Document of The World Bank Report No: 33502-CV IMPLEMENTATION COMPLETION REPORT (IDA-32240 PPFI-Q0710 TF-29258 IDA-32241) ON A CREDIT IN THE AMOUNT OF SDR 14.8 MILLION (US$20.1 MILLION EQUIVALENT) TO THE REPUBLIC OF CAPE VERDE FOR A SOCIAL SECTOR DEVELOPMENT PROJECT September 28, 2005 Human Development II Country Department 14 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective April 1999 (Appraisal)) Currency Unit = Escudos 94.515 Escudos = US$ 1 US$ 1 = 0.01058 FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS AfDB Africa Development Bank INE National Statistics Institute ACDI/VOCA-DAgricultural Cooperative Development INERF Instituto Nacional de Engenharia GASP International-General Directorate of Rural e Florestas (National Institute Agriculture, Silviculture and Fisheries of Rural Engineering and Forestry) AGETIP Agence d'exécution de travaux d'intérêt KPI Key Performance Indicator public (Executing Agency for Public Works) CAS Country Assistance Strategy MIS Management Information System CVE Cape Verdian Escudos MOD Maître d'ouvrage délégué (Delegated Contract Management) DCMA Delegated Contract Management MTR Mid-term Review Agency D.G. Director General of AGECABO NCB National Competitive Bidding E.U. European Union NGO Non-Governmental Organization FAM Fonds d'appui aux municipalités NPAP National Poverty Alleviation (Support Fonds to Municipalities) Program FAIMO Frente a Alta Intensidade de Mano de PCU Program Coordination Unit Obra (Labor Intensive Funds) GoCV Government of Cape Verde PPF Project Preparation Facility GTI Inter-Municipal Technical Cabinet PROMEF Education and Training Consolidation and Modernization Project (Projecto de Consolidação e de Modernização da Educação e Formação) GTM Municipal Technical Cabinet PRSP Poverty Reduction Strategy Paper ICB International Competivie Bidding PRSC Poverty Reduction Support Credit IDRF Inquérito às despesas e receitas as SSDP Social Sector Development Project familiares (Expenditures and Income Household Survey) IFAD International Fund for Agricultural TOR Terms of Reference Development Vice President: Gobind T. Nankani Country Director: Madani M. Tall Sector Manager: Alexandre V. Abrantes Task Team Leader/Task Manager: Johanne Angers CAPE VERDE Social Sector Development CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 5 5. Major Factors Affecting Implementation and Outcome 14 6. Sustainability 16 7. Bank and Borrower Performance 18 8. Lessons Learned 20 9. Partner Comments 21 10. Additional Information 21 Annex 1. Key Performance Indicators/Log Frame Matrix 22 Annex 2. Project Costs and Financing 25 Annex 3. Economic Costs and Benefits 31 Annex 4. Bank Inputs 35 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 38 Annex 6. Ratings of Bank and Borrower Performance 39 Annex 7. List of Supporting Documents 40 Annex 8. Borrower's Contribution 43 MAP: IBRD 33383 Project ID: P000432 Project Name: Social Sector Development Team Leader: Johanne Angers TL Unit: AFTH2 ICR Type: Core ICR Report Date: September 28, 2005 1. Project Data Name: Social Sector Development L/C/TF Number: IDA-32240; PPFI-Q0710; TF-29258; IDA-32241 Country/Department: CAPE VERDE Region: Africa Regional Office Sector/subsector: Other social services (94%); Central government administration (6%) Theme: Improving labor markets (P); Social safety nets (P); Participation and civic engagement (S); Poverty strategy, analysis and monitoring (S) KEY DATES Original Revised/Actual PCD: 04/03/1997 Effective: 10/04/1999 10/04/1999 Appraisal: 02/11/1999 MTR: 10/09/2001 10/09/2001 Approval: 05/25/1999 Closing: 08/31/2003 03/31/2005 Borrower/Implementing Agency: GOVERNMENT OF CAPE VERDE/MINISTRY OF LABOR AND SOLIDARITY Other Partners: STAFF Current At Appraisal Vice President: Gobind T. Nankani Jean-Louis Sarbib Country Director: Madani M. Tall Mahmood A. Ayub Sector Manager: Alexandre V. Abrantes Nicholas Burnett Team Leader at ICR: Johanne Angers Tonia Marek ICR Primary Author: Maurizia Tovo; Johanne Angers; Serge Theunynck 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: U Project at Risk at Any Time: Yes 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The Project Development Objective was to contribute to poverty reduction among the 30 percent of the population living below the poverty line by: (a) assisting the Government of Cape Verde (GoCV) in the restructuring of the labor-intensive public works program (FAIMO) by creating an institutional capacity (AGECABO) to execute public works that are: (i) economically and/or socially useful; (ii) executed cost-effectively by the private sector; and (iii) based on labor-intensive techniques; (b) building the capacity of municipalities, communities and NGOs, in planning, implementing and monitoring cost-effective poverty alleviation interventions; and (c) establishing a central capacity to coordinate, monitor and evaluate the 1998-2002 National Poverty Alleviation Program (NPAP). The FAIMO system was instituted in the early 1980s as the main Government's strategy to provide food security through labor-intensive public works programs (FAIMO) that created jobs, particularly during the rainy season (July-August) when food is most scarce. Given the recurrent famines and the low-income levels at the time, the dependency of the poor on the FAIMO employment system was quite strong, particularly in the agricultural sector. Hence FAIMO was used predominantly as a welfare mechanism for the poor, which provided no incentive for improved productivity. With Cape Verde approaching middle-income levels, concerns for a sustainable approach to employment and improved productivity were the main reasons for Government's decision to decrease FAIMO dependency. The intention was to move from a centrally planned, forced account approach to one driven by community demand and handled at the municipal level. Project objectives evolved out of the NPAP, which formed an integral part of the broader National Development Plan (1997-2001), and were fully consistent with the overall country strategy of Cape Verde as well as with the Bank's 1997 Country Assistance Strategy (CAS, report no. 17206-CV). In particular, they (a) supported NPAP objectives to improve the coordination, allocation, and effectiveness of both government and donor resources to alleviate poverty, and (b) supported the priorities identified in the CAS, namely, sustaining macroeconomic stability, encouraging private sector development, improving human development --especially of women and children, and decentralizing responsibilities and resources to municipalities. However, the project objectives could have been expressed in a more streamlined fashion, without "layering" (one overall objective with three specific objectives), and it probably would have been better to limit the objectives to the three specific objectives, especially given the inherent difficulty of proving the contribution of a single project to a goal as ambitious and general as poverty reduction. 3.2 Revised Objective: The project objectives were not revised. 3.3 Original Components: The Social Sector Development Project (SSDP) consisted of two components, which were: Public Works and Employment Creation (US$16.6 million with original credit and US$4.0 million with supplemental credit). This component aimed at (a) providing a viable alternative for FAIMO workers; (b) decentralizing public works based on community-defined priority activities using private sector principles through a delegated contract management approach; (c) reinforcing NGO capacity to work with communities, as NGOs were to be contracted to mobilize poor communities and assist them in designing and implementing activities; and (d) building capacity of municipalities, NGOs, and small entrepreneurs through training and targeted technical assistance. The project, therefore, was to finance economic and social sub-projects, technical assistance, operating costs, and training. The objectives of this component - 2 - were to be achieved primarily by setting up a delegated contract management agency (AGECABO) drawing on the success and lessons learnt from the AGETIPs experience in Sub-Saharan Africa. Assistance to form local small-scale associations was to be provided to "FAIMO-type" workers with training in technical skills and contract management. Only sub-projects between US$20,000-800,000 equivalent were eligible for financing. Community activities costing less than US$20,000 equivalent could consist of self-help projects that aim at improving the living conditions of the poor especially in the areas of health, hygiene, education, environment, nutrition and food security. Capacity Building (US$1.5 million with original credit and US$0.4 million with supplemental credit). This component aimed at (a) providing support to the Program Coordination Unit (PCU) created in August 1998 for the implementation of the NPAP, whose main roles were to (i) coordinate the NPAP; (ii) liaise with donors to finance poverty reducing interventions; (iii) check eligibility of subprojects included in municipal annual programs and transmit them to a Program Approval Commission; and (iv) select implementation agencies for the execution of sub-projects; (b) finance limited studies and suveys (beneficiary assessments) to monitor the project impact, thus complementing the existing poverty monitoring activities of the Government; and (c) finance technical and financial audits. The project was to finance investment and recurrent costs of the PCU, technical assistance, training, equipment, and limited operating costs. The two components were clearly related to the specific Project Development Objectives. 3.4 Revised Components: No formal revision of the components took place. However, on June 20, 2001, about US$1.1 million of the original credit was reprogrammed towards financing the implementation of the National Household Income and Expenditure Survey (known as IDRF) in view of an urgent national need to develop a poverty profile to underpin the Poverty Reduction Strategy Paper (PRSP) when no alternative sources of financing, either domestic or foreign, were available. Given the close relationship with the development objective of increasing capacity to monitor poverty, the addition contributed significantly to the institutional development impact of the project. At mid-term review (October 2001) it was agreed that the employment indicators contained in the PAD had to be revised in light of the new disbursement projections, the wages for "FAIMO type" workers and the attainable levels of employment in the subprojects. In particular, it was aknowledged that original estimates for employment creation had been overly optimistic, as they assumed an 85 percent labor content, which is virtually impossible for anything but streetsweeping, while the most labor-intensive type of work in the portfolio, rehabilitation of cobblestone roads, has only a 45 percent labor content. The initial key project performance indicators were therefore reviewed (e.g., overall employment creation from 104,000 person/months to 28,271), but not formally revised, that is, the Credit Agreement was not amended. A formal revision of the project key performance indicators was instead done at the time of the supplemental credit, which was approved by the Board on December 16, 2003. This revision addressed the gross overestimate of employment creation by (a) lowering the target for overall employment creation from 104,000 person/months to 16,750 person/months (including the supplemental credit), and (b) lowering the target for direct employment of FAIMO-type workers from 20,500 to 13,800 (including the supplemental credit). At the same time, a new indicator was introduced to monitor activities planned to foster sustainability. 3.5 Quality at Entry: The Quality at Entry is considered unsatisfactory. - 3 - Strategic relevance. Bank's support clearly responded to the Government overall strategy on poverty and the Bank's CAS priorities as stated in their respective documents (see Section 3.1 above). Weaknesses in Quality at Entry. Several weaknesses in key areas of project design were identified by the ICR team. Perhaps the most glaring was the gross overestimation of the original Key Performance Indicators (KPI) related to employment creation, both overall direct employment and for FAIMO-type workers. Initially set at 104,000 person/months and 20,500 person/months, respectively, they were reviewed twice during project implementation in light of the obvious impossibility of reaching them --informally during the mid-term review, and formally at the time of the supplemental. While the final outcome in terms of employment creation compares favorably to similar projects, having started the operation with such overambitious targets gives the impression that preparation was not carried out as carefully as it should have been. Linked to this shortcoming was the lack of an adequate monitoring and evaluation system to ascertain the reliability of employment creation. The project collected an impressive amount of information, but there were no systems in place to exploit this wealth of information for monitoring and evaluation purposes. This may be due in part to efforts to keep recurrent costs low, and therefore limit AGECABO's and PIU personnel, but it appears that more thinking would have been needed from the onset on how to use the data collected. The capacity building component as well could have benefitted from more careful definition. The original project design described in Annex 2 of the PAD envisaged training for the PCU staff (topic not specified) as well as limited training for NGOs to facilitate implementation of Component 1. On the other hand, the section on expected benefits included "capacity building of municipalities and participating NGOs in planning, monitoring and financial accounting [for] a more rational and efficient approach to social sector investments". Including a KPI related to capacity building would have facilitated monitoring and assessment of this component. Finally, anther weakness found in quality at entry is the quality of the economic analysis presented in the project document. Choice of Approach. The approach to be adopted for implementing sub-projects was the origin of much debate between the Bank and the Government. While the Bank recommended the establishment of an Agetip-type outfit (AGECABO) based on demonstrated successful experience in the sub-region, Government argued for open competition among the various national entities (e.g., firms, NGOs) to participate as potential Delegated Contract Management Agencies (DCMAs). During negotiations, it was decided that any eligible DCMA besides AGECABO be allowed to execute Component 1, provided that the said DCMA was selected in accordance with the Bank's guidelines through a competitive process of quality-cost based method of selection. Given the fact that Capo Verde at the time was transitioning from a socialist to a market-based economy, a delegated contract management agency was probably the best compromise to support the development of a nascent private sector without sacrificing implementation effectiveness. In addition, the choice of supporting the decentralization process while empowering civil societies and municipalities was sound. Readiness for Implementation. Key preparatory activities supported by a PHRD trust fund (Yen 16.6 million) and a PPF (US$750,000) helped finance: studies, project manuals, capacity assessments, pilot activities in four municipalities, the establishment of AGECABO and of a project management information system. However, the component on capacity building could have benefited from greater definition at its onset, particularly on the type of training and targeted groups (see above), considering that the project was to be originally implemented within a relatively short period (four years) at a time when little institutional capacity existed at the central and, more importantly, at the decentralized levels. The PCU was created by a Government Resolution in 1998 to design, coordinate, monitor and ensure the execution of the NPAP. Therefore, some of its staffing was already in place prior to effectiveness. Delays in project start-up occured due to government lack of familiarity with the delegated contract management approach, work overload of the PCU/NPAP (which was simultaneously involved in the preparation of projects financed by - 4 - other donors), changes in the PCU/NPAP leadership, local government turn-over due to municipal elections in April 2000, as well as few factors specific to the project such as delays in the signature of the first convention with AGECABO and related dibursement of the first tranche. Candid Assessment of Risks. The risks identified were: (a) AGECABO would not develop into an effective and efficient agency and no other DCMA would be available; (b) insufficient and ineffective coordination of the various donors; (c) inadequate cash flow available by donors and Government to support program activities; (d) delay in the implementation of FAIMOs reform; (e) no sharing of information among government agencies and other partners; (f) difficult collaboration between political parties that may hinder devolving responsibilities to municipalities --particulalry, those controlled by the opposition; and (g) unwillingness by certain municipalities to delegate certain responsibilities to DCMAs. The ICR team concurs that the project risks were appropriately identified and the minimization measures were correct. During project implementation such measures were employed and gave good results as none of the risks was ever fully realized. The only risk partially realized was the one related to municipalities unwillingness to delegate certain responsibilities to outside agencies, but the final evaluation suggests that most municipalities had come around and readily recognized the advantages of the DCMA approach. Bank's safeguard policies. Environmental risks were adequately addressed through systematic environmental assessments of all sub-projects according to the Environmental Guidelines for Social Funds, October 1998. Environmental analysis evaluation grids were filled in for each sub-project and if further analysis were warranted, the local contrators and/or municipalities were to ensure that mitigating measures would be undertaken in a timely fashion. The approach was reviewed at the time of the supplemental in late 2003, and it was found acceptable. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The overall outcome of the project is moderately satisfactory. The project objectives remain highly relevant in today's context. At the time of project preparation, the reduction of unemployment was correctly identified as an effective way to alleviate poverty on the basis of the information available (the 1988-89 household budget survey). Analysis carried out with the 2001-02 data from the IDRF confirms that unemployment is still a crucial determinant of poverty, with the poverty incidence being 49 percent among the unemployed, as opposed to 37 percent among the general population (see Poverty Diagnostic green cover, para. 3.6). Beneficiary assessments carried out in 2003 and 2005 confirm that unemployment is largely perceived (i.e., by over 80 percent of respondents) as one of the main causes of poverty, and the Government Poverty Reduction Strategy Paper (PRSP) in 2004 notes that poverty remains high because unemployment remains high, regional disparities exist and emigration has declined. Thus, the Government sees the improvement of the productive capacity of the poor as a major element of its strategy. The objectives related to capacity building at the central and decentralized levels are also still relevant, as suggested by the inclusion of capacity building elements in the PRSC (e.g., improving human resource capacity in the public administration is a medium-term objective identified in 2004). An alternative to the traditional labor-intensive public works programs (FAIMO) was successfully established by creating a delegated contract management agency (development objective 1). The main instrument for poverty reduction envisaged by the project was "the creation of institutional capacity to execute works that are (a) economically and/or socially useful; (b) executed cost-effectively by the private sector; and (c) based on high labor intensive techniques". Interviews carried out by the ICR team suggest - 5 - that AGECABO is regarded by all stakeholders as an example of efficiency and effectiveness. Indeed, the performance of AGECABO appears to have been an important element in the decision of the United States Government to sign an agreement for US$180 million under the Millennium Challenge Compact as it indicated that Government systems were not solid enough to manage money directly. In particular: l Public works managed by AGECABO were economically and/or socially useful : As mentioned above, infrastructures financed by the project significantly improved access to important services, and the Final Evaluation confirms beneficiary satisfaction: 82 percent of beneficiaries interviewed reported that the infrastructures reflected the priorities expressed by the communities, and 79 percent found them an important contribution in the fight against poverty. l Public works managed by AGECABO were executed cost-effectively by the private sector : Comparing the cost-effectiveness of public works across programs is problematic because of the different types of public works executed, so the analysis carried out by the ICR team is limited to primary schools and rural roads. In the first case, the average cost of schools built by AGECABO was 24 percent lower than that of schools built by the Ministry of Infrastructure and Transport (MIT); as for roads, the cost per linear meter when AGECABO managed the works was two-third the average cost incurred when the MIT was in charge (see Annex 3 for more details). Technical audits conducted in 2002 and 2004, as well as Bank supervision missions and visits conducted by the ICR team, leave no doubt as to the good quality of infrastructure built under AGECABO's management, thus excluding the possibility that lower costs were achieved by sacrificing quality. Cost-effectiveness was also attained by AGECABO as an agency: (a) the agency's operating costs have remained close to the 10 percent set as the KPI, and financial management, procurement and sub-project monitoring have been good throughout project implementation (see PSR/ISR and aide-memoires), and (b) works were generally carried out within the agreed deadlines and in any case much faster than when the government is directly in charge (e.g., a sports center was built by the municipality of Mosteiros in about four years, while AGECABO built a similar one for the same municipality in less than a year). l Public works were based on labor-intensive techniques: All works were executed using technologies that were the most labor-intensive possible for the type of infrastructure and their related technical quality norms. Municipal FAIMO programs, which are arguably the public works programs most comparable to AGECABO, had an average labor ratio of 20 percent, compared to AGECABO's ratio of 34 percent. Thus, it would appear that municipalities wanting to maximize employment creation would be better off adopting AGECABO's approach or working through it. Municipalities, communities and NGOs are now better able to plan, implement and monitor sub-projects (development objective 2). Capacity building at the local level was planned to complement the execution of civil works as a way to support the decentralized approach promoted by the project and also to further sustainability. While initially attention was concentrated on infrastructure sub-projects at the detriment of the capacity building component, the results by project end are fully satisfactory (see Annex 1 for more details, as well as section 4.2 below). All 17 municipalities received training in project planning and management, and training evaluations consistently reveal high levels of satisfaction. Participants included members of the Municipal Partners Commissions set up by the project. These commissions, which comprise representatives of municipal governments, decentralized government services, NGOs and community associations, are the first example in Cape Verde of local development actors getting together to decide on investments and receive the means to implement their decisions. In line with the projected KPI, all 17 Commissions are still functional. Their members, interviewed by the ICR team and by the Final Evaluation team, speak with pride of the results achieved together. In addition, two Commissions have registered themselves officially as associations to ensure their long-term sustainability, - 6 - and several Commissions have taken on responsibilities going well beyond the project scope, such as running AIDS prevention programs or setting up micro-finance services. All agreed that if it weren't for the support received through the Program Coordinating Unit (PCU), their achievements would not have been possible. Capacity building was also provided to both Municipal Partners Commissions and small contractors on infrastructure management and maintenance. The good conditions of infrastructures found by the Final Evaluation and the ICR team suggest that skills were indeed transferred to the local level. On the other hand, the Final Evaluation provides a somewhat mixed picture with regards to the capacity building provided to communities: while 84 percent of beneficiaries who expressed an opinion (45 percent of the total had none) felt positively about the sustainability of the sub-projects, less than 10 percent considered that the community had been trained sufficiently to be able to manage the infrastructure autonomously. There exists a central capacity to coordinate, monitor and evaluate the National Poverty Alleviation Program (development objective 3). A PCU was established by Government resolution in 1998 to coordinate the NPAP. The PCU has done a superior job in integrating the three donor projects (IDA, IFAD, AfDB) that provide funding under parallel financing for the NPAP into a coherent program. Examples of economies of scale and synergy abound. Many running costs of the PCU are shared by the three projects (e.g., information and dissemination, monitoring and evaluation). Infrastructure built with IDA funding is equipped with AfDB funding and also used by IFAD beneficiaries, and the technical assistance for its efficient use is provided by all the three projects, each focusing on different aspects. A good database maintained by the PCU has allowed for close monitoring of the NPAP and facilitated Bank supervision, while regular information events (radio spots, media news reports, various meetings) have favored transparency and visibility. The project also showed flexibility in agreeing to finance a household survey (the IDRF), which was carried out by the National Statistics Institute (INE). The IDRF has provided a wealth of much needed, high quality data that have been instrumental in shaping the country poverty reduction strategy, through, for example, updating the Consumer Price Index (CPI) and developing a targeting system; it also allowed for more in-depth analysis of specific poverty-related topics (see also next section). Because the government has opted for a broader approach to poverty reduction as reflected in the PRSP, the NPAP has now become a part of the overall strategy, whose central coordination is the responsibility of the Ministry of Finance. It is difficult to determine the extent to which the project contributed to poverty reduction (overall development objective). The overall objective of the project was to contribute to reduce poverty among the population living below the poverty line by allocating resources to sub-projects on the basis of: (a) a maximum allocation by municipality according to a poverty-based formula; and (b) municipal targeting of poverty pockets identified during project preparation through detailed poverty assessments conducted at the municipal level. Between 1988-89 and 2001-02, the share of the population living below the poverty line went from 49 percent to 37 percent, a reduction of 1 percent per year in line with the key performance indicator (KPI) for the CAS-related objective (note that at project preparation the poverty headcount had been estimated at 30 percent, while subsequent calculations with a more sophisticated methodology revealed that a more correct estimate was 49 percent; in any case, regardless of the point of departure, a consistent methodological approach leaves no doubt that poverty went down considerably). To gauge the extent to which the project can be partly credited for this impressive progress the project team considered the following elements: l Targeting effectiveness: The table below suggests that targeting by municipality did not take place, as there is no clear pattern in allocations by municipality, or in number of projects (e.g., the highest allocation per capita went to Maio, which is one of the five least poor municipalities, while the average allocation for the four poorest municipalities, which have over half of their population below the - 7 - poverty line, was less than the national average and less than one third the allocation to Maio). On the other hand, there are reasons to believe that targeting did take place within municipalities. While the ICR team could not obtain evidence on whether municipalities had consistently concentrated subprojects in poverty pockets, the characteristics of the employment offered by the project are most likely to have resulted in self-targeting. Average monthly salaries paid to unskilled workers were about 14,000 CVE (about twice the salaries paid by FAIMO programs), well below the 28,833 CVE poverty line for extreme poverty, therefore it can be assumed that only the poor would have been interested in working for SSDP-financed sub-projects. The reverse of the coin, of course, is that wages low enough to result in self-targeting are generally too low to lift families out of poverty. However, given the temporary nature of most employment created and considering that in Cape Verde most households have diversified sources of revenues including significant remittances from abroad, the project was probably right in adopting an approach that had a strong potential for self-targeting. Overall, the employment generated by the project injected approximately US$ 4.7 million in wages for unskilled workers into the economy. Municipality % Total Per capita Population subproject subproject Number of under expenditures expenditures subprojects poverty line (`000 CVE) (CVE) Paul 56 63,647 7,634 9 S. Domingos 55 64,667 4,885 14 S. Miguel 55 62,598 3,912 11 S. Cruz 54 122,601 3,731 27 Ribeira Gr. 49 98,710 4,608 15 Porto Novo 49 84,476 4,890 22 Mosteiros 47 85,260 8,977 7 S. Felipe 45 66,067 2,367 9 S. Nicolau 43 88,484 6,448 9 Tarrafal 43 51,121 2,868 8 Brava 43 43,465 10,223 8 S. Catarina 43 134,103 2,689 31 Maio 35 110,132 16,320 6 S. Vicente 29 153,045 2,295 23 Praia 22 198,670 1,896 18 Boa Vista 21 32,258 7,608 5 Sal 20 25,358 1,712 3 Source: Project Data, Census Data 2000, Poverty Diagnostic 2004 (World Bank). l Improved quality of life for the poor: The SSDP made a significant contribution to improve access to social services through the construction of social infrastructure (see table below). Of particular importance is the extension of access to water and sanitation services, given Cape Verde's challenges in this sector especially in remote areas, and the large coverage of community multi-use centers. These centers represent a best practice model for smaller, dispersed populations, as they make optimal use of an infrastructure and become a one-stop-center for a variety of services. Services offered in multi-use centers include preschool, training rooms, community library, feeding programs for children and - 8 - elderly, health outreach, community meeting rooms for local associations, and an administrative area where local employees provide outreach for official municipal services (thus bringing government closer to remote communities and poor peri-urban areas). Municipal governments contract with local associations for the management of the services, like training or day care, and recurrent costs are shared between the municipality, community associations and user fees for services. It is also worth noting SSDP contribution to extending access to preschools, a particular important service in a country where 40 percent of households are female-headed and where female-headed households have a higher likelihood to be poor (urban households headed by a female are associated with a 6 percent loss in income per capita). According to the Carta Social, out of a total installed capacity of 24,817, the project contributed 1,122 or a net expansion of 6 percent of preschool capacity in the country. The real contribution of the project, however, is likely to be higher because preschool services offered in multi-use centers were not included in the estimates of the Carta Social. Type of Number of Out of national target % benefiting infrastructure beneficiaries population from SSDP Health infrastructure 23,302 431,989 total population 5 Education 16,205 90,543 primary school aged 18 infrastructure children (6-12) Water and sanitation 53,973 431,989 total population 13 Community and 123,780 431,989 total population 29 multi-use centers Roads 77,064 431,989 total population 18 Sports facilities 3,958 193,222 youth (7-25) 2 Markets 14,724 431,989 total population 3 Environment 3,736 431,989 total population 1 Total 316,742 Source: Project data, Carta Social (INE). l Employment creation for unskilled workers: According to the original outcome KPI, the project was to create 20,500 person/months of unskilled labor. By the end of the project, employment created was FAIMO-type workers amounted to only 12,800 person/months. What is the reason for such discrepancy and, more importantly, should this difference be considered an indication of project failure? As mentioned earlier, the original estimate was overly optimistic, and in the end the average labor intensity was 34 percent --lower than the projected 85 percent but higher than the level observed in other DCMAs in the region (e.g., Senegal's Agetip averages 25 percent). Also, the goals of the project were two-fold --to create good infrastructure that would improve the lives of the poor, and to create temporary jobs. Municipalities were free to optimize between these two following locally determined investment priorities. They gave priority to expansion of social infrastructure (which has a relatively low labor content) more than foreseen at beginning, perhaps due to availability from other sources of financing (EU, IDA) for roads (which have a higher labor content). The discrepancy between the original KPI estimate and the final estimate should therefore be considered a shortcoming in project preparation rather than a project failure. l Beneficiary assessments: The independent Final Evaluation, which is informed by two beneficiary assessments (over 2000 people were interviewed in 2003 and 2005) as well as by interviews with local and central government authorities, NGOs and local associations, reports a widespread impression that the project contributed to poverty reduction. Of the 1,126 households interviewed in 2005 in relation to - 9 - 151 subprojects, 79 percent considered the subproject as important or very important in the fight against poverty. Most useful were schools, health posts, roads, community centers and water; least useful were markets and sport fields. The opinion of municipal authorities, local associations and NGOs is more cautiously optimistic. Four out of five believe that SSDP contribution to poverty reduction was only moderate. In fact, if on the one hand the project improved the quality of life of poor communities through key infrastructure, on the other the employment created was only temporary. 4.2 Outputs by components: Component 1: Public Works and Employment Creation (moderately satisfactory) Public works financed by the project built infrastructure that improved the quality of life of over 315,000 people. The final breakdown of infrastructures built and related costs by type of project is presented below. As mentioned earlier, these infrastructures were found to be consistent with the priorities identified by communities and 82 percent of beneficiaries considered them important or very important in the fight against poverty; this compares favorably to the original output indicator (PAD, Annex 1) that aimed at a 70 percent rate of satisfaction among beneficiaries. As for the number of direct beneficiaries, the revised KPI set as a target a cumulative total of 318,000, equivalent to almost 80 percent of Cape Verde's population. Final estimates put the figure at 316,742, equivalent to about 73 percent of the 2000 Census (note that these figures do not necessarily represent the total number of direct beneficiaries, as some people may benefit from more than a project). While the final total comes slightly short of the original estimate, benefiting such a large percentage of a country's population is a very respectable achievement, and the discussion in the previous section indicates that the contribution of the project in expanding access to basic social services has been significant (the PAD designates the percentage of people with access to various social services as one of the original output KPI, but no target percentages are specified and in any case this would be an outcome indicator). Also and according to the data collected by the PCU, 80 percent of the new facilities were found to be functioning properly at closing. Total number of infrastructures financed under SSDP Type of Infrastructure Quantity Cost (million CVE) Health Facilities 21 62.38 Education Facilities (including preschool) 57 173.37 Water and Sanitation 26 36.04 Roads 46 596.68 Other Community Social Infrastructure (e.g., sports centers, 84 477.47 community centers for the handicapped) Community Economic Infrastructure (e.g., markets) 3 21.86 Total 237 1,367.80 Source: Project data, Conventions between PCU/NPAP and AGECABO. 2000-2005. Employment creation was much less than originally envisaged, and also less than envisaged in the Supplemental Credit. As mentioned earlier, the project potential for employment creation was grossly overestimated at inception, and it was revised twice during project implementation as follow: - 10 - Direct Employment Creation through Public Works: Number of person/months of unskilled labor Timeline Total Ex-FAIMO At appraisal 104,000 20,500 Revised at mid-term review 28,271 22,930 Formally revised at 16,750 13,800 Supplemental Actual at closing 15,081 12,800 (direct + indirect) (31,999) (29,698) Source: Project data, AGECABO. March 2005 At the time of the Supplemental Credit, it was estimated that the objective for the whole project, i.e., original plus supplemental, was to create a total of 38,500 person/months of (direct and indirect) employment, of which 35,000 person/months for ex-FAIMO workers (i.e., unskilled laborers). The project created nearly 32,000 person/months (equivalent to 83 percent of the revised objective), of which about 29,700 for FAIMO-type workers (85 percent of the objective --note that the project did not collect data on whether unskilled laborers hired by contractors had actually participated in FAIMO programs before). The results are slightly better as regards direct employment only (respectively, 90 percent and 93 percent of the target set in the supplemental credit; see Table 1 in Annex 1 for data by year, as well as Annex 3 for a more detailed description). Component 2: Capacity Building and Monitoring & Evaluation (Satisfactory) This component aimed at strengthening capacity both at the decentralized and the central levels, but the means to this end were not clearly defined in the PAD. Considering this lack of specificity in defining this component and the fact that there are no output KPI for this component (see Annex 1 of the PAD), the final output for the component confirms the ICR team positive impression of the dedication and competence of the project staff in identifying and implementing a relevant training program. As for monitoring and evaluation (M&E), the outputs mentioned were monthly and quarterly reporting by the PCU, yearly beneficiary assessments, and biannual impact evaluations. In reality, only two beneficiary assessments were carried out (2003 and 2005) and no project-specific impact evaluation was ever undertaken, even though analyses done for the preparation of the Carta Social provide important insights. Implementation went well beyond what was initially envisaged and produced highly satisfactory results. A slow start in the execution of this component was more than compensated in the last three years, although funding limitations did not allow the project to satisfy the demand generated by the high quality and pertinence of the training provided. The project also showed flexibility in reallocating part of its funding to cover the expenses of the IDRF, a decision that proved wise in light of the central role played by the IDRF in shaping the current poverty reduction strategy and of the extensive use made of the data collected (see below). Outputs in terms of capacity building encompassed a large range of development actors and addressed important needs. They can be summarized as follows (see Annex 1 for more quantitative information on the capacity building program financed by the project): l All 17 municipalities have an operating Municipal Partners Commission in accordance with the NPAP (see previous section for their composition). In some cases these commissions have taken over new responsibilities, such as HIV/AIDS prevention. With the support of an IFAD project also operating - 11 - under the NPAP, nine Municipal Partners Commissions have evolved from an ad-hoc commission into a legally established association (Regional Partners Commission), hence the likelihood of sustainability appears high. It is important to note that these mixed commissions have made it possible to set up successful collaborations among municipal authorities and NGOs/associations in social service delivery. For example, most multipurpose community centers have been built by municipalities and are operated by community associations or NGOs, with municipalities contributing to running costs. l After a needs assessment, all 17 Municipal Partners Commissions received training in participatory techniques, project design and management, and infrastructure management and maintenance. All training evaluations were extremely positive, both immediately after the training and later on after the opportunity of applying the skills newly acquired (Final Evaluation and ICR mission interviews). l 385 small contractors and infrastructure managers (the latter include community associations, NGOs, community workers and private individuals) were trained in infrastructure management and maintenance. All training evaluations were extremely positive and the demand generated for this type of training remains high. l Manuals and guides for infrastructure management and maintenance were produced and over 500 of them were distributed. l To facilitate employment creation, training in painting was offered to 212 unskilled laborers in collaboration with a paint factory (SITA). Three months after the training, 40 percent of the trainees have already found employment, and one has even set up his own painting business thanks to micro-credit offered by another project under the NPAP. Institutional capacity to monitor and evaluate poverty has greatly increased thanks to the project. The IDRF has received very positive evaluations from AFTPM (including a "best practice" label for the sensitization campaign to prepare the population) and has served as the basis for the PRSP. INE has now acquired the skills to competently run and analyze similar surveys in the future. Data from the IDRF have been widely exploited by INE, which has produced several studies on specific poverty-related topics (e.g., health expenditures, fiscal incidence, pensions). In addition, a number of other studies are under way; among them is the production of a Carta Social containing the location of all social infrastructures, both public and private, thus complementing the existing basic health and school maps. Poverty monitoring will be carried out by INE with a two-pronged approach: (a) a Database of Official Data and Statistics, containing all poverty-related indicators and making them accessible through a unique portal; and (b) a Database of Means and Projects, containing all information on poverty-related output (see Project Document of the PRSC, Report No. 30056-CV). In terms of data gathering, a Demographic Health Survey is about to be launched and INE is planning a CWIQ-type (Core Welfare Indicators Questionnaire) survey every two years. The NPAP Program Coordinating Unit (PCU) has been able to build on the strengths and complementarities of projects financed by four donors: IFAD, AfDB, UNDP, and IDA. The synergy created among the projects has produced economies of scale (e.g., information campaigns and monitoring and evaluation have been shared) and has resulted in a more comprehensive and sustainable approach. For example, IDA finances the infrastructures and IFAD the organizational support for running the services housed in the infrastructures. A National Commission for Poverty Alleviation (Comisão Nacional da Luta contra Pobreza, CNLP) was created in 2003 to ensure sustainable government coordination, but its operationalization is unlikely in the context of the PRSP. In fact, as the government approach to poverty alleviation has broadened, the NPAP now feeds its experience into the PRSP and overall coordination of poverty-related efforts is no longer the responsibility of the NPAP, but of the Ministry of Finances. 4.3 Net Present Value/Economic rate of return: One important objective of the NPAP supported by the Project was to develop an economically - 12 - sustainable alternative to the traditional labor-intensive public works system. This system, known as FAIMO, was set up in the 1980s to counteract the high rural unemployment resulting form periodic droughts and consists mainly of rural roads and infrastructure. FAIMO represents an important social safety net for the poor, but it has been criticized for its low efficiency and for generating a welfare mentality. By creating AGECABO, the SSDP aimed at assisting the Government in developing the institutional capacity to execute public works that can provide employment for the poor in a cost-effective manner. The model followed in setting up AGECABO was that of the public works employment agencies known as AGETIP, successfully pioneered in Senegal in 1989 and replicated throughout West Africa. The comparative analysis is based on FAIMO programs such as ACDI/VOCA (small agriculture infrastructure implemented through farmers associations), INERF (larger rural engineering works), municipalities (labor-intensive road repairs, clean-up activities); or non FAIMO programs such as PROMEF for school construction and the Road Sector Support Project (RSSP), which used the Ministry of Infrastructure and Transport (MIT) to centrally manage public works. Comparison of AGECABO's performance to other labor-intensive public works programs concludes that, overall, the delegated contract management approach used by AGECABO is cost-effective. It out-performed (in cost and in time) the municipal FAIMO program, which is the most comparable to it. In addition, comparative analysis with the other programs indicates that AGECABO was effective in (a) managing public works that are labor-intensive; (b) creating employment; and (c) managing the overall program. The table below presents a summary of the comparative analysis (details are found in Annex 3). Effectiveness and AGECABO's Comparable programs' achievements cost-effectiveness criteria Achievement Labor-intensity of public-works Equivalent to previous FAIMO Employment generation (direct+indirect 13,142 person-month (2003) programs of ACDI-VOCA and INERF employment) respectively 11,000 and 12,000 person-month (2001) Employment creation to catch-up with Employment creation fully synchronized Employment creation not synchronized in seasonal unemployment periods during with unemployment periods (Nov-June) the case of ACDI-VOCA's and INERF's the dry season FAIMO programs Cost-effectiveness of Employment creation 996,000 CVE: FAIMO program managed Cost per worker employed 518,000 CVE (2003) by municipalities Labor share in the construction costs 34% of total expenditure 20% of total expenditure in FAIM0 (including Agency cost) program managed by municipalities Cost-effectiveness of Public Works Cost-effectiveness of school Average cost of schools managed by Average cost of schools managed by the construction AGECABO: 3,110,000 CVE MIT under PROMEF: 13,000,000 CVE (AGECABO average cost is 24% lower) Cost-effectiveness of rural road Average cost of roads managed by Average cost of similar roads managed by construction AGECABO: 11,000 CVE per linear meter the MIT in the RSSP: 17,000 CVE per linear meter (AGECABO average cost is 33% lower) Cost-effectiveness of Agency Cost of the Agency as a ratio of the cost 11% in average (2000-2005); Norm in other AGETIP located in of the works it managed 8.6% in 2003 (Agency cost during full similar countries is 10% capacity period is below objective) 4.4 Financial rate of return: Not applicable - 13 - 4.5 Institutional development impact: The institutional development impact of the project is rated as substantial, because the project has demonstrably and sustainably built the capacity of local development actors (including public and private agencies) and central institutions on many fronts: established a programmatic approach to alleviating poverty; instituted private sector principles in contract management among central and decentralized governments, contractors, and service providers; strengthened national capacity to select, measure, monitor, supervise, and evaluate interventions according to their impact on poverty. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: l Loss in Credit Value. By the end of 2001, the PCU had signed four main Conventions (1/99, 2/99, 1/00 and 1/01) with AGECABO for a total value of 1.3 billion CVE, or US$11 million equivalent at time of signing. By 2003, the Government and the Bank estimated a total of almost US$1 million in exchange rate-induced losses in the escudo-denominated contracts due to the depreciation of the US dollar. The bulk of actual disbursements made on these sub-projects has occurred over 2002 and 2003, when the US dollar had lost about 25 percent of its value in relation to the CV escudo. The last Convention (01/01) between the PCU and AGECABO financed under the original credit was signed before the dollar depreciation and by that time, many of the technical studies were completed and bidding had already taken place for a number of contracts. When the dramatic depreciation in credit value occurred, the only possible response was to suspend the execution of many of the sub-projects, and to reduce the scope of on-going works where possible. This led the Government to formulate a request for supplemental financing to cover all or part of the portfolio of sub-projects already identified by the municipalities. l Unanticipated Cost Increases. Coming from the FAIMO tradition of low-pay, force account public employment schemes, there was little capacity within municipalities or NGOs to design and execute small-scale civil works using private contractors. Under the SSDP, municipalities presented their demand based on their original estimates of subproject cost. The PCU then approved a package and entered into a contract (Convention) with AGECABO to implement the subprojects, including technical designs and bidding. The first Conventions suffered from average cost increases of 60 percent, well above the standard 20 percent price contingency included in the Credit Agreement. This rate was reduced to 36 percent and 17 percent in Convention 01/00 and 01/01 as technical capacity improved thanks to their experience working with AGECABO. However, despite the substantial improvements, cost increases in the last Convention under the original credit, equivalent to US$800,000, could only be met by suspending execution of part of the Convention given the concomitant loss in credit value. Note that this factor was at least in part subject to government control, as the limited planning capacity of municipalities should have been known and pointed out to the project preparation team. 5.2 Factors generally subject to government control: l Salary Reduction. The government decision to reduce salary levels and harmonize contract terms among Project Coordination Units in July 2003 had a significant impact at the time of contract renewal of all NPAP personnel (June 1st, 2004), with the exception of the NPAP coordinator, whose contract was due for renewal October 31, 2004. The Government decided that the old salary scale be maintained while the revised contract would now require that all contractual staff be responsible for paying income taxes; in addition, by Arrêté 31/04 of July 21, 2004, the Ministry of Finances and Plan informed the PCU staff that they would be responsible for paying income tax arrears from January 2003 within a six-month period. According to estimates made at the time, this created a monthly salary loss ranging between 38 and 92 percent depending on seniority. Needless to say, the impact on - 14 - staff morale and performance was highly negative, thus putting at risk the timely completion of project implementation. However, the contract of the NPAP coordinator was not renewed. Fortunately, he was replaced by the PCU Program Officer for Institutional Development who adequately ensured the interim until shortly after project closing. l Counterpart Funds. While counterpart funds were systematically deposited under the original credit, the requirement for the supplemental credit suffered a delay of about eight months from the effectiveness date of March 2, 2004 (see section 5.4 for additional details on counterpart funds). This caused important payment arrears to suppliers and AGECABO, and the PCU had to move financing of some operating costs to other NPAP programs (IFAD, AfDB). However, the Government took the decision to cover the accountant's salary during the four-month grace period (April 1 - July 31, 2005) granted under the IDA credits to ensure proper follow-up on timely payment of accruals. 5.3 Factors generally subject to implementing agency control: While the PCU has demonstrated its management capacity over time, the tracking and the exploitation of the information on project investments remained a challenge in terms of sub-projects under implementation. AGECABO had also difficulties in consolidating and providing reliable data on the status of indicators related to employment creation. The wealth of information being generated by the contractors on each sub-project was unfortunately not sufficiently exploited by the PCU and AGECABO. Unless requested during Bank missions, the information was not systematically recorded and consolidated, and, therefore, not easily made available as expected. 5.4 Costs and financing: The original project cost estimate was US$18.1 million, of which US$16.1 from IDA and US$2.0 million from Government, communities and municipalities. An additional amount of US$4.4 million, including US$4.0 million from IDA and US$0.4 million from Government, communities and municipalities was provided under the supplemental credit for the reasons mentioned under sections 5.1 and 5.2 above. Disbursement levels at closing reached US$21.6 million for IDA and US$1.4 million for Government, communities and municipalities. Two reallocations of the original credit proceeds were carried out : (a) in 2001 for the financing of the National Household Survey; and (b) in 2003 to complete project activities, which essentially moved funds from the unallocated category and unused PPF to the subprojects category. Counterpart funds posed a number of challenges. The main issues are summarized below: l The government contribution was slightly less than anticipated (US$500,000 over US$600,000), but this did not affect the implementation of the activities against which counterpart funds were required, except for the 8-month delay in the deposit of the counterpart funds for the supplemental. As mentioned earlier, funds for operating costs for the PCU/NPAP were also provided from projects financed by the other donors involved (AfDB, IFAD), particularly during the delay in the deposit of the counterpart funds associated to the IDA supplemental credit (see section 5.2 above). l The appraisal team of the original credit indicated that Government and beneficiaries contribution would be around 10 percent of the investment cost, with municipalities (a) providing in-kind contributions such as land, and (b) contributing to recurrent and operating costs using either government transfers to the municipalities (equivalent to 6 percent of the government revenue for the previous fiscal year) or by cost recovery and locally generated revenues. However, municipal contributions were not properly monitored either by AGECABO or by the PCU, and information provided at project closing by the NPAP was based on the estimated value of land provided by municipalities for sub-projects. The land value for social infrastructure was estimated on a range going - 15 - from 1200 CVE/m2 for Praia to 400 CVE/m2 for Porto Novo; estimates for roads were based on an average rate of return of 1400 CVE/m2 (see Table 1, Annex 2). l Evidence shows that most municipalities provided land (sometimes bought from private owners) or technical studies as their contribution to the project, for a value well below the anticipated 10 percent. In light of this situation, the Government requested to increase the percentage of expenditure to be financed by IDA for Category 1 (sub-projects) from 90 to 100 percent of amount disbursed. The request was approved less than two years after effectiveness (July 2001), in part because taking into account these in-kind contributions in the withdrawal applications proved very difficult. In the end, of the US$1.75 million expected (of which US$1.4 million under the original credit), it appears that only US$0.9 million came from the municipalities. 6. Sustainability 6.1 Rationale for sustainability rating: Project sustainability is rated as likely. l Government commitment to poverty reduction continues to be high and the PRSP produced in 2004 clearly identifies the NPAP as one of the building blocks, as well as the forbearer, of the government strategy. The specific objectives of the project are also in line with present priorities (employment creation, restructuring of the FAIMO system, provision of infrastructure, decentralization), suggesting that the government will continue to sustain them. While with a new party in power (elections took place in 2003) support to the project appeared less strong (see aide-memoire of October 2004), the last months of project implementation have been smooth. The PRSC will make it easier for the government to continue pursuing the NPAP objectives within the context of budgetary support, building on the SSDP experience. l The NPAP PCU set up by the project is likely to continue operating at least in the short to medium term. Sustainability of the PCU would be assured if it were integrated into the ministry under whose oversight it operates, the Ministry of Labor and Solidarity. However, the Minister has indicated that he prefers to keep the PCU in its present semi-autonomous status, while two of his advisors closely liaise with the PCU to ensure consistency and coordination with other ministerial programs. As the first PRSP was only approved in February 2005, institutional responsibilities for its implementation are still in the process of being established, and the Minister's hesitation is understandable. In the meantime, the PCU will continue coordinating the other two projects in its portfolio, one financed by AfDB ($16 million) closing in 2007 and the other by IFAD ($18 million) closing in 2008. l Capacity to monitor poverty has been improved, and will be used to track implementation of the PRSP. The training and experience provided to INE by the project have equipped it with state of the art skills, and budgetary support through the PRSC will provide the means needed to carry out its responsibilities. See section 4 on Component 2. l The continued operation of the delegated contract management agency established by the project appears certain. A new convention was signed in October 2004 with AfDB for 139 million CVE (US$1.74 million equivalent), and another one for 100 million CVE (US$1.3 million equivalent) will be signed for 2005. In addition, the Ministry of Education will also enter into a convention (90 million CVE or $1.125 million equivalent) with AGECABO to manage its 2005 program. In light of the construction needs identified in the education and health sectors, implementation of the PRSC in 2006 and 2007 would certainly have to rely on AGECABO involvement, as there are simply no other - 16 - institutions able to deliver with its speed and quality. Discussions surrounding the Millennium Challenge Account (MCA) financed by the United States may lead to the decision to use AGECABO's services in managing contracts as is the case for the Ministry of Education and the AfDB programs. Finally, the Country Procurement Assessment Report (CPAR) and action plan of June 30, 2004, supports the adoption of a law to allow for delegated contract management. l The experience of AGECABO has definitely, and probably irreversibly, changed for the better the way in which municipal actors approach public works, and especially the way in which they provide assistance through labor-intensive public works. Municipal authorities have not always been keen on using AGECABO's services because this deprives them of control over the distribution of contracts and employment, and also because some municipality representatives have indicated that their own personnel could do the job just as well (ICR team interviews and final evaluation workshop, March 2005). But AGECABO's demonstration effect has been powerful and the impact will be long lasting (in the words of the Minister of Economy, AGECABO has been a "quiet revolution"). Interviews with municipal authorities make it clear that the main principles followed by AGECABO have been internalized with contractual arrangements clearly spelling out responsibilities and deadlines, attention to productivity, reliance on small private contractors, etc. Some municipalities have started concluding contracts with small local entrepreneurs following these principles, avoiding AGECABO but building on its experience. Most importantly, all stakeholders are now convinced that productivity matters also in the case of labor-intensive public works offered as a safety net. In this sense, the old FAIMO system is probably dead, and the final blow was dealt by the project. l The project contributed to beneficiary empowerment and social capital development, both of which tend to have long-lasting benefits. The SSDP offered local development actors their first opportunity to analyze their needs and select infrastructures that helped addressing them. Interviews with both Municipal Partners Commissions and beneficiaries confirmed their overall satisfaction with such an approach, even though some civil society organizations complained that they were not sufficiently involved (see Final Evaluation). Voluntary contributions in the form of planning meetings and supervision have been frequent, and the assistance provided by the IFAD-financed project in building social capital at the community level has further strengthened social accountability and commitment around the sub-projects. The beneficiary assessment carried out in 2003 indicates that beneficiaries felt empowered by this approach. Empowerment was measured by: (a) a perception of improved decision-making mechanisms and legitimate representation in the community; and (b) a perception of improved power distribution within and between families, groups, communities and institutions. Results varied considerably among islands. Also, younger and better educated respondents were more positive than the average. and social capital was strengthened (the majority of respondents reported improved interpersonal relations and improved relations between communities). l Operation and maintenance have been satisfactory. While at the beginning of the project, municipal authorities waited until construction was completed to start making arrangements for its use, perhaps wary of previous experiences with much delayed executions, at project closing over 80 percent of infrastructures were found to be operational within 4 to 6 months of works completion, which is the minimum timeframe given to consider whether the infrastructure has become functional after completion. Facilities appear to be well maintained, well run and well attended. Arrangements for covering recurrent costs reflect flexibility and ingenuity on the part of municipal authorities in an effort to contain their contribution. For example, the municipality of San Felipe started assuming almost all recurrent costs for a kindergarten, asking parents for only a symbolic contribution; it has now made an arrangement with the Capucin Fathers to take over the management of the kindergarten, including all recurrent costs, knowing that parents will not mind increasing their contribution to a facility run by a - 17 - religious order with a reputation for good pedagogy, while they would expect the municipality to provide the services for free. 6.2 Transition arrangement to regular operations: A summary of the understandings between the Bank and the Government for the operationalization of the Project outcomes for the coming years is presented below: l The PCU/NPAP will continue to have a key role in coordinating, monitoring and evaluating the NPAP, which continues to receive financial support from other donors (AfDB and IFAD); l Government has opted for a broader approach to poverty reduction as reflected in the PRSP. As such the NPAP has now become a part of the overall strategy whose central coordination is the responsibility of the Ministry of Finances; l Two technical advisors from the Ministry of Labor and Solidarity liaise with the PCU on monitoring the poverty alleviation agenda under the PRSP and PRSCs; l The PRSC (US$15.0 million credit), approved in February 2005, builds on the project experience, which focused on involving municipalities in decision making through a participatory process while strengthening their capacity to manage and supervise contracts ; l The INE, which has received substantial support from the project, has now the capacity to take the lead on poverty monitoring and evaluation responsibilities in the context of the PRSP. 7. Bank and Borrower Performance Bank 7.1 Lending: The Bank's performance during the design phase was moderately unsatisfactory. As explained in Section 3.5, several shortcomings were identified by the ICR team on the project design and preparation. Those include the gross overestimation of the original KPIs for employment, which were reviewed twice during project implementation (informally during the mid-term review, and formally at the time of the supplemental), the lack of an adequate monitoring and evaluation system, the failure to take into account the weak capacity of municipalities to estimate sub-projects cost, and the poor quality of the economic analysis presented in the project document. However, the Bank's recommendation on simplifying the project scope early on through (a) the cancellation of: the micro-credit component (at project concept document, PCD) in view of other donors involvement; and (b) the integration of the community activities window (social funds) with the municipalities sub-projects into one single category "subproject" (at negotiations), was sound. It demonstrated an ability to adjust its strategy to the government and other stakeholders vision and involvement in the sector. The Bank highlighted its legitimate concerns over the Government's degree of ownership with the establishment of the AGETIP approach and the need for reforming FAIMO (at PCD stage). Also of concern for the Bank, was other donors' potential resentment in having the Bank taking a lead role on supporting the NPAP. Hence, the Bank ensured continued dialogue with the various partners and, whenever possible, would invite other donors to participate in key missions. Furthermore, the task team leader, located in Dakar, carried out several formal and informal missions to Cape Verde to follow closely progress. As a result, the project processing schedule was essentially kept on track (PAD, Annex 7). Recommendations emanating from the various reviews were adequately addressed by the team, which demonstrated flexibility on responding to Government's reluctance to solely use AGECABO as the DCMA for the public works and training programs. - 18 - 7.2 Supervision: The Bank's performance is rated moderately satisfactory. On the negative side, the shortcomings as identified by the ICR team could have been addressed by the supervision team during the early stages of project implementation. Difficulties encountered in capturing reliable information on the KPIs, not formally revising the said KPIs, not establishing a simple M&E system, and not providing municipalities with the necessary tools to better estimate sub-project costs (i.e., training, standard price lists, etc.) contribute to the decision by the ICR team to assign the rating. On the positive side, the quality of the Project Status Reports (PSR) and aide-memoires was very good, as it focused on results and outcomes, the early detection of critical bottlenecks to implementation, the quality of advice provided, the effort made by the TTLs and the team to have their recommendations implemented, and the positive feedback the ICR team received from the client on the quality of the support they received. The ICR team reviewed all PSR and one Implementation Status for Results (ISR) and related aide-memories, where applicable, and found in all of them a good level of detail, a clear identification of the problems in project implementation, a clear and strategic approach to address such problems, and substantive sector and country management involvement. The Bank supervision also proved to be flexible in deciding for alternative solutions, and to respond to emerging client needs (i.e., supplemental financing). The Bank's role was essential in reaching an agreement on specific operating principles, particularly important in a programmatic environment such as the NPAP. This created synergy among the various key donors to the NPAP. For instance, multipurpose community centers financed under the project were also used by groups supported under the IFAD's financing. Infrastructures would also be equipped by the AfDB's funding. Issues of compliance with legal covenants were systematically assessed by the supervision team. 7.3 Overall Bank performance: It should also be pointed out that task team leadership was assigned to staff who had already been involved as core team member for the operation. Hence, assuring overlapped in task management allowed to minimize disruption. Overall Bank performance was moderately satisfactory for the reasons outlined above. Borrower 7.4 Preparation: The Borrower's performance during preparation was satisfactory. The Government was highly engaged in pursuing its strategy through the implementation of the NPAP. It did not shy away from expressing its concerns on the AGETIP approach suggested by the Bank, which was driven, in part by political pressure of key stakeholders and by the lack of experience in using this approach in the country. The Government and the Bank did, however, reach a compromise on adapting the approach to allow for other national entities to participate in a competitive process aimed at selecting DCMAs in accordance with the Bank's guidelines on procurement. A better assessment of municipality's technical capacity to estimate the cost of sub-projects (by types) as part of project preparation could have helped minimize cost under-estimation found in the first conventions. This accompanied with appropriate training and the provision of a standard list of unit cost could have prevented the cost increases ranging from 60 to 36 and 17 percent (see section 5.1 above). 7.5 Government implementation performance: The Government's performance during implementation was satisfactory despite problems during the last year of the project (May 2004), which were satisfactorily resolved by December 2004 (see section 5.2 on salary reduction). The Ministry of Labor and Solidarity could have played a more prominent role in managing the NPAP, particularly in view of the ministry's increased responsibility with regard to the - 19 - PRSP/PRSC agenda. The Ministry did however, assign two technical advisors to closely follow the implementation of the NPAP and liaise with its PCU. 7.6 Implementing Agency: The two agencies, PCU and AGECABO, entrusted with implementation, carried out in a satisfactory manner their respective mandate as the original and supplemental credits were essentially disbursed and or committed by the closing date of March 31, 2005 with less than SDR 50,000 remaining at the end of the four-month grace period. The training program more than achieved its target thanks to the dedication and commitment of the PCU institutional development program officer. The professionnally-led team focus its energy on the program aspects of project implementation instead of the administrative details. However, the ICR team felt that the wealth of data collected by AGECABO could have been better exploited rather than being limited to responding to the Bank's mission demands in its raw form. 7.7 Overall Borrower performance: The overall Borrower performance is rated satisfactory. 8. Lessons Learned l Even if the main output of a project is infrastructure, "softer" aspects such as social capital development and capacity building should receive careful attention, especially with a decentralized approach. The SSDP included a capacity building component, but the component was only vaguely defined. Thanks to the dedication and competence of project staff, the SSDP was able to make up for this shortcoming. A sensible capacity building program was put together and, more importantly, the project benefited from the design of the two other projects financed by AfDB and IFAD in the NPAP, both of which included attention to "softer" issues. But both the beneficiary assessments and the Final Evaluation pointed out that failure to address "social mobilization" was a weakness in the project. l KPIs should be carefully determined at project start and limited to those most relevant to the operation. Difficulties with the KPIs relative to employment creation underline the importance of chosing KPI carefully, making explicit the assumptions behind the choice of particular values. Should the KPI reveal inadequate, they should be changed formally, rather than simply discussed in aide-memoires, as even when there is an understanding between bank and Borrower on the need to use different KPIs, failure to make formal changes will inevitably reflect poorly in ICR. Also, there were 13 KPIs under the original credit (see Annex 1 of the PAD), and their positioning in the logframe did not always seem appropriate, suggesting that the logframe had been developed as an afterthought once the KPI had already been decided (e.g., why would overall employment creation be an output indicator, while employment creation for unskilled laborers was an outcome indicator?). l Monitoring and evaluation tools should be simple and user friendly. The wealth of data collected by AGECABO from the contractors for each subproject was not systematically recorded and consolidated, thus becoming very difficult to exploit. The system in place was based on series of spreadsheets linked to each other, which required lengthy verification for potential errors in data entry. This underscores the importance of thinking about a thorough M&E system during preparation, starting from an assessment of the information needs of different stakeholders (central government, municipalities, PIU, Bank) and ending with the production of reports. l Use of National Competitive Bidding Procurement Method instead of Small Works Procurement Method. With a country made of over ten islands, it is difficult to envisage requirements for planned - 20 - bulk procurement. There should have been flexibility in setting the aggregate threshold for small works. Municipalities felt that by awarding contracts to more qualified firms of the Islands of Santiago and São Vicente, it prevented local contractors to gain access to the economy and strengthen their capacity while creating local employment. As such, a review by AGECABO of local capacities could have provided the necessary justification to allow for an increase in the threshold level per contract for Small Works Procurement Method. 9. Partner Comments (a) Borrower/implementing agency: The Borrower prepared an end-of-project report (Annex 8), in which the Borrower's own assessment of project performance and constraints is presented. The Borrower has also confirmed the content of the final draft Implementation Completion Report by its letter of September 27, 2005. (b) Cofinanciers: Not applicable, but in view of the importance of their contribution to the overall NPAP, this report was shared with representatives of IFAD, UNDP, and AfDB (partners under parallel financing). (c) Other partners (NGOs/private sector): Not applicable 10. Additional Information Not applicable. - 21 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate 1. Total number of person/month 16,750 15,081 employment created (direct only) 2. Total number of person/month of 13,800 12,800 ex-FAIMO workers employed (direct only) 3. Number of operating Municipal Partners 17 17 Commissions Output Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate 1. Number of direct beneficiaries of 318,000 (cumulative) 316,742 completed sub-projects. 2. % of new facilities fully functioning. 90% 80% 3. % allocated to AGECABO for operating 9% 11% costs. 4. Number of members of local facility 100 568 management committees trained in operations and maintenance 1End of project Project Key Performance Indicators were formally revised through the supplemental process approved by the Board on December 16, 2003. The table below presents the result of the weighted rating of the performance indicators according to the rating methodology found in the OPCS/OED Joint Guidelines for Use in ICRs and in OED Assessments (January 2005). Weighted Rating of Performance Indicators* Original Target Revised Target Indicators at Rating against Rating against Indicators Indicators Project Original Target Revised Weighted Performance Indicators (May 1999) (Oct. 2003) Completion and Weighted Target and Average (March 2005) Value Weighted Value Outcome Indicators 1. Total number of HU=1 MS=4 direct person/month 104,000 16,750 15,081 (.77) (.92) 1.69 employment created 2. Total number of U=2 MS=4 person/month of 20,500 13,800 12,800 (1.54) (.92) 2.46 ex-FAIMO workers employed 3. Number of operating S=5 S=5 Municipal Partners 17 17 17 (3.85) (1.15) 5.00 Commissions Output Indicators 4. Number of direct HS=6 MS=4 beneficiaries of 225,000 318,000 316,742 (4.62) (.92) 5.54 completed sub-projects 5. Percentage of new S=5 MS=4 facilities fully functioning 80% 90% 80% (3.85) (.92) 4.77 6. Percentage allocated <10% 9% 11% S=5 MS=4 4.77 to AGECABO for (3.85) (.92) operating costs Overall Project 4.04 Weighted Rating: Moderately Satisfactory *Weight assigned according to disbursed credit amounts at revision: original targets 77%, revised targets 23%. Ratings as follows: Highly Satisfactory (HS)=6, Satisfactory (S)=5, Moderately Satisfactory (MS)=4, Moderately Unsatisfactory (MU)=3, - 22 - Unsatisfactory (U)=2, Highly Unsatisfactory (HU)=1. Table 1: Essential data on AGECABO performance during 2000-2005 Total from 2000 2,000 2001 2002 2003 2004 2005 to 2005 (a) (a) (c) (d) 1 AGECABO total expenditures(all contracts + Agency's cost) 69,548,304 146,966,990 455,899,992 507,265,122 282,900,252 67,743,240 1,530,323,900 2 Total Contracts managed by AGECABO (b) ECV 69,548,304 146,966,990 455,899,992 507,265,122 282,900,252 67,743,240 1,530,323,900 3 Total Contracts for Works managed by AGECABO (b) ECV 60,765,570 132,890,424 423,383,625 469,561,432 260,578,805 62,398,151 1,409,578,007 4 Total Contracts for Site Supervision ECV 8,782,734 14,076,566 32,516,367 37,703,690 22,321,447 5,345,089 120,745,893 5 Ratio cost of supervision / cost of works % 14% 11% 8% 8% 9% 9% 9% 7 Total labor cost (direct+indirect employment) (b) ECV 21,267,950 55,171,308 152,373,489 172,784,901 96,307,504 23,061,776 520,966,929 8 Ratio total labor costs / construction costs % 35% 42% 36% 37% 37% 37% 37% 9 Ratio labor costs / Total AGECABO costs % 31% 38% 33% 34% 34% 34% 34% 10 Labor cost of direct employment (b) ECV 14,480,050 33,264,310 103,996,273 92,377,706 58,542,245 14,018,515 316,679,099 11 Labor cost of indirect employment (b) ECV 6,787,900 21,906,998 48,377,216 80,407,195 37,765,259 9,043,262 204,287,830 - - 12 Cost of low-skilled labor (ex-FAIMO) (b) ECV 15,191,393 48,805,508 140,497,276 156,367,974 86,538,687 20,722,537 468,123,375 13 Number of contracts (d) U 8 56 66 26 37 44 237 14 Average cost per contract ECV 7,595,696 2,373,043 6,414,903 18,060,055 7,042,670 1,418,140 5,947,587 15 US$ 94,946 29,663 80,186 225,751 88,033 17,727 74,345 16 Number of pers-months per contract p-m 175 57 145 505 85 36 135 17 Total employement (direct + indirect) (b) p-m 1,400 3,182 9,577 13,142 3,130 1,568 31,999 18 - direct employment (b) p-m 990 1,699 5,831 4,347 1,475 739 15,081 19 - indirect employment (b) p-m 410 1,483 3,746 8,794 1,655 829 16,917 20 Total low skilled labor (direct+indirect) ex-FAIMO (b) p-m 1,190 2,963 8,670 12,543 2,875 1,457 29,698 21 - direct employment (b) p-m 780 1,480 4,924 3,749 1,239 628 12,800 22 - indirect employment (b) p-m 410 1,483 3,746 8,794 1,636 829 16,898 23 % low-skilled labor / total employment % 85% 93% 91% 95% 24 Average daily wages for workers (all categories) ECV 608 694 636 526 1,231 588 651 25 US$ 7.6 8.7 8.0 6.6 15.4 7.4 8.1 26 Average wages for ex-FAIMO workers ECV 511 659 648 499 1,204 569 631 27 US$ 6.4 8.2 8.1 6.2 15.1 7.1 7.9 28 Total employement (direct+indirect) p-y 117 265 798 1,095 261 131 2,667 29 Cost of generating one job during one year ECV 596,128 554,244 571,244 463,185 1,084,602 518,522 573,894 30 Cost of the Agency 40,360,638 50,386,000 43,843,539 33,181,866 167,772,043 31 Ratio Agency cost / total cost of contracts 27.5% 11.1% 8.6% 11.7% 11.0% Source: AGECABO, May 2005 - 23 - Table 2: Capacity Building Program Content Type of No. of Geographic Level of satisfaction Impact Comments beneficiaries beneficiaries coverage Presentation and rules of the Municipal 75 National All municipalities have SSDP hired consultant to game for SSDP/NPAP authorities, NGOs, participated in SSDP help municipalities Local Associations, identifying sub-projects Technicians from Municipalities Presentation and rules of the NGOs, Local 88 National High Knowledge of This training helped game for SSDP/NPAP and Associations, procedures and rules for expedite the process ­ from Project Management Technicians from submitting projects for selection to financing of Municipalities financing under activities SSDP/NPAP Procedures, administrative, Small local 261 National Very satisfied Improved their capacity financial and accounting entrepreneurs to work, present management documents according to procedures and rules, and better management of own business Use of local building materials Contractors' 81 Porto Novo, Santa High Trained workers could Provided by Consultant workers Cruz e Mosteiros apply knowledge acquired both in SSDP works as well as other projects under the NPAP Management and maintenance Associations, 385 (201 men, 17 municipalities in High satisfaction, Provided by the NGO of infrastructure NGOs, Personnel of 184 women) 18 sessions demands exceeded MORABI (who also Municipalities in (Municipality of Praia training capacity produced the manual). charge of had 2 sessions due to Responded to identified Infrastructure high demand) need maintenance, Infrastructure managers. Painting and micro-business Youth searching for 212 8 Municipalities in 8 Very satisfied Within 3 months about Training delivered and skills to make people more first employment sessions 40% found employment cofinanced by main paint employable and 1 set-up a manufacturer (SITA); micro-enterprise with synergy with IFAD project micro-credit (hired trainees) and AfDB (micro-credit program). Used market niche to complement technical training provided by IFAD Participative techniques, Municipal 183 (100 men 17 Municipalities in Very Satisfied Dynamisation of the Project Management and Committee of and 83 women) 6 sessions Municipal Committee of Infrastructure Maintenance Partners, NGOs, Partners; its operation Local Associations goes beyond the SSDP projects Capacity Building on analysis Technicians of the 37 5 training sessions High Better knowledge of of poverty , data from IDRF, National Institute of with different poverty data and its use of SPSS software, Statistics, Personnel duration analysis. econometrics and modeling, of the PCU, simulation of poverty data, etc Directorate General of Planning, Ministry of Justice, MORABI, Central Bank. - 24 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appraisal Component US$ million US$ million 1. Public Works and Employment Creation 19.42 20.87 107 2. Social Sector Capacity Building 1.81 2.11 117 Total Baseline Cost 21.23 22.98 Physical Contingencies 0.90 Price Contingencies 0.40 Total Project Costs 22.53 22.98 Total Financing Required 22.53 22.98 Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.41 15.11 1.48 0.35 17.35 (0.41) (13.91) (1.35) (0.00) (15.67) 2. Goods 0.00 0.00 0.02 0.00 0.02 (0.00) (0.00) (0.01) (0.00) (0.01) 3. Services 0.00 0.00 2.94 0.00 2.94 (0.00) (0.00) (2.89) (0.00) (2.89) 4. Training 0.00 0.00 0.78 0.00 0.78 (0.00) (0.00) (0.70) (0.00) (0.70) 5. Operating Costs 0.00 0.00 0.26 0.30 0.56 (0.00) (0.00) (0.06) (0.00) (0.06) 6. PPF Refinancing 0.00 0.57 0.31 0.00 0.88 (0.00) (0.56) (0.20) (0.00) (0.76) Total 0.41 15.68 5.79 0.65 22.53 (0.41) (14.47) (5.21) (0.00) (20.09) - 25 - Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 17.37 0.00 0.00 17.37 (0.00) (16.49) (0.00) (0.00) (16.49) 2. Goods 0.00 0.00 0.21 0.00 0.21 (0.00) (0.00) (0.19) (0.00) (0.19) 3. Services 0.00 0.00 4.08 0.00 4.08 (0.00) (0.00) (3.67) (0.00) (3.67) 4. Training 0.00 0.00 0.54 0.00 0.54 (0.00) (0.00) (0.49) (0.00) (0.49) 5. Operating Costs 0.00 0.00 0.36 0.00 0.36 (0.00) (0.00) (0.32) (0.00) (0.32) 6. PPF Refinancing 0.00 0.00 0.42 0.00 0.42 (0.00) (0.00) (0.42) (0.00) (0.42) Total 0.00 17.37 5.61 0.00 22.98 (0.00) (16.49) (5.09) (0.00) (21.58) 1/Figures in parenthesis are the amounts to be financed by the IDA Credit. All costs include contingencies. 2/Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Project Financing by Component (in US$ million equivalent) Percentage of Appraisal Component Appraisal Estimate Actual/Latest Estimate IDA Govt. CoF. Bank Govt. CoF. IDA Govt. CoF. 1. Public Works and 18.60 2.00 19.67 1.20 105.8 60.0 Employment Creation 2. Social Sector Capacity 1.50 0.43 1.91 0.20 127.3 46.5 Building Govt. = Government, municipalities and community contribution. CoF. = Cofinancing is not applicable for this project. Funding from other donors to the Government's NPAP (AfDB, IFAD, UNDP, etc.) are under parallel financing. - 26 - Table 1: Municipalities Contribution Municipalities contribution per sub-project based on Land Municipalities Estimated Value (CVE) m2 Total cost/sub- Sub-projects Value Cost/m2 project PRAIA 198,669,896 5,863,020 Const.Centro Social dos Deficientes Visuais e Mot. 32,280,337 380 1200 456,000 Const. Centro Social Multi-uso-Achada Grande 14,529,558 240 1200 288,000 Const. 1espaço infantil de lazer-Alto Safende 4,912,862 170 1200 204,000 Const. 1 centro multi-uso-João Varela 13,080,143 220 1200 264,000 Const. 1 centro multi-uso-Achada S.Felipe 17,939,455 220 1200 264,000 Const. 1 pequeno mercado- Achada Grande Freb«nte 9,354,476 300 1200 360,000 Const. 1 centro social Armindo Lála- Vila Nova 5,256,825 120 1200 144,000 Correcção Torren e Dre - Várzea 22,888,000 0 Corr Torrencial e Pr Encostas-Achad B. Tcheu 20,484,867 0 Extensão da Rede Viaria -Achada Mato 16,321,000 11657 140 1,631,980 Const 1 Passadeira Sub - S.Pedro Lat 8,816,171 1763 140 246,820 Const 1 Jardim infantil- SAF 9,629,958 170 1200 204,000 Connstrução 2 salas de aula 6,132,407 80 1200 96,000 Extensão da rede viária- Casa Lata 17,043,839 12173 140 1,704,220 S. DOMINGOS 64,667,045 3,909,260 Calcetamento da Via Portal/Vale d'Acusta 15,435,323 11025 140 1,543,500 Calcetamento da Via Rui Vaz/Pedra Galinha 8,404,830 6002 140 840,280 Const. 6 chafarizes 6,284,440 120 500 60,000 Const. 1 jardim infantil - Chaminé 5,156,534 170 500 85,000 Const. de 1 Centro Comu - Moia Moia 5,943,000 160 500 80,000 Const. de 1 Centro Com- Capela 3,972,400 120 500 60,000 Const. 1 Centro Comu- Ach Lama 2,527,518 100 500 50,000 Const 1 Centro Comuni- Rª Chiqu 5,838,000 160 500 80,000 Estrada Portal a Vale d' Acusta 2ª Fase 11,105,000 7932 140 1,110,480 SANTA CRUZ 122,600,649 8,286,020 Const. 1 Centro Comunit-Ach.Fazenda 7,130,567 160 600 96,000 Const. 1Centro Comunit-Boaventura 3,879,784 120 600 72,000 Const. 1 Centro Comunit-Funcos 5,673,131 160 600 96,000 Recuperação da Ponte Cutelinho e Arruamento 3,403,485 2430 140 340,200 Repar. Centro saúde e USB- João Teves e Renque Pu. 5,054,535 170 600 102,000 Const.1 jardim infantil e vedação escola EBI-Cancelo 2,099,261 170 600 102,000 Const.2 salas de aula - Ponta Achada Igreja 3,152,355 80 600 48,000 Conc Const 1 Placa Desport -Órgãos Peq 5,765,537 1100 600 660,000 Conc Const 1 Placa Despor-Merc órgãos 4,297,616 1100 600 660,000 Const 1 Centro Comunitário- Cancelo 9,173,067 140 600 84,000 Const 1 Comple Sanitario-Varzea IG 2,763,482 60 600 36,000 Const 1 Fontená c/ Lavandaria-Mato Ra 1,008,000 60 600 36,000 Const 1 Fontená c/ Lavand-João Gôtô 1,008,446 60 600 36,000 Const 1 Fontená c/ Lavand-Serelho 765,000 60 600 36,000 Extençãob Rede Viária - Cutelinho 6,987,282 4990 140 698,600 Extençãob Rede Viária-Ponta Belém 7,551,313 5392 140 754,880 Extençãob Rede Viária - Cutelinho Super 14,518,542 10370 140 1,451,800 - 27 - Construction 5 complexos sanitários 12,710,638 300 600 180,000 Conclusão da Const de 1 placa Desp - Ponta Ahada 4,291,329 1100 600 660,000 Extensão da Rede Viária - Cutelinho Superior 6,987,281 4990 140 698,600 Extensão da Rede Viária -Cancelo 14,380,000 10271 140 1,437,940 SANTA CATARINA 134,103,035 6,562,200 Const. de CdB e Cist. em Esc ach.leitão, EP, RI 3,615,079 116 600 69,600 Const. de CdB e Cist. em Esc ach.Tossa, Fundura 2,881,661 116 600 69,600 Const. de CdB e Cist. em Esc João B, Librão e Telhal 6,705,271 116 600 69,600 Const. de CdB e Cist. em Esc Ach Galego, EP Reda, PR 5,584,403 116 600 69,600 Const. 2 escolas em Mato Sanches/Cruz Grande 15,766,218 280 600 168,000 Const. 2 centros comunitarios entre Picos Reda/Librão 21,683,601 300 600 180,000 Construção de 1 Viaduto - Faveta 7,826,520 521 600 312,600 Const 1 Centro Comunit-Boaentra 6,278,915 160 600 96,000 Constr de 1 Centro Comun - Furna 7,585,588 160 600 96,000 Calcetam 2 Arrua Ach Riba-Cumbem 10,979,737 7842 600 4,705,200 Const 2 Salas Aula Chã Lag- Leitãoz 7,828,000 80 600 48,000 Construção de 1 USB -Pau Verde 5,312,000 120 600 72,000 Construção de 1 USB -Palha Carga 4,511,942 80 600 48,000 Const de 1 Sanitário Público-Acha Igreja 3,429,864 60 600 36,000 Conclusão de 1 Casa de Juventude- Ribeirão Manuel 2,127,000 120 600 72,000 Conclusão de 1 Casa de Juventu- Entre Picos de Reda 3,114,000 120 600 72,000 Construção Jardim Infantil - Cabeça Carreira 4,331,000 170 600 102,000 Construção Jardim Infantil - Pingo Chuva 4,600,000 170 600 102,000 Construção Jardim Infantil - Chão Rodrigues 4,640,235 170 600 102,000 Conclusão Centro Comunitário - Mato Gegé 5,302,000 120 600 72,000 S. Miguel 62,598,246 830,000 Construção de 1 Centro Dia para Idosos 5,651,225 140 500 70,000 Const. 2 jardins infantis- Aha. Monte e Palha Carga 17,241,649 340 500 170,000 Const. 1 jardim infantil - Pilão Cão 7,573,370 170 500 85,000 Const 1 J. Infantil- Gongon/Chacha 3,167,000 170 500 85,000 Const 1 Centro Social - Veneza 5,293,925 140 500 70,000 Const 1 Centro Social -Ach Monte 555,813 140 500 70,000 Construção 1 Centro Social -Pilao cao 5,430,408 140 500 70,000 Const 1 Co Sanitário-Flamengos 5,138,752 140 500 70,000 Const 1 Centro Comunitário-Palha Carga 6,964,000 140 500 70,000 Construção de 1 Centro Social em Achada Monte( Conc) 5,582,104 140 500 70,000 TARRAFAL 51,121,339 1,571,960 Const. 2 jardins infantis- Milho Branco e Chão Bom 9,812,463 340 500 170,000 Const. de 1 Centro Social-Chão Bom 10,697,741 220 500 110,000 Const. de 1 Centro Social-Covão Sanches 10,025,135 220 500 110,000 Calcet Vila tarrafal e chão Bom 10,520,000 7514 140 1,051,960 Const 1 Centro comunitário- Figuei Muita 4,474,000 140 500 70,000 Construção de 1 USB - Achada Tenda 2,796,000 60 500 30,000 Construção de 1 USB - Trás os Montes 2,796,000 60 500 30,000 S. VICENTE 153,044,644 15,507,040 Reabilitação do Centro Comunitário de S.Pedro 1,919,339 120 500 60,000 Const. de Instalações Sanitárias em Alto Doca 2,005,915 60 500 30,000 Extensão Rede Viária 27,546,610 21189 500 10,594,500 Saneamento Básico, Extensão Rede Esgostos 11,800,912 500 0 Extensão Rede Viária - Ribeira Craquinha 7,704,205 5926 140 829,640 Extensão Rede Viária - Fonte Inês 12,179,880 9368 140 1,311,520 Extensão Rede Viária - Rª.Passarão e Alto Mira Mar 10,488,736 8067 140 1,129,380 Ampliação Escola -Pedra Rolada 1,269,149 80 500 40,000 Const. 1 Placa Desp: Escola-Fonte Ines 997,000 1100 500 550,000 Reabilita 1 Cent Comunit-Rª Craquinha 2,679,000 140 500 70,000 Const1 Jardim Infantil-Calhau 7,752,715 220 500 110,000 Const 1 Posto Sanitário -Chã Alecrim 7,067,027 100 500 50,000 Const 1 Centro Acolhim-Chã Alecrim 9,052,262 120 500 60,000 Reabilitação 2 centros Comunit 3,292,670 240 500 120,000 Const. 4 reservatórios 2,120,000 64 500 32,000 Const 1 jardim Infantil 9,395,181 240 500 120,000 - 28 - Reabilitação do Bairro De Alto de S.joão- Alto S. João 35,774,044 800 500 400,000 RIBEIRA GRANDE 98,710,113 1,140,000 Const. Centro Comunit. de Pescadores-Sinagoga 4,936,000 160 500 80,000 Const. de 1 Centro de Dia para Idosos-Serrado 16,674,740 300 500 150,000 Const. de 1 Unidade Sanitária-Pinhão 5,223,371 100 500 50,000 Const. de 1 Centro Socio-Comunit.-Figueiral 6,133,748 130 500 65,000 Const. 1 centro Dia para Idosos - Ponta do Sol 9,705,658 200 500 100,000 Const. 2 jardins infancia - Pia Cima e Lugar Guene 8,479,187 340 500 170,000 Const. 1 matadouro - Ponta do Sol 3,828,192 80 500 40,000 Const 1 Centro Desenv Comun- Coculi 7,013,309 160 500 80,000 Const 1 Cent Desenv. Comu-Chã Igrja 10,369,072 180 500 90,000 Const 1 Centro Dese. Comun- Caibros 2,259,000 120 500 60,000 Const 1 Centro De Comu-Figueiras 10,189,710 180 500 90,000 Constr 1 Centro De Comu-Fontaính 5,593,055 160 500 80,000 Const 1 jardim infant-Lombo Santa 4,034,572 170 500 85,000 Reparação Cami Vicinais-Figueiras 4,270,500 PORTO NOVO 84,476,353 1,180,000 Const. 1 USB - Aldeia Norte 4,216,618 60 400 24,000 Const. 1 jardim infantil - Martiene 4,553,936 120 400 48,000 Const. 1 jardim infantil - Alto Mira 4,935,307 120 400 48,000 Const. 1 Centro dia/lar idosos- Rª das Patas 7,018,215 140 400 56,000 Const. 1 polivalente - Rª das Patas 11,733,384 1100 400 440,000 Conclusão 1 USB - Curral das Vacas 1,752,966 60 400 24,000 Conclusão 1 USB - Lajedos 1,396,163 60 400 24,000 Const. de 1 Mercado de Peixe - Vila PN 8,136,516 150 400 60,000 Muro de protec da praia do armaz e Conc de Praceta 10,164,000 160 400 64,000 Const 1 Jardim Infantil- Manuel Lopes 2,922,000 170 400 68,000 Conc 1 Centro Multiuso - Alto S. Tomé 4,292,739 150 400 60,000 Const 1 Jardim Infantil - Casa Meio 2,894,133 120 400 48,000 Const 1 Jardim Infantil - Chã Norte 2,972,000 120 400 48,000 Reabilitação 5 sentinas 4,751,693 100 400 40,000 Construção 1 jardim Infantil 5,154,721 120 400 48,000 Construção 1 USB 2,363,413 60 400 24,000 Recuperação escola Mato Estreito 5,218,549 140 400 56,000 PAÚL 63,647,155 2,462,400 Const. 1 centro juventude-Vila das Pombas 13,964,961 160 500 80,000 Const. 1 jardim infantil - Santa Isabel 5,162,281 120 500 60,000 Const. 1 sentina municipal - Pico da Cruz 3,598,242 40 500 20,000 Const. De Parque infantil-Vila das Pombas 10,127,516 300 500 150,000 Const 1 Centro Comunitário- P Moças 6,697,315 170 500 85,000 Const 1 Centro Comu - Lombinho 7,178,840 170 500 85,000 Cal Chã M Sant/ Chã João Vaz 9,818,000 5454 140 763,560 Calcet Pontinha de Janela 7,100,000 3994 140 559,160 Conc calcet de Chã Manuel Santos a Chã de João V 8,482,000 4712 140 659,680 S. FILIPE 66,067,368 5,414,740 Const 1 Ce Comun Polivalen-Lomba 17,582,983 1200 500 600,000 Const. 1 Jardim Infantil-Figueira Pavão 3,071,000 170 500 85,000 Const 1 Centro Comunitário- Roçadas 18,037,708 140 500 70,000 Reabilitação 1 Estrada- Camp. Baixo 15,243,000 7621 500 3,810,500 Const. 1 Estrada de Penetração- Achada Fora 12,132,677 6066 140 849,240 MOSTEIROS 85,259,548 5,009,880 Const 1 Polidesportivo-Mst trás 18,658,503 1200 500 600,000 Const 1 Jardim Infantil-Rª Ilhéu 4,816,000 170 500 85,000 Conclusão de 1 Estrada - Achada Grande a Relva 39,273,000 19636 140 2,749,040 Construção da Estrada Baia do Corvo - Mosteiros 22,512,045 11256 140 1,575,840 MAIO 110,131,990 10,736,200 Construção da Rede Viária Principal- Calheta 15,714,844 12652 140 1,771,280 Construção Bloco de Apoio Escola EBI - Calheta 2,381,162 60 400 24,000 Const de 2 Rese Água Potável- P. Vaz e Pr.Gonç 2,687,956 16 400 6,400 Construção Estrada P. Vaz / Alcatraz 34,997,698 24997 140 3,499,580 - 29 - Conc. Est Pedro Vaz a Alcatraz-Pedro Vaz e Alcatraz 54,350,329 38821 140 5,434,940 S. NICOLAU 88,483,927 6,635,300 Const de 1 Chafariz -Lompelado 1,032,947 20 500 10,000 Const de 2 Sentinas-V. Tarrafal 4,686,025 40 500 20,000 Conc 1 Centro Social-Caleijaão 2,357,082 120 500 60,000 Conc 1 Centro Social-Carriçal 5,611,346 170 500 85,000 Const da Estrada Caleijão a Vila Rª Brava 28,633,000 20452 140 2,863,280 Const Passad Subm -Vila R-ª Brava 2,020,189 336 140 47,040 Concl 1 Centro Social- Juncalino 9,336,540 140 500 70,000 Conclusão da Estrada de Caleijão a RªBrava-João Baptista 34,806,798 24857 140 3,479,980 BOA VISTA 32,258,382 2,687,120 Extensão da Rede Viária, Vila Sal Rei 3,900,000 3250 140 455,000 Extensão Rede Viária, Rabil, est.Baix 2,784,546 2320 140 324,800 Const 1 Centro de Infor Juvent-Sal Rei 9,566,506 100 400 40,000 Ext Rd Viária, J.Galego, FF,CT 6,095,621 5079 140 711,060 Const da Estrada Rabil a Pov Velha 9,911,709 8259 140 1,156,260 SAL 25,358,000 2,359,860 Const 1 Com Sanitário-Chã Matias 2,150,000 60 600 36,000 Extensão da Rede viaria em hortelã 21,260,000 15185 140 2,125,900 Pavim Via Acesso ao EBI Bairro IFH 1,948,000 1414 140 197,960 BRAVA 43,465,393 321,600 Reabilitação de 2 USB-Cachaço e N:Sra 3,571,520 120 600 72,000 Reabilitação de 1 USB-Furna 2,413,771 60 600 36,000 Const. 1 Cisterna Comunitária- Pau 8,036,110 16 600 9,600 Const. 1 Jardim Infantil-Mato 7,156,000 170 600 102,000 Const. 1 Jardim Infantil-Fajã d´Agua 7,258,604 170 600 102,000 Reab. caminhos carroçáveis-Laranjeiras 8,670,388 Reab. caminhos carroçáveis-Tapume a Mato 6,359,000 TOTAL 1,484,663,082 Total Municipalities 80,476,600 contribution - 30 - Annex 3. Economic Costs and Benefits One important objective of the National Poverty Alleviation Program (NPAP) that was supported by the SSDP was to develop an economically sustainable alternative to the traditional labor-intensive public works system. This system, known as FAIMO, was set up in the 1980s to counteract the high rural unemployment resulting form periodic droughts and consists mainly of rural roads and infrastructure. FAIMO represents an important social safety net for the poor, but it has been criticized for its low efficiency and for generating a welfare mentality. By creating AGECABO, the SSDP aimed at assisting the Government in developing the institutional capacity to execute public works that can provide employment for the poor in a cost-effective manner. The model followed in setting up AGECABO was that of the public works employment agencies known as AGETIP, successfully pioneered in Senegal in 1989 and replicated throughout West Africa. This Annex analyzes the performance of AGECABO and then compares it to other labor-intensive public works programs. These include FAIMO programs, such as ACDI/VOCA (small agriculture infrastructure implemented through farmers associations), INERF (larger rural engineering works) and municipal FAIMOs (labor-intensive road repairs, clean-up activities), as well as non FAIMO programs, such as the IDA-financed PROMEF (school construction) and the Ministry of Infrastructure and Transport (MIT) programs (roads). AGECABO was able to generate a significant amount of employment for the poor. The AGECABO program generated more than 32,000 person-month of direct and indirect employment during the 2000-05 period, equivalent to 2,667 person-year (indirect employment is included because private contractors tend to subcontract labor intensive components such as cut stone, cement blocs, carpentry). The vast majority of this employment (95 percent) consisted of low-skilled jobs accessible to the individuals traditionally targeted by FAIMO programs. This means that, on average, AGECABO created 6,200 person-months of work per year. However, when it was operating at its full capacity in 2003, the Agency generated 13,142 person-months (43 percent of the total). This result is in the same order of magnitude as the two main FAIMO programs, ACDI-VOCA and INERF, that generated each between 11,000 and 12,000 person-months per year in the period 2000-01. 1 The seasonality of AGECABO's employment creation was fully synchronized with unemployment periods. In Cape Verde, agricultural works are during the rainy season (June-October) and rural unemployment is high during the dry season (November-June). The AGECABO's pattern of generated employment was fully synchronized with the labor demand, as shown by the graph below for the year 2003. This compares favorably to ACDI-VOCA, whose job creation was slightly out of step with the needs, and INERF, which actually did not create jobs in the high unemployment period. In both cases, the misalignment of job creation was due to inadequate availability of funding. - 31 - Graph: Employment generated by AGECABO per month during year 2003 Emprego em Pessoas/Mês 2500 2000 1500 1000 500 0 01/01/2003 02/01/2003 03/01/2003 04/01/2003 05/01/2003 06/01/2003 07/01/2003 08/01/2003 09/01/2003 10/01/2003 11/01/2003 12/01/2003 Source: AGECABO, March 2005. Wages paid to low-skilled workers by small private contractors or communities under contracts with AGECABO were much higher than those of FAIMO programs. During 2000-05 daily wages paid by contractors to low-skilled workers averaged CVE 630 equivalent to about US$8 (see table 1 of Annex 1), in line with market prices. These wages were higher than those paid by the FAIMO programs, which ranged between CVE 270 and CVE 400 (in particular, ACDI/VOCA paid CVE 400, INERF paid CVE 310 and municipal FAIMOs paid CVE 270). While the wages paid by the FAIMO system were kept below the market rate to encourage self-targeting, they were considered by workers quite inadequate to provide a decent living and resulted in lower productivity that lead to long implementation delays. The wages paid by ACECABO's contractors, on the other hand, promoted productivity as evidenced by the ratio of timely deliveries. Cost-Effectiveness of AGECABO in Employment Generation The cost of the AGECABO program per worker employed is estimated to be half the cost of a comparable FAIMO program. Calculations including all costs (all contracts plus the Agency cost) and all employment (direct plus indirect) show that the average cost per worker employed during one year was CVE 575,000 during the 2000-05 period, and was at its lowest in 2003 at CVE 518,000 (see table). While this is substantially higher than the costs of the ACDI/VOCA and INERF programs (CVE 310,000 in 2001, the only year for which data are available), it is almost half of cost per worker of the municipal FAIMO program. The results of the comparison with ACDI/VOCA and INERF are not surprising as both 2 programs focused on soil and water conservation works and rural engineering works that are by nature more labor-intensive than AGECABO's program of socio-economic infrastructure. Also, the INERF ratio is probably significantly underestimated as the available data do not include the cost of DGASP to plan and design the projects or the cost of INERF's vehicles and equipment. Comparison with the municipal FAIMO program is far more relevant as the AGECABO program was demand-driven by municipalities (hence, municipalities were simultaneously planning their own FAIMO programs and AGECABO - managed programs) and the type of public works carried out was similar. As a whole, AGECABO was highly efficient in keeping a high labor level in public works contracts for the construction of basic socio-economic infrastructure. Labor accounted for 34 percent of total - 32 - expenditures (i.e., including the Agency cost) in the AGECABO program; as could be expected, it was much higher for road construction (19 percent of the contracts) than for buildings such as schools and health centers. ACECABO labor share is much higher than in other comparable civil works projects. Comparisons with other public works programs have to be made with caution because of the reasons explained above. Because of the more labor-intensive nature of the works executed by the ACDI/VOCA and INERF programs, their labor ratio is higher, respectively 52 percent and 37 percent. But municipal FAIMO programs, which are more similar to ACECABO, had a labor ratio of 20 percent, suggesting that municipalities would be better off in working through AGECABO if they want to maximize employment creation in their public works programs. Cost-Effectiveness of ACECABO in Executing Public Works Primary schools built under ACECABO management are cost-effective. Comparing the cost-effectiveness of public works across programs is problematic because of the different types of public works executed, so the analysis is limited to primary schools and rural roads. The average cost of schools built by small contractors in the AGECABO program (CVE 3.11 million) was about 24 percent lower than the MIT cost through centralized procurement. Because of this difference, and because ACECABO contractors delivered in a timely manner, the Ministry of Education (MoE) decided in 2001 to shift from MIT to AGECABO the management of its IDA-financed school construction program (PROMEF). In 2005, the MoE signed a new contract with AGECABO for the school construction program financed by the Netherlands. In the meantime, the MoE has learned lessons form AGECABO and, form time to time, directly manages small contracts for school construction, awarded according to the competitive method promoted by AGECABO (three quotations) with results similar to AGECABO. Rural roads built under AGECABO management as well appear cost cost-effective. The prime objective of the road projects financed by AGECABO was to improve access to social and administrative services and market opportunities for isolated community. Since there was no reference data on initial traffic nor means to estimate future traffic, it was not possible to conduct a cost-benefit analysis and a cost-effectiveness approach was used instead. The "least cost solution" was adopted as the .3 cost-effectiveness indicator for comparing alternative investments; in this case the least cost solution for surfacing roads with little or no existing traffic was cobblestone, which was always chosen by AGECABO 4 . The average cost of road construction was about CVE 11,000 per linear meter , equivalent to US$138. 5 This cost corresponds to two-third of the costs estimated by the MIT for such roads in the Road Sector Support Project. The MIT assesses positively AGECABO's contribution to the national road program as: (i) the MIT does not have the capacity to answer the important national needs, particularly for small rural roads; (ii) AGECABO is following MIT's technical standards, and (iii) the quality of works executed by contractors under contract with AGECABO is as high as that of contractors working with MIT. Compared to contract management agencies in other countries, AGECABO has achieved good results in terms of cost-efficiency. Cross country comparison can be based on school construction since design and technical characteristics are similar. AGECABO's performance can be compared to AGETIP's performance in Senegal which uses identical implementation arrangements. In the AGECABO system, schools are built at a cost that is 25 percent higher than schools built by AGETIP, but the cost of inputs is much higher in Cape Verde, due to its island status. For example, cement is 42 percent more expensive, and wood and low-skilled labor are twice as expensive. - 33 - Cost-Effectiveness of AGECABO as an Agency The Agency has shown considerable flexibility in adjusting its cost to its volume of activities. As shown in Table 1 of Annex 1, the cost of the Agency peaked in 2002 at CVE 50 million, and was down to 33 million in 2004. The management of the Agency took appropriate steps to adapt to changing volume of operations, including a 15 percent salary cut in 2003 and an adjustment of the number of construction specialists to the workload of the Agency. On average the Agency was composed of nine staff during the whole period 2000-05, with a maximum of 13 staff in 2003 when its work volume peaked. AGECABO has shown that it can manage its program in a cost-effective manner. AGECABO's cost-effectiveness in program management can be measured by the ratio of its total cost to the financial volume of the works managed by the Agency. As shown in table 1, this ratio averaged 11 percent during the 2000-05 period. However, it was 8.6 percent during 2003 when the Agency was working at its full capacity. Although the average is slightly above the 10 percent target performance indicator, the Agency's performance in 2003 shows that it succeeded in managing its program in a more cost-effective manner than targeted. Initially, operating costs were higher as typical of AGETIP-type agencies. There is no doubt that the 8.6 percent ratio could have been maintained overtime if the volume of work of the Agency would have remained at the 2003 level or expanded. Notes 1. Siri, G., Goovaerts, P. Analysis of the Costs and Benefits of Public Works Programs in Cape Verde. July 2002, p. 20. 2. Siri, G., Goovaerts, P. Analysis of the Costs and Benefits of Public Works Programs in Cape Verde. July 2002, p. 35. 3. Road Sector Support Project, Project Appraisal Document, Annex 9, p. 64. 4. It is also a higher labor intensive solution. 5. Aide memoire ICR mission. March 2005. - 34 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 01/1996 4 TASK TEAM LEADER (1); NUTRITIONIST (1), ECONOMIST - UNDP (1); POVERTY SPEC. (1) 09/1996 2 TASK TEAM LEADER (1), ECONOMIST (1) 12/1996 1 TASK TEAM LEADER 03/1997 3 TASK TEAM LEADER (1), ECONOMIST (1), IMPLEMENTATION SPEC. (1) 05/1997 4 TASK TEAM LEADER (1); SOCIAL FUND SPEC. (1); AGETIP SPEC. (1); NGO PARTICIPATION SPEC. (1) 11/1997 1 TASK TEAM LEADER (1) 01/1998 5 TASK TEAM LEADER (1); SOCIAL FUND SPEC. (1); IMPL. SPEC. (1); AGETIP SPEC. (1); NGO PART. SPEC. (1) 07/1998 2 TASK TEAM LEADER (1); ECONOMIST (1) 10/1998 2 TASK TEAM LEADER (1); ECONOMIST (1) 12/1998 2 TASK TEAM LEADER (1); ECONOMIST (1) Appraisal/Negotiation 02/1999 7 TASK TEAM LEADER (1); SOC. FUND SPEC. (1); OPER. MGT SPEC. (2); ECONOMIST (2); AGETIP SPEC. (1) Supervision 05/2000 1 TASK TEAM LEADER (1) S S 07/2000 1 TASK TEAM LEADER (1) S S 11/2000 1 FIN. MGT SPEC. (1) S S 06/2001 5 TEAM LEADER (1); S S FINANCIAL MANAGEMENT (1); PROCUREMENT (1); ENGINEER/INSTIT. DEVEL (1); ENVIRONMENT/COMM. DEVP (1) - 35 - 10/2001 5 TASK TEAM LEADER (1); S S PROCUREMENT SPECIALIST (1); FINANCIAL MANAGEMENT (1); CONSULTANT (2) 05/2002 1 SR. STATISTICIAN (1) S S 10/2002 4 TASK TEAM LEADER (1); S S FINANCIAL MANAGEMENT (1); PROCUREMENT SPECIALIST (1); ENGINEER - INSTITUTION (1) 04/2003 5 TASK TEAM LEADER (1); S S FIN. MGT SPECIALIST (1); SR MUNICIPAL ENGINEER (1); CONSULTANT (1); SR STATISTICIAN (1) 10/2003 4 TASK TEAM LEADER (1); S S INFRA. SPEC. (1); OPER. OFFICER. (1); FIN. MGT. SPEC. (1) 05/2004 4 TASK TEAM LEADER (1); S S INFRA. SPEC. (1); LEAD SP SPEC. (1); POVERTY SPEC. (1) 10/2004 4 TASK TEAM LEADER (1); U U LEAD SP SPEC. (1); PROC. ANAL. (1); INFRA. SPEC. (1) ICR 03/05 3 TASK TEAM LEADER (1); S S SR. SP SPEC. (1); SR. IMPLEMENTATION SPEC. (1) - 36 - Supplemental Credit (P086628) Stage of Project Cycle No. of Persons and Speciality Performance Rating Implementation Development Month/Year Count Speciality Progress Objective Identification/Preparation 07/2003 3 TASK TEAM LEADER (1); INFRA. SPEC. (1); FIN. MGT. SPEC. (1) Appraisal/Negotiations 10/2003 4 TASK TEAM LEADER (1); INFRA. SPEC. (1); OPER. OFF. (1); FIN. MGT. SPEC. (1) Supervision 05/2004 4 TASK TEAM LEADER (1); S S INFRA. SPEC. (1); LEAD SP SPEC. (1); POVERTY SPEC. (1) 10/2004 4 TASK TEAM LEADER (1); U U LEAD SP SPEC. (1); PROC. ANAL. (1); INFRA. SPEC. (1) (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 89.1 239.8 Appraisal/Negotiation 12.7 28.3 Supervision 92.7 271 ICR 8.0 21 Total 202.5 560.1 ICR staff inputs combine the original and supplemental credits. Supplemental Credit (P086628) Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 2.0 6.7 Appraisal/Negotiations 1.1 3.5 Supervision 6.9 22.8 ICR -- -- Total 10.0 33.0 - 37 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA - 38 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 39 - Annex 7. List of Supporting Documents Official Documents, Sources of Data, and Analytical Reports for Cape Verde Avaliação do Impacto na Perspectiva dos Beneficiários dos Projectos de Investimento no âmbito do Programa Nacional de Luta Contra a Pobreza em Cabo Verde; Relatório Final. June 2003. Cap Vert. Rapport analytique du système de passation des marchés. 30 juin 2000. Cap Vert. Rapport analytique du système de passation des marchés. 30 juin 2004. Estatutos do Instituto Nacional de Estatística, Decreto Regulamentar N. 9/2000 de 4 de Setembro, Conselho de Ministros. Étude pour la constitution d'un fonds de Développement Social au Sein du Programme National de Lutte Contre la Pauvreté au Cap Vert (Document Final), July-August, 1997. Furtado, Claudio, Relatório da Revisão a Meio-Percurso do Programa de Desenvolvimento do Sector Social (PDSS); Ministério da Saúde, Emprego e Solidariedade,Unidade de Coordenação, Programa Nacional de Luta contra a pobreza. July-August, 2001. Gabinete da Secretária de Estado para a Luta contra a Pobreza, Célula de Mobilização e Sensibilização Social, Plano de Actividades para 1999, Março 1999. Gabinete da Secretária de Estado para a Luta contra a Pobreza, Célula de Mobilização e Sensibilização Social. Programa de Actividades para o ano de 1999, Fevereiro de 1999. Governo de Cabo Verde, Cabo Verde; Estratégia de Crescimento e de Redução da Pobreza (2005/2007, Praia, November 2004. Ministério da Saúde, Emprego e Solidareidade (Observatório da Pobreza). Bolsas de Pobreza; Proposta de Seguimento e Avaliação. Agosto 2001. Ministério das Finanças e Planeamento, Documento de Estratégia de Crescimento e de Redução da Pobreza, Praia, September 2004. MORABI. Programa Nacional de Luta Contra a Pobreza (PNLP); Guia prático de gestão e manutenção de infra-estruturas, 2004. Programa Nacional de Luta contra a Pobreza (PNLP), Aplicação informática do Sistema de documentação do PLPR, Banco de dados Documentação (versão 0), Janeiro 2001. Programa Nacional de Luta contra a Pobreza (PNLP), Projet de promotion socio-économique des groupes défavorisés (PSGD); financement BAD. 2005-03-24 Programa Nacional de Luta Contra a Pobreza, Unidade de Coordenação, Relatório de Actividadtes, Quarterly Reports. Years 1999 through 2005. Programa Nacional de Luta contra a Pobreza. Inquérito Sobre as Condições de Vida das Famílias, - 40 - Convenção Nº 002/2001, Procedimentos e instrumentos para a gestão financeira da Convenção Programme National de Lutte Contre La Pauvreté (PNLP), Composante «Reconversion des FAIMO», Rapport de mission 18-21 mars 1997. Republic of Cape Verde, Micro Enterprise Training and Lending Program, Cape Verde PL-480 Monetization Program, A Project Proposal, November 1995. República de Cabo Verde, AGECABO, Audit technique(annuel). República de Cabo Verde, AGECABO, Audit technique; Phase Pilote 1999 ­ 1º Semestre et 2º Semestre 2000, October 2002. República de Cabo Verde, AGECABO, Relatório de Actividades. Years 2000-2005 República de Cabo Verde, AGECABO, Relatório Seminário de Formação de Pequenos Empreiteiros da Fase Piloto, May 2002. República de Cabo Verde, Ministério da Saúde, Emprego e Soldariedade, Unidade de Coordenação do Programa Nacional de Luta contra a Pobreza, Workshop report on Qualification Institutionnelle du PNLP , Chefia do Governo, May 3, 2002 República de Cabo Verde, Declaração da Política de Desenvolvimento do Sector, Programa Nacional de Luta contra a Pobreza. Rural Poverty Alleviation Programme (PLPR), Appraisal Report, Volume I, Main Report, Confidential, Report No. 0967-CV-July 1999. Siri, Gabriel and Goovaerts, Piet, Analysis of the Costs and Benefits of Public Works Programs in Cape Verde, Executive Summary and Full Report (with the collaboration of Antonio Sabino), Program Coordination Unit, National Poverty Alleviation Program, July 2002. Wirrankoski, Paul-Henri, AFRISTAT, Rapport de Mission, Observatoire Economique et Statistique d'Afrique Subsaharienne, December 11-18, 2000. World Bank Documents Bank's Aide mémoire (from identification to completion missions). IDA. First Poverty Reduction Support Credit. Program Document. Report No. 30056-CV. January 26, 2005. Lucas, Frank R, Draft Consultant Report, Mission 19 May ­ 21 June 2001, International Analysis of the National Poverty Alleviation Program and IDA Social Sector Development Project (PDSS). OPCS/OED. Joint Guidelines for Use in ICRs and in OED Assesssments; Rating the Outcome of Projects with Formally Revised Objectives. January 2005. Poverty in Cape Verde, A summary Assessment and a Strategy for its Alleviation, June 30, 1994. - 41 - Other Analytical Reports FIDA, Programme de lutte contre la pauvreté (PNLP), Mission intercycles opération et programme de travail pour la Phase 2, Rapport No. 1552-OV, January 2004. FIDA, Rapport de la Mission de Supervision, 14-26 mai 2001, Programme National de Lutte contra la Pauvreté Rurale, Prêt FIDA 510-CV. Miscellaneous Convenção-Quadro Reguladora das Relações entre o Governo de Cabo Verde e a Agência de Promoção do Emprego e do Desenvolvimento Local "AGECABO" no âmbito do programa de luta contra a pobreza. Etude pour la Constitution d'un fonds de Développement Social. Document Complémentaire de Synthèse ­ Expériences en cours au Cap Vert ­ Résumé de l'action de quelques «Fonds». Fonds Africain de Développement. Projet de Promotion Socio-Economique des groupes défavorises. Rapport d'évaluation. Octobre 1999. International Monetary Fund, Cape Verde: 2001 Article IV Consultation, Staff Report and Public Information Notice on the Executive Board Discussion, Country Report No. 01/174, October 2001 - 42 - Additional Annex 8. Borrower's Contribution The Borrower prepared the following end-of-project report for credits 3224-CV and 3224-1-CV. The English translation is followed by the original report in Portuguese. Republic of Cape Verde MINISTRY OF LABOR AND SOLIDARITY Office of the Minister Social Sector Development Project (SSDP) Evaluation by the Government of Cape Verde Praia, August 30, 2005 The Government of Cape Verde has made a favorable assessment of the Social Sector Development Project (SSDP) carried out with World Bank financing. The major goal of that project, which begun in 1999 and ended March 31, 2005, was a sustained 30% reduction in the poverty of the population living below the poverty line. Two priority approaches were: (1) public works and employment; and (2) skills-building, follow-up, and evaluation. From this standpoint, the following were established as the principal specific objectives of the project: (i) assisting in the restructuration of the High Labor-Intensive Public Works Program (FAIMO) by: (1) developing institutional capacity (AGECABO) [Agency for Public Works and Employment] for responsibly executing public works that are: (a) economically and socially useful; (b) effectively carried out by the public sector; and (c) based essentially on methods that make intensive use of labor. (ii) strengthening institutional capacity within the community, local government, and NGOs, a condition deemed necessary to effective and efficient implementation of actions to combat poverty; and (iii) establishing a capacity to coordinate, follow up, and evaluate the National Anti-Poverty Program at the different levels (central, local, and community). IMPLEMENTING THE SSDP The preparation phase of the SSDP began in 1997, and was financed by a grant of US$150,000 from Japan. A total of US$22.1 million was projected as needed for implementation. Fundraising was distributed among the partners as follows: (i) IDA/World Bank US$20.1 million; (ii) Government of Cape Verde US$1 million; and (iii) local governments/communities US$1.1 million. The loan agreement was signed in May 1999, and actually went into effect on September 30 of that year. The Public Works and Employment Component. Basic social infrastructures was constructed to improve access by the poor to vital equipment and services. Prominent in this regard were the construction of Basic Health Units [Unidade Sanitária de Base ­ USB], kindergartens, classrooms, sanitary facilities ( sentinas), multi-use centers, highways, roads, markets, water and sanitation, environmental protection, and multi-use sports centers, with priority given to previously-identified pockets of poverty. - 43 - The Project completed 237 units of infrastructure, corresponding to almost 100% of the programmed target. These contributed to a substantial improvement in people's living conditions. The Municipal, Community and NGO Skills-Building Component. A sizeable diversified training team was put in place to gradually build up local capacity for planning, carrying out, and managing actions aimed at reducing poverty. Furthermore, training and skills-building activities were planned. Priority was given to involving former FAIMO workers and small local contractors. We should also make special mention of the seminars, workshops, and working visits conducted in a decentralized fashion. These played an important role in consolidating the partnership relationships between the Program Coordination Unit (PCU) and the partners. Strengthening Institutional Capacity. We found that the capacity for follow-up and evaluation of the NPAP [National Poverty Alleviation Program] has been significantly improved. Gains were made in the financial and technical managerial skills of the PCU. Indeed, we were quite satisfied to see that they were able to respond efficiently and effectively to the requirements of a program of multidimensional nature and of undeniable complexity. By adopting a simple and conveniently-organized structure, the PCU demonstrated skills and capacities that will enable them to ensure proper coordination of programmed activities. Creation of Institutional Capacity ­ AGECABO The Government recognizes the significant impact resulting from the decision made to implement projects using the Delegation Contract Management approach. This translated into (1) strengthening the capacity of the central administration and of the local governments to manage and execute public works, (2) establishing highly successful alternative to the FAIMO system, (3) increasing the ability to create jobs and distributing income in the least favored communities, (5) introducing sound and disciplined practices into the market of small-scale public works for employment, and (6) contributing to poverty reduction. In accordance with the initial objective of the SSDP, establishment of AGECABO means that there is now an institutional capacity for executing projects quickly, efficiently, and transparently. The innovative practices adopted by the staff of the agency constitute a benchmark and asset to be preserved. This is why the Government will continue to use the agency's services in the realm of projects to be carried out by budgetary support such as the PRSP [Poverty Reduction Strategy Paper], in the context of the MCA [Millennium Challenge Account] or in other activities where the agency's involvement would seem to be useful. AGECABO is now working under a formal agreement with the Ministry of Education to build school infrastructures in the towns of the Island of Santiago, and under another agreement signed with the NPAP in the context of the PSGD [Project for Social and Economic Promotion of the Least Favored Groups]. One explanation for the discrepancy between the initial objective of creation of 20,000 person/month jobs and the 12,800 jobs achieved at project-end may have to do with the fact that buildings (such as community centers, schools, parks) were constructed. Although these activities are extremely important for social utility and impact on poverty, they, unlike road construction, are not really labor-intensive. This question will be examined in greater detail in the final report. Contribution to Poverty Alleviation In as much as poverty in Cape Verde is a multifaceted problem, it has been attacked on several fronts. Some of these efforts involved direct intervention by the SSDP in coordination with other components of the NPAP. It is, however, difficult to measure the direct contribution by the SSDP to the reduction of poverty in Cape Verde, since it is practically impossible to separate the net effects of the SSDP from those supported by the central government, local governments, and other agencies. - 44 - Nevertheless, one important contribution by the SSDP was the financing of the IDRF [National Household Expenditure and Income Survey]. Its results enable us to calculate and more thoroughly analyze the determinants of poverty. Poverty measured in relative terms in Cape Verde using a threshold of 43.250$, is a predominantly rural and asymmetric condition. The IDRF figures demonstrate the existence of a significant cleavage between the rural and urban environments, as well as striking differences among the islands of this Republic. Pursuing the analysis of those figures further, to reach more disaggregated geographic levels, reveals situations that are even more asymmetric, which emphasizes the need to sharpen the focus of anti-poverty policies. The documents known as the Poverty Profile and Cape Verde Poverty Diagnostic show that the determinants of poverty can be found in the absence of income and in low educational levels, among other factors. The action by the SSDP in promoting an improvement in the quality of life for the poor by giving them better access to potable water, basic sanitation, education, and health contribute directly to that objective. And so we can be certain that the SSDP made a positive contribution to the battle against poverty in Cape Verde by providing temporary jobs and income to the least favored groups and increasing the coverage of basic social services in the identified pockets of poverty. It is estimated that actions launched as part of the project have potentially benefited more than 270,000 people. Nevertheless, one must be cautious in interpreting these figures. Although the proportion of individuals lifted out of poverty by the SSDP cannot be accurately measured, the project's contribution in that regard has certainly been important, inasmuch as the program benefits the workers in the FAIMO and the rural population, primarily the least favored. Beneficiaries seem to agree that the SSDP work made a positive impact, since nearly three out of four (72.8%) believe that the proposed objective of the project was indeed achieved. Contribution by the Central and Local Governments One of the mistakes in Project follow-up was the failure to set up a system that would permit continuing quantification and evaluation of what the national contribution really would be. We know that this participation has been much more than a simple transfer of funds or the furnishing of land for buildings. For example, the Ministry of Environment, Agriculture, and Fishery have certainly played a technical and physical role in several projects, but their contributions were not recorded on the books. The same can be said of the municipalities and the beneficiaries themselves. The Government is greatly pleased that the Bank has put its faith in the abilities of this country and has financed this important development tool, one that leaves behind an extremely rich experience that can be tapped to tackle other challenges. Job creation, vocational training, literacy, increased availability of running water for household use, bathrooms for all, affordable housing, micro-credit, communication and information for development continue to be major areas where interventions are needed to reach on a larger scale the neediest target population, especially women and children, young people who lack skills, the handicapped, and the unemployed. Sidónio Fontes Lima Monteiro Minister of Labor and Solidarity (Government's report in Portuguese--original version) - 45 - Projecto de Desenvolvimento do Sector Social (PDSS) Avaliação do Governo de Cabo Verde MINISTERIO DO TRABALHO E SOLIDARIEDADE Gabinete do Ministro Praia, 30 de Agosto de 2005 O Governo de Cabo Verde faz uma avaliação positiva do Projecto de Desenvolvimento do Sector Social (PDSS), executado com o financiamento do Banco Mundial. Esse Projecto iniciado em 1999, cujo término foi a 31 de Março de 2005, elegeu como grande meta a redução sustentada da pobreza dos 30% da população residente que vive abaixo do nível da pobreza, destacando dois eixos prioritários de intervenção: 1) Trabalhos Públicos e Emprego e, 2) Capacitação, seguimento e avaliação. Nesta perspectiva, fixaram-se como principais objectivos específicos do projecto: i) o apoio a reestruturação das frentes de alta intensidade de mão-de-obra (FAIMO), através da: 1) criação de uma capacidade institucional (Agecabo) para a execução responsável dos trabalhos públicos, que sejam, simultaneamente; a) económico e socialmente úteis; b) executados de forma eficaz pelo sector privado e; c) baseados essencialmente em técnicas de utilização de mão-de-obra de forma intensiva; ii) o reforço da capacidade institucional a nível das comunidades, dos municípios e das ONG, condição julgada necessária à implementação eficaz e eficiente das intervenções de combate a pobreza, e; iii) o estabelecimento de uma capacidade de coordenação, seguimento e avaliação do Programa Nacional de Luta contra Pobreza, nos diferentes níveis (central, local e comunitário). IMPLEMENTAÇÃO DO PDSS - 46 - O PDSS teve a sua fase de preparação iniciada em 1997, com o financiamento do donativo japonês de US$ 150.000. Para a sua implementação, um total de 22, 1 Milhões USD foi projectado, e sua mobilização distribuída da seguinte forma entre os parceiros: i) IDA/Banco Mundial o montante de USD $20,1 milhões, ii) o Governo de Cabo Verde o montante de USD $1 milhão e; iii) os Municípios/comunidades o valor USD $1,1 milhões. O acordo de crédito foi assinado em Maio de 1999, tendo entrado em vigor , efectivamente, a 30 de Setembro do mesmo ano. No âmbito da componente ­ Trabalhos Públicos e Emprego, foi construído um conjunto de infra-estruturas sociais de base como forma de melhorar o acesso dos pobres aos principais equipamentos e serviços. Destacam-se, neste particular, a construção de USBs, Jardins Infantis, Salas de Aula, Sentinas, Centros Multi-Uso, Estradas, Caminhos, Mercados, Água e Saneamento, Protecção Ambiental e Polidesportivos localizados prioritariamente nas bolsas de pobreza previamente identificadas. Foram concluídas 237 infra-estruturas, correspondentes praticamente à 100% do programado que contribuem para uma melhoria substancial das condições de vida das populações. Na componente ­ Capacitação de Municípios, Comunidades e ONG's ­ foi implementado um amplo e diversificado quadro de formação por forma a criar progressivamente uma capacidade local de planificação, implementação e gestão de intervenções voltadas para a redução da pobreza. Por outro, acções de formação e capacitação foram projectadas, tendo como grupos destinatários prioritários, os ex-trabalhadores das FAIMO e os pequenos empreiteiros locais. Merece ainda destaque, os seminários, workshops e as visitas de trabalho conduzidas descentralizadamente e que tem tido um papel importante na consolidação das relações de parceria entre a UCP e os parceiros. No reforço da Capacidade institucional, conclui-se que as capacidades de seguimento e avaliação do PNLP sofreram melhorias substanciais. Houve ganhos na capacidade de gestão financeira e técnica de UCP. Conseguiu de facto, de forma bastante satisfatória, responder eficiente e eficazmente às exigências de um programa multidimensional e de complexidade inegável. De estrutura leve e organizada convenientemente, a UCP apresenta-se dotada de habilidades e capacidades que permitiram assegurar uma coordenação adequada das actividades programadas. Criação de uma capacidade institucional ­ AGECABO: O Governo reconhece o impacto significativo resultante da abordagem adoptada para a implementação dos projectos de acordo com a metodologia de Delegação de Contratos, traduzido no (1) reforço da administração central e dos municípios na gestão e execução de trabalhos públicos, (2) na constituição - 47 - de uma alternativa bem sucedida a abordagem FAIMO, (3) na capacidade de criação de empregos e distribuição de rendimentos nas comunidades mais desfavorecidas, (5) na disciplina e rigor introduzidos no mercado de pequenas obras públicas para o emprego e (6) na sua contribuição para a redução da pobreza. Conforme objectivo inicial do PDSS, a criação da AGECABO se traduz actualmente na institucionalização no país, de uma capacidade institucional para execução de projectos, de forma rápida, eficiente e transparente. As práticas inovadoras adoptadas no quadro da agência, constituem um referencial e capital a ser preservado, pelo que o Governo, continuará a utilizar os serviços da agência no âmbito dos projectos a serem executados através do apoio orçamental como o PRSP, seja no quadro do MCA, ou outros onde a intervenção da agência se mostrar relevante. De notar que a AGECABO actualmente tem em execução uma convenção assinada com o Ministério de Educação para construção de infra-estruturas escolares nos municípios de Santiago e outra assinada com o PNLP no quadro do PSGD. Sobre a diferença entre o objectivo inicial de criação de 20 mil postos de trabalho pessoa/mês e o resultado alcançado de 12.800 postos no final do projecto, uma possível explicação teria a ver com a construção edifícios (centros comunitários, escolas, jardins) em que não obstante a utilidade social e impacto na redução da pobreza serem de extrema importância, não são no entanto trabalho intensivo, como a construção de estradas por exemplo. No entanto, no relatório final apresentar-se-á uma análise mais detalhada sobre essa questão. Contribuição para a Diminuição da Pobreza Sendo um fenómeno multifacetado, a pobreza em Cabo Verde tem sido combatida em várias frentes, algumas delas através da intervenção directa do PDSS, em articulação com outras componentes do PNLP. A contribuição directa do PDSS para a redução da pobreza em Cabo Verde constitui um objectivo de difícil medição uma vez que é praticamente impossível dissociar os efeitos líquidos do PDSS dos das demais intervenções conduzidas tanto pelo Governo Central como local e demais intervenientes. Uma das contribuições importantes do PDSS foi o financiamento do IDRF, cujos resultados permitem calcular e aprofundar a análise das determinantes da pobreza. A pobreza, medida de forma relativa em Cabo Verde, em função de um limiar de 43.250$, é um fenómeno predominantemente rural e assimétrico. Esses dados mostram a existência de uma clivagem sensível entre o meio rural e o meio urbano, bem como diferenças gritantes entre as ilhas. O aprofundamento da análise desses dados para níveis - 48 - geográficos mais desagregados mostra situações ainda mais assimétricas, enfatizando a necessidade de uma maior focalização das políticas no combate à pobreza. O documento de Perfil da Pobreza e o Cape Verde Poverty Diagnostic mostram que as determinantes da pobreza assentam na ausência de rendimento e no baixo nível educacional, entre outros factores. A acção do PDSS ao promover a melhoria de qualidade de vida dos pobres através de um melhor acesso à água potável, ao saneamento básico, à educação e à saúde, concorre directamente para esse objectivo. Assim pode-se assegurar que, o PDSS teve um contributo positivo na luta contra a pobreza em Cabo Verde, proporcionando emprego e rendimento temporário aos grupos mais desfavorecidos e aumentando a cobertura de serviços sociais de base nas bolsas de pobreza identificadas. Estima-se que as acções desencadeadas no quadro do projecto tenham beneficiado potencialmente mais de 270 mil pessoas. Todavia, a leitura destes números requer prudência. A proporção dos indivíduos excluídos da pobreza pelo PDSS não pode ser rigorosamente medido, mas constitui um contributo certamente importante visto que o programa beneficia os trabalhadores das FAIMO e a população rural, maioritariamente os mais desfavorecidos. Os beneficiários parecem concordar que o impacto das obras do PDSS foram positivas na medida em que praticamente 3 em cada 4 (72,8%) consideram que o objectivo proposto com a realização da obra foi alcançado. Contribuição do Governo e dos municípios. Uma das falhas no seguimento do Projecto foi a não implementação de um sistema que permitisse uma quantificação e avaliação permanente do que ia ser realmente a contribuição nacional. Sabemos que essa participação tem sido muito mais do que a simples transferência monetária ou a disponibilização de terrenos para construções. Por exemplo o Ministério de Ambiente, Agricultura e Pesca tem garantido uma participação técnica e material em vários projectos mas não tem sido contabilizado. No mesmo sentido, pode-se falar dos municípios e dos próprios beneficiários. O Governo congratula-se pelo facto de o Banco ter acreditado nas capacidades do País e ter financiado esse importante instrumento de desenvolvimento e que deixa uma experiência extremamente rica para levar a cabo outros desafios do País. A criação de emprego, a formação profissional e alfabetização, a generalização de água domiciliária e casas de banho para todos, a habitação social, o micro crédito, a comunicação e informação para o desenvolvimento, continuam a ser os grandes sectores de intervenção para atingir o público alvo mais carente, designadamente a mulher e a criança, os jovens sem formação, os deficientes e os desempregados de uma forma global. O Ministro Sidónio Fontes Lima Monteiro - 49 - 25°W 24°W 23°W This map was produced by the Map Design Unit of The World Bank. RIBEIRA The boundaries, colors, denominations and any other information GRANDE shown on this map do not imply, on the part of The World Bank Ribeira Grande Group, any judgment on the legal status of any territory, or any Vila das Pombas endorsement or acceptance of such boundaries. Ribeira da Cruz PAÚL Lajes Santo Antão 17°N PORTO (1803 m) Porto Novo 17°N NOVO Mindelo SÃO VINCENTE W I N D W A R D Calhau São Pedro Madeiral São Vicente Santa Luzia Sal Espargos SÃO NICOLAU Ilhéu Branco SAL Ilhéu Raso Riberia Brava Santa Maria Tarrafal Preguica I S L A N D S São Nicolau AT L A N T I C O C E A N Sal-Rei BOA VISTA Norte Boa Vista Povocão Velha Povocão 16°N 16°N CAPE VERDE DS CAPE VERDE LEEWARD ISLANSão Tiago TARRAFAL SANTA Maio Tarrafal CRUZ MAIO SANTA Santa Cruz SELECTED CITIES AND TOWNS CATARINA Vila do Maio Santa Catarina NATIONAL CAPITAL Mosteiros Ilhéus Do Igreja MOSTEIROS São Domingo São RIVERS Rombo 15°N Mt. Fogo DECEMBER SÃO MAIN ROADS (2,829 m) 0 10 20 30 Kilometers Brava Furna PRAIA DOMINGOS IBRD São PRAIA COUNTY (CONCELHO) BOUNDARIES BRAVA Filipe SÃO FILIPE 0 10 20 30 Miles 33383 2004 INTERNATIONAL BOUNDARIES Fogo 25°W 24°W 23°W