee/Implementing Agencies/Donors (a) Grantee/Implementing agencies N/A (b) Cofinanciers/Donors N/A (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) N/A 20 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD (USD millions) Appraisal millions) PILOT SERVICE DELIVERY 4,303,000.00 4,300,000.00 99 SUB-PROJECTS SERVICE DELIVERY SUB- 431,200.00 430,000.00 99 PROJECT TRAINING PROJECT MANAGEMENT, MONITORING AND 264,900.00 270,000.00 98 EVALUATION (PIU SUPPORT) Total Baseline Cost 4,999,100.00 5,000,000.00 Physical Contingencies 0.00 0.00 0.00 Price Contingencies 0.00 0.00 0.00 Total Project Costs 0.00 0.00 Project Preparation Costs 0.00 0.00 .00 Total Financing Required 0.00 5,000,000.00 (b) Financing Appraisal Actual/Latest Type of Percentage of Source of Funds Estimate Estimate Cofinancing Appraisal (USD millions) (USD millions) State and Peace Building Fund 4.90 5.00 .00 21 (c) Disbursements by Category Categories Initial Initial Payment as of Payment as of Payment as Payment as Cumulative Payments Remaining Allocation (1) Allocation + 31/12/2013 30/06/2014 of 31/12/ of 31 July Payment as of / balance at U$ 15% 2014 U$ 2015 U$ 31 July 2015 Allocation project (2) / (1) closing (%) (1) Cash transfer to 4,030,000.00 3,884,500.00 1,306,336.28 1,263,166.00 243,813.24 828,925.27 3,642,240.79 93.76% 242,259.21 Beneficiaries (2) Goods 205,000.00 235,750.00 24,763.35 161,017.24 24,947.49 26,687.93 237,416.01 100.71% -1,666.01 (3) Consultant 645,000.00 741,750.00 283,431.03 157,535.00 213,236.04 84740.31 738,942.38 99.62% 2,807.62 Services , training , workshops and audits (Services consultants, formation, workshops, audits) (4) Operating Costs 120,000.00 138,000.00 7,576.38 32,566.98 39,268.89 56,617.10 136,029.35 98.57% 1,970.65 Total 5,000,000.00 5,000,000.00 1,622,107.04 1,614,285.22 521,265.66 996,970.61 4,754,628.53 95.09% 245,371.47 23 Annex 2. Outputs by Component Component (as per PAD) Output Component 1: Pilot Local Service Delivery Sub-  All facilitators have been trained Projects  PIU UTSS has disseminated  Capacity building for community facilitators has communication to local UTSS, been conducted local associations and community All facilitators have been trained communication outlets.  Local communication strategy has been  Sub-projects have been developed (messaging, outlets, and materials for implemented. calls for proposals and job opportunities and roster of local associations has been created)  40 sub-projects are initiated in Tataouine and Medenine (or 4,000 workers have been employed) Component 2: Local Service Delivery Sub-Project  All workers have received training Training certification for service delivery  Service delivery skills training certification has sub-projects. been granted to employed workers  At least one participatory  Local participatory monitoring assessments monitoring report per sub-projects have been disseminated publicly (available on has been published. UTSS website)  All facilitators have been recruited.  Local facilitators have been selected and training for service delivery has been conducted. Component 3: Project Management, Monitoring and  All PIU staff have been trained Evaluation  Improved M&E system in place  Training has been conducted on Operations  Evaluation reports are routinely Manual, capacity building, and operational published planning  Workshops arranged for  Establishment of M&E framework and data dissemination of mid-term and collection plan for future scale-up final evaluation reports  Evaluation, lessons documentation and planning for subsequent rounds has been conducted  Mid-term and final independent evaluation and dissemination 24 Annex 3. Economic and Financial Analysis While there was no ERR/IRR estimated at appraisal, the Project’s investments yielded relatively high returns as compared to traditional, non-workfare based social safety nets. The value of the Project lies its capacity to increase productivity of social safety net beneficiaries and improve local community development incrementally to a greater extent than do traditional social safety nets. Given availability of the data to date, a basic framework is proposed here for evaluating cost-effectiveness based on the additional benefits accrued in terms of earnings, and more broadly, community externalities. The expectation was that within 1 year of program exit, at least a share of these beneficiaries would be expected to have found income-generating activities, as a result of the skills, work and collaborative experience gained. These activities would provide at least the equivalent of Tunisia’s minimum wage (286 TND per month, which is also the average for self - employed, low-skilled earners). The program would cost the equivalent or marginally more than traditional social safety nets, but which would increase that overall cost-effectiveness of public spending due to (i) incrementally more positive externalities to local community development, services and infrastructure, and (ii) boosting purchasing power and local economic growth. A more rigorous impact evaluation in the future may help collect the detailed data required for a full analysis in the future. The comparator is the current social safety net program in Tunisia (PNAFN). PNAFN provides benefits to 235,000 heads of households, or on average TND 130 per month per household over the course of 2012-2015 when this project was implemented. Over the course of 18 months, therefore, the average head of household would have received approximately 2,340 TND in total. Overall, during the period of 2012-2015, the Government spent approximately 420 million TND per year on PNAFN, or the equivalent of 0.4 percent of GDP. In terms of generosity, these cash transfers amount to nearly 21 percent of average household expenditure. The SPF Project provided work experience and 220 TND per month for six months to one person per household. For the first six months of the program, beneficiaries received a total of TND 1320. Among the sample of 1430 beneficiaries surveyed within 1 year of program exit, 13 percent had been employed for at least six months, or approximately 186 beneficiaries. Assuming this rate remains constant, of the 6180 total beneficiaries, this means that 803 beneficiaries were working for at least 6 months at minimum wage within 1 year of program exit. 286 TND per month is the equivalent of approximately 30 percent of total household expenditure. These benefits also indirectly provider broader benefits not monetized or taken into account in this analysis due to limited data. For example, as a result of one such sub-project in a remote oasis town, non-beneficiary community members opened microenterprises to cater to beneficiaries who were now earning. In principle, these dynamics boost broader community and economic development over time. Were the project to be scaled-up to working-age, able-bodied beneficiaries of PNAFN, this would equate to approximately 30 percent of current beneficiaries, or 70,500 heads of households (leaving 164,500 who would continue receiving unconditional cash transfers). On average, the workfare program, assuming the same design, would cost 94 million TND per year in cash transfers and an additional 14 percent in training and operating costs, or a total of approximately 107 million TND. The return on investment (ROI) in terms of productivity as measured by additional earnings would assume that at least 13 percent of these beneficiaries are working after program exit at (at least) minimum wage for six months, or earning a total of nearly 16 million TND. For the 107 million TND investment, this yields a rate of return (ROR) of approximately 15 percent. 25 These additional benefits are expected to accrue at a financial cost that is nearly equivalent to the current PNAFN budget and in line with average global best practice. Adding the cost of the productive social safety net (107 million) with PNAFN only among the 70 percent remaining (294 million), gives a total of budget of 401 million TND (or just under 0.4 percent of GDP). This total amount is lower than the current social safety net spending of 420 million TND, and is expected to have a higher return on investment. Even under assumptions that the budget would be marginally higher, the returns on investment are likely positive. This level of public spending on a scaled-up program, the coverage and transfer amount would be in line with that of well- performing programs productive social safety net programs elsewhere, such as Brazil’s Bolsa Familia. The results of this basic analysis are tentative, given limitations related to the assumptions and the availability of data, but are nonetheless expected to hold directionally under other scenarios. Although a full impact evaluation of the project has not been done currently Tunisia’s PNAFN program does not have exit mechanisms, so most beneficiaries are assumed to remain in the program. While in principle their inclusion is based on not having other sources of income, this criteria is self-reported, so in this analysis it is assumed they mainly rely on monthly cash transfers for income. Overall, the total cost-effectiveness of the Project needs to be viewed in terms of both the increases in productivity (earnings) and broader, non-monetized externalities to local community development. The program is therefore relatively cost-effective in comparison to the overall individual and community benefits. 26 Annex 4. Grant Preparation and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending/Grant Preparation Hebatalla Elgazzar Senior Economist GSPDR Team Leader Shirley Foronda Financial Management Specialist GGODR Team Member Jean-Charles Marie De Senior Counsel LEGAM Team Member Daruvar Maged Mahmoud Hamed Regional Safeguards Adviser OPSOR Team Member Walid Dhouibi Procurement Specialist GGODR Team Member Brooks Fox Evans Consultant GSPDR Team Member Karen Majli Vega-Coronel Team Member Mehdi El Batti Financial Management Specialist GGODR Team Member Christina Wright Operations Officer GEDDR Team Member Peter McConaghy Financial Sector Specialist GFMDR Team Member Supervision/ICR Aissatou Maisha Dicko Operations Analyst GSPDR Team Member Jean-Charles Marie De Team Member Daruvar Senior Counsel LEGAM Amr Moubarak Social Protection Economist GSPDR ICR Co-author Walid Dhouibi Procurement Specialist GGODR Team Member Suzanne Essama Operations Officer GSPDR ICR Co-author Mehdi El Batti Financial Management Analyst GGODR Team Member (b) Staff Time and Cost Project: Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending 0.0 0.0 Supervision/ICR Total: 36.78 248,628.84 27 Annex 5. Beneficiary Survey Results Background: An external beneficiary survey was conducted after the mid-term evaluation by a survey firm recruited by the PIU in line with terms of reference approved by the World Bank. Objectives: The survey was conducted to assess beneficiary, community and partner organizations’ perceptions of process and implementation of the project and, among beneficiaries, the relevance of the workfare sub-project and training modules. The assessment included qualitative focus groups and a quantitative survey among 1000 beneficiaries and 1000 community members. Main results: Perceptions of project benefits: Overall, the main benefits of the Project were perceived to be work experience gained by beneficiaries and the services delivered to communities. 75 percent of beneficiaries reported that they benefited by gaining new work experience. However, only 66 percent were satisfied by the content of the training module they received, mainly due to the lack of formal certification offered by the primarily informal training they received from individual instructors. Beneficiaries reported additional benefits in terms of local development (Table 2): (i) greater social cohesion and commitment to local community development; (ii) gaining entrepreneurial know-how, which led many beneficiaries and community members to form small micro-enterprises in areas where the project was implemented, given stronger purchasing power among beneficiaries; (iii) in terms of gender, female beneficiaries and community members expressed becoming more active publicly and socially as a result of sub-project community and follow-up activities; and (iv) development of new local market and social networks as a result of connections formed between and among beneficiaries, NGOs and local municipalities. Table 2 Benefits perceived by beneficiaries in terms of local development (%) Dimension % de ceux qui pensent que l’impact est positif L’intégration sociale de la femme rurale 73% Satisfaction des besoins au quotidien 68% Initiateur de nouveaux projets 73% Un coup de main donné aux institutions publiques 67% Une solidarité sociale à travers la sensibilisation de la 72% population à propos de certains problèmes Perceptions of process and implementation: In terms of implementation, partner NGOs reported having had adequate levels of technical and financial resources to support local sub-projects. However, the average duration of the training modules of 2 months was considered too short by beneficiaries, and most suggested at least 4 months would be needed in the future to fully acquire new technical trade-related skills. 28 An important negative effect reported by NGOs and beneficiaries was related to delays in receiving payments and cash transfers. These delays were due to lags in financial management and disbursement processing, largely due to familiarity with forms required and Central Bank processing timelines. Main recommendations: Moving forward, the main recommendations from the beneficiary survey were to: (i) strengthen the rigor and quality of the supplemental training offered to ensure technical and vocational skills are certified; (ii) strengthen the oversight and resources offered to partnering NGOs supporting supervision and training; and (iii) improve financial management and disbursement procedures for more rapid processing and payments. 29 Annex 6. Stakeholder Workshop Report and Results Objectives: A stakeholder workshop was organized by the PIU towards project closing, held on February 17-19 in Zarzis, Medenine. The main objectives of the workshop were to present the main goals and results of the Project and discuss ideas for collaboration moving forward, as well as to honor and present beneficiaries and local partner CSOs. Participants: All central and local government, PIU, beneficiary survey firm (Emrhod Consulting) and NGO implementation partners attended the workshop, as well as beneficiaries and other local NGOs, for a total of approximately 200 participants. Themes: The workshop was organized as three days of presentations, working sessions and awards ceremonies, highlighting the range of diverse projects and local community impacts. Key sessions included: (i) presentations from government authorities (Governors, Ministry of Industry, Ministry of Vocational Training and Employment and Ministry of Social Affairs) on sustaining interventions from the project; (ii) panel on operational mechanisms to support low-skilled unemployed youth; (iii) panel on enhancing vocational training; (iv) panel on mechanisms for collaborating with CSOs on professional development; (v) panel on supporting entrepreneurship among youth through local collaborations; (vi) A wrap up session on summarizing scale-up options and sustaining the program in the future. Results: The workshop resulted greater awareness of local and central decision-makers to project results, which further boosted interest in scaling up the intervention moving forward. Subsequently, a request for future World Bank support for scale-up was sent from MoSA to the Ministry of Investment and International Cooperation (MICI) in May 2015, as a follow-up to an earlier request sent at mid-term review (April 2014). As of August 2015, the request is being considered as a potential component of the Government’s forthcoming 2016-2020 National Development Plan. 30 Annex 7. Summary of Grantee's ICR and/or Comments on Draft ICR The Participatory Service Delivery Project for Reintegration in the governorates of Medenine and Tataouine is an initiative of the Tunisian government, which is supported by the World Bank with a US$ 5 million grant. The project was implemented over 30 months period. This project was conceived as an emergency response to provide additional income to young people from disadvantaged families in these two governorates and at the same time improve the employability of a class of young people with low-skills (some education and no diploma). As a pilot project, its implementation was a true experiment in terms of intervention approaches in the field of learning, delivery of basic services, mobilization of civil society and mobilization of young people and women in community and proximity approaches. Thus, the project was implemented through the mobilization of 66 associations and through 92 sub- projects in the field of social assistance, education, basic health and the environment. The actions undertaken were distributed among services and works and have contributed to the facilitation of the access to information and services to the communities targeted by the project. Furthermore, the project has tried to answer the problem of employment for unqualified youth with little education through informal training and exercise on site the theoretical knowledge acquired in class and that, through the delivery of basic services to marginalized groups and especially with special needs. The statistical study conducted as part of the project showed that the majority (87%) of direct beneficiaries had only primary or secondary education and about 74% of them are older between 18 and 32 years. These direct beneficiaries were satisfied with their participation in the project; 41% of them were enticed by the new experience and more than two thirds (68%) reported having acquired professional experience during this project. While 54% found that the project helped them to develop their communication skills, and 41% say that the project helped them to find a job opportunity. Only 40% of beneficiaries have seized the opportunity to develop raw technical skills. Service Delivery: As for community members (indirect beneficiaries) their satisfaction with the services provided has been estimated at 68%. Participation in decision-making: 56% of direct beneficiaries say that the project gave them the opportunity to participate in a social activity while 47% of community members say they have participated in project activities. Lessons learned from the project: the project shows that the provision of opportunities to young people, to acquire knowledge and professional experience relating to a trade, improves their employability. Thus, to ensure high employability, training must be: 1) adapted to the profile of the unemployed and 2) tailored for local markets while involving different stakeholders at the local level i.e., the public services concerned by the technical specialty from the training, the professionals in the private sector and civil society. Moreover, these in-service training activities as they were tested in the project must be associated with a system of validation of competence by government technical services and the assurance of complementary training in management, marketing, life skills, entrepreneurship and financial education. 31 And to complete the process, it is essential to have a fund to finance income-generating activities with ideas deriving from training. Another important aspect that has been treated in the project is "community services". While these services have helped to visualize the impact of the project on the communities and to ensure their participation, they should, however, be accompanied by an upstream mechanism guaranteeing the sustainability of their achievements. Moreover, these services related to matters which fall within the priority concerns of the targeted communities in both targeted governorates and services provided as part of the implementation of sub-projects, have highlighted the need, even the urgency of their consideration in the intervention of public services programs. This will justify the need for the institutional anchoring of the project in the future. The associations have played an important role in the project and that, in formulating participatory proposals that meet the expectations of communities and defend their interests with local authorities and mobilizing various partners (public, private and OSC) for the implementation and success in their projects. These associations need to be more implicated in the training process the non- graduates of and in any employment initiative at the local level. The partnership with public facilities and some private sector professionals should bring added- value to the sub-projects but prevent that it follows a standard for random and specific and ad hoc initiatives. It is recommended that this partnership be managed by an incentive system and therefore an obligation for results for different stakeholders. As contribution to the strengthening of the local governance, the project acted through associations to encourage participation and citizen engagement with these different levels; namely, information, consultation, cooperation, implementation of actions and evaluation of service delivery. The project has gone through difficulties with administrative procedures (period devoted to the amendment of the letter of grant agreement) and disbursement procedures that have influenced the achievement of project objectives, but these difficulties have been overcome through the flexibility demonstrated by the project management at the World Bank. Therefore, and to minimize these difficulties, it is recommended, on one hand, to involve the implementing agency in the discussions on the Agreement between the Bank and the Beneficiary and to require that the process of direct payment via the system Client Connection be well stipulated in the agreement. Furthermore, and for similar projects that are based on the participatory approach, it is recommended: (1) to provide more time for the implementation period and (2) to devote more funding to the project management components (20% allocated for this project were not sufficient). Also, it is essential to provide a more transparent beneficiaries targeting mechanism that should be accessible through online registration or through an independent office and it, along with an institutionalized and effective complaints handling mechanism. Project management has generated a large data flow for which it is essential to make provision for the establishment of a monitoring web application integrating the collection, analysis, verification / validation and centralization of the information on project performance indicators. 32 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders None 33 Annex 9. Additional Supplemental Information Annex 9.A. Additional Information on Financial Management (FM) Arrangements The Project’s FM arrangements were deemed adequate and provided (i) accurate financial information and (ii) reasonable assurance that project funds are being used for the purpose intended. However, there were initially a few challenges that were addressed in order to fully comply with Bank’s fiduciary requirements. They namely were: (a) Systematic delays in submission of financial reports (IFRs, Audited PFS) that were progressively reduced during implementation. The PIU was trained on several occasions (joint fiduciary training, fiduciary clinics and specific on-demand trainings), which enabled the PIU to submit IFRs on time particularly during the last year of the project. Nevertheless some issues with the audited PFS persisted: (i) submission delays albeit reduced; and (ii) quality issues regarding audits that were discussed with the PIU and that led to the revision of the auditor’s TORs based on standard World Bank TORs. The final audit is expected to be submitted by the end of August 2015, which covers the grace period (March 31 through July 31, 2015). (b) The Project’s budgeted activities per components and categories were revised as the level of commitments for some categories exceeded planned amounts, namely operating costs and training. As a result, the submission of requests for overdrafts to the World Bank were recommended to allow the PIU to settle final payments for these categories as there were no Government funds available otherwise. (c) Disbursement of grant proceeds was made exclusively through the advance to the designated account via the submission of withdrawal applications by the Central Bank of Tunisia on behalf of the PIU, following the procedures and guidelines already in place and used for existing projects financed by the Bank. The PIU requested a ceiling increase for the designated account from USD 500,000 to USD 900,000 to accommodate project expenditures, which was approved and reflected in the disbursement letter. Replenishment of the designated account by the Central Bank of Tunisia had been relatively slow according to the PIU and the ceiling increase allowed delays in payments to be reduced. The disbursement ratio reached the amount of USD 4.756 million as of August 21, 2015, or 95.12% of the total grant amount USD 5 million. 34 Annex 9.B. Detailed Description of Achievement of Objectives by Key Project Development Indicator (i) Indicator 1: At least 4,000 workers are employed to implement projects. Based on the PIU January 2015 Implementation Report and information from the Project M&E database, a total of 6,182 beneficiaries had been enrolled in 70 local sub-projects. 91 percent of beneficiaries had either a primary or secondary education level, suggesting that the right target groups were included (Figure 1) and most were youth (60 percent) (Figure 2). The total number of direct beneficiaries exceeded the initial target number of 4,000 beneficiaries. This equates to a total of approximately 6182 persons x 6 months per person x 15 days per month (on average) of person-days worked, or 556,380 person-days worked. Figure 1 Distribution of beneficiaries by educational level (%) 60% 49% 50% 42% 40% 30% 20% 10% 7% 1% 2% 0% Analphabète Enseignement de Niveau Formation études base secondaire professionnelle supérieures Source: Tunisia SPF Project Database, November 2014. N=4260. Figure 2 Distribution of beneficiaries by age group %() 45% 40% 35% 30% 25% 20% 40% 34% 15% 10% 15% 5% 11% 0% moins de 25 ans entre 26 et 32 ans entre 33 et 39 ans plus de 40 ans Source: Tunisia SPF Project Database, November 2014. N=4260. 35 In terms of total enrolment by sub-project, most beneficiaries were enrolled either in “social assistance” subprojects (37 percent of all beneficiaries, followed by “environmental” subprojects (31 percent) (Figure 3). In terms of sub-project type, the vast majority of sub- projects have been “social protection and local development” in nature, followed by “education” and “health” sub-projects. The sub-projects have been selected through participatory approaches where CSOs, in consultation with local citizens, propose projects (as described in the project document and PIU mid-term implementation report). Of the 24 sub-projects in Tataouine, nearly 55 percent of sub-projects are classified as “social protection” projects (such as support to older persons and those with handicaps), 27 percent as “education” projects (such as basic civil works at primary schools) and 18 per cent as “health” projects (notably reproductive health awareness and education, as well as environmental health projects such as those that rehabilitate oases and wells). Figure 3 Distribution of beneficiary enrolment by sub-project sector (%) Sante 9% Social Assistance (Services, Textiles, Education Maconnerie) 22% 37% Environment (Commerce agri-peche, infrastructure) 31% Source: Tunisia SPF Project Database, November 2014. N=4260. In Medenine, 74 percent of projects were “social protection” projects, 19 percent have been “health” projects and 7 percent have been “education” projects. Data from sub -projects showed that that sub-projects mainly comprised service delivery, with some involving public works. 78 percent of sub-projects have been service delivery projects (not involving civil works) and 22 percent were minor civil works projects (such as landscaping and painting considered category ‘C’ projects). (ii) Indicator 2: At least 60 percent of workers perceive improvement in job skills as a result of sub-project training. Based on the PIU January 2015 Implementation Report and information from the project M&E database, 54 percent of workers perceive an improvement in jobs skills as a result of sub-project training. Over 75 percent of beneficiaries rated their workfare “plus” on -the- job training experience favorably, compared to an average of 16 percent among low-skilled beneficiaries of public employment programs. The single biggest reason was “work experience”, cited by 72 percent of SPF beneficiaries. (iii) Indicator 3: At least 30 percent of workers employed are women. Based on the PIU January 2015 Implementation Report and information from the Project’s M&E database, 70 percent of workers employed were women. The sectors with 36 the highest shares of women were education and social assistance, with at 77 and 75 percent of all beneficiaries, respectively (Figure 2). Figure 4 Distribution of beneficiaries by gender and by sector (%) % Male % Female Total 30% 70% Health 31% 69% Education 23% 77% Environment (Agri-Fishery commerce, 41% 59% Utilities) Social (Services, Textiles, Masonry) 25% 75% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Source: Tunisia SPF Project Database, November 2014. N=4260 (iv) Indicator 4: At least 12,000 community members receive services through local sub- projects. Based on the PIU January 2015 Implementation Report and information from the Project’s M&E database, the target number of indirect beneficiaries has also largely been surpassed (62,669 achieved versus 12,000 targeted). The number of indirect beneficiaries is based on estimates of catchment areas rather than direct surveying or monitoring due to budget and feasibility in implementation constraints, which may have inflated the number of indirect beneficiaries. In the future, additional resources and tools could help improve monitoring and evaluation such as the use of mobile phones and digital technology. (v) Indicator 5: At least 60 percent of community members perceive services received respond to local needs. Based on the PIU January 2015 Implementation Report reflecting information from the project M&E database collected by the PIU and partner CSOs through local beneficiary assessments, 67 percent of community members reported perceiving that services responded to local needs on average. Qualitatively, communities, included neighboring households that benefited by services provided by project beneficiaries, and authorities alike found the project opened up career opportunities in sectors and services with unmet needs, either because public services were unable to reach certain households, or because the services had never been created to begin with. Given that the project did not specifically target improving the quality of services (only basic supplemental access), these results are considered realistic and the assessment credible. 37 (vi) At least 60 percent of workers and community members perceive greater involvement in local decision-making regarding service delivery. Based on the PIU January 2015 Implementation Report reflecting information from the project M&E database collected by the PIU and partner CSOs through local beneficiary assessments, 56 percent of workers and 47 percent of community members perceive greater involvement in local decision- making regarding service delivery. 8 Given that the project did not specifically target radically strengthening local governance and decision-making, these results are considered realistic and the assessment credible. 8 Source: PIU (UTSS), in partnership with local associations. 38 Annex 9.C. Additional Figures on Employment and Insertion at Six Months Following Project Exit Figure 5 Follow-up tracer data on employment status at six months following project exit (%) Number of beneficiaries in sub-projects Job insertion rate, unadjusted (%) 1600 Number of beneficiaries surveyed 1400 1200 1000 800 600 400 50% 6% 3% 200 7% 13% 0 Source: Tunisia SPF Project Database, November 2014, Preliminary tracer survey at 6 months following project exit, N=1360. Figure 6 Employment status by type (%) Self-employment Salaried employment 80% 69% Share of employment within 6 70% 65% 60% 50% 50% months (%) 50% 40% 35% 31% 30% 20% 10% 0% Total Medenine Tataouine Region Source: Tunisia SPF Project Database, November 2014, Preliminary tracer survey at 6 months following project exit, N=183 employed beneficiaries. In the Project, for sub-projects where these networks were the strongest and best coordinated, the transition from training to jobs was the smoothest, reaching a 90 percent job placement rate for one such project in fishery, capturing 64 percent of all beneficiaries employed among the sample 39 (Figure 7). As project beneficiaries were predominately female, most of the employed after the project were also female (Figure 8), with a small advantage in Tataouine, suggesting that the project did not disproportionately benefit males in finding employment opportunities. This results is also promising for future efforts to boost female labor force participation. Beneficiaries and their new professional networks had created unforeseen opportunities, despite the fact that the project had not explicitly included specific support to job placement or entrepreneurship. Figure 7 Employment by sector (%) Maconnerie; Agro- Textile/commerce business (Auto- (Auto-emploi et emploi et Salarié) Salarié) 3% 7% Commerce général (Auto-emploi et Salarié) 6% Agro-business Pêche (Auto-emploi) (Salarié) 20% 64% Source: Tunisia SPF Project Database, November 2014, Preliminary tracer survey at 6 months following project exit, N=183 employed beneficiaries. Figure 8 Employment status by gender and region (%) % Male % Female Tataouine 30% 70% Medenine 40% 60% Total 37% 63% 0% 20% 40% 60% 80% 100% Source: Tunisia SPF Project Database, November 2014, Preliminary tracer survey at 6 months following project exit, N=183 employed beneficiaries. 40 Annex 10. List of Supporting Documents Emrhod Consulting (2014). Rapport d’evaluation finale. Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2014). Manuel d’Operations. Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2014). Plan de Passation des Marchés. Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2014). Rapport d’Etat d’Avancement, mai 2014 (mi- terme), Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2015). Note d’Etat d’Avancement, janvier 2015 (pré- clôture), Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2015). Tableaux de Bord des Interventions à Medenine et Tataouine, janvier 2015. Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2015). Tableaux des Bénéficiaires à Medenine et Tataouine, janvier 2015. Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2015). Suivi du Budget, juin 2015. Projet des Services Participatifs pour la Réintégration. Union Tunisienne de la Solidarité Sociale (2015), Projet des Services Participatifs pour la Réintégration, Site-Web : http://www.buildingcapacities.com/ World Bank (2012). Project Appraisal Document (PAD): Tunisia Participatory Service Delivery for Reintegration Final Project Proposal. Project number P127212. World Bank (2012). Tunisia Participatory Service Delivery for Reintegration Implementation Status and Results Report, Sequence 01. Report No: ISR7697 (Archived 26-Sept-2012). World Bank (2013). Tunisia Participatory Service Delivery for Reintegration Implementation Status and Results Report, Sequence 02. Report No: ISR9240 (Archived 21-May-2013). World Bank (2013). Tunisia Participatory Service Delivery for Reintegration Implementation Status and Results Report, Sequence 03. Report No: ISR13326 (Archived 30-Dec-2013). World Bank (2014). Tunisia Participatory Service Delivery for Reintegration Implementation Status and Results Report, Sequence 04. Report No: ISR16424 (Archived 12-Nov-2014). World Bank (2014). Restructuring Paper on a Proposed Project Restructuring of the State and Peace-Building Fund Participatory Service Delivery for Reintegration TF011069 April 20, 2012 to the Republic of Tunisia. Report No: RES13680. Annex 11. Country Map 41 42