Document of The World Bank Report No: ICR2023 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-H0300) ON A GRANT IN THE AMOUNT OF SDR 41.6 MILLION (US$ 55 MILLION EQUIVALENT) TO THE REPUBLIC OF MOZAMBIQUE FOR A HIV/AIDS RESPONSE PROJECT December 28, 2011 Human Development Sector Mozambique Country Department Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective 10/19/2011) Currency Unit = Mozambique New Metical MZM 1.00 = US$ 0.03759 US$ 1.00 = MZM 26.3 FISCAL YEAR January 1 – December 31 ABBREVIATIONS AND ACRONYMS ACTAfrica AIDS Campaign Team for the Africa Region AIDS Acquired Immuno-deficiency Syndrome APE Polyvalent Elementary Health Agent ART Anti Retroviral Treatment ARV Anti Retroviral Pharmaceuticals ASAP AIDS Strategy and Action Planning - service of GHAP BCC Behavior Change Communication CF Common Fund CNCS National AIDS Council CSO Civil Society Organization DGA Development Grant Agreement DP Development Partners ECoSida Entrepreneurs Against AIDS FM Financial Management GACOPI Office of Coordination of Investment Projects GAMET Global AIDS Monitoring and Evaluation Team GHAP Global HIV / AIDS Program – a unit in the WB GNP Gross National Product GOM Government of Mozambique HARP HIV/AIDS Response Project HCSP Health Commodity Security Project HIV Human Imuno-deficiency Virus ICR Implementation Completion (and Results) Report IEC Information, Education and Communication INSIDA National Survey on Prevalence, Risks, Behavior and Information re AIDS ISR Implementation Status and Results report M&E Monitoring and Evaluation MICS Multi-Indicator Cluster Survey (a UNICEF survey instrument) MOH Ministry of Health NGO Non-governmental Organization NPCS Provincial Nucleus for AIDS Control PEN National Strategic Plan (for HIV/AIDS) PEP Post-exposure prophylaxis PEPFAR President‘s Emergency Plan for AIDS Relief (USA) PIU Project Implementation Unit PLHIV People Living with HIV PPTCT Prevention of Parent to Child Transmission RRF Rapid Results Fund RRI Rapid Results Initiative TAP Regional HIV/AIDS Treatment Acceleration Project TB Tuberculosis UN United Nations UNAIDS Joint United Nations Programme on HIV/AIDS UNDP United Nations Development program UNFPA United Nations Population Fund UNGASS United Nations General Assembly Special Session on AIDS UNICEF United Nations Children‘s Fund USAID United States Agency for International Development USG United States Government VCT Voluntary Counseling and Testing Vice President: Obiageli Katryn Ezekwesili Country Director: Laurence Clarke Acting Sector Manager: Jean J. De St Antoine Project Team Leader: Humberto Albino Cossa ICR Team Leader: Kees Kostermans MOZAMBIQUE HIV/AIDS RESPONSE PROJECT Table of Contents A. Basic Information .................................................................................................. i B. Key Dates .............................................................................................................. i C. Ratings Summary ................................................................................................... i D. Sector and Theme Codes ....................................................................................... ii E. Bank Staff ............................................................................................................. ii F. Results Framework Analysis ................................................................................. iii G. Ratings of Project Performance in ISRs ............................................................... vii H. Restructuring (if any) ......................................................................................... viii I. Disbursement Profile ............................................................................................ ix 1. Project Context, Development Objectives and Design............................................. 1 1.1 Context at Appraisal ..................................................................................... 1 1.2 Original Project Development Objectives (PDO) and Key Indicators ................. 2 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification ......................................................................................... 2 1.4 Main Beneficiaries ....................................................................................... 2 1.5 Original Components (as approved) ................................................................. 3 1.5 Revised Components.................................................................................... 4 1.7 Other significant changes ................................................................................. 4 2. Key Factors Affecting Implementation and Outcomes ............................................ 4 2.1 Project Preparation, Design and Quality at Entry ............................................... 4 2.2 Implementation ................................................................................................ 5 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization ...... 7 2.4 Safeguard and Fiduciary Compliance ................................................................ 8 2.5 Post-completion Operation/Next Phase ........................................................... 10 3. Assessment of Outcomes ..................................................................................... 10 3.1 Relevance of Objectives, Design and Implementation ..................................... 10 3.2 Achievement of Project Development Objectives ............................................ 11 3.3 Efficiency ...................................................................................................... 12 3.4 Justification of Overall Outcome Rating ......................................................... 13 3.5 Overarching Themes, Other Outcomes and Impacts ........................................ 14 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops ... 14 4. Assessment of Risk to Development Outcome ...................................................... 15 5. Assessment of Bank and Borrower Performance ................................................... 15 5.1 Bank Performance .......................................................................................... 15 5.2 Borrower Performance ................................................................................... 17 (a) Government Performance ................................................................................... 17 (b) Implementing Agency or Agencies Performance ................................................. 18 (c) Justification of Rating for Overall Borrower Performance ................................... 18 6. Lessons Learned .................................................................................................. 19 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 20 Annex 1. Project Costs and Financing ...................................................................... 23 Annex 2. Outputs by Component ............................................................................. 24 Annex 3. Economic and Financial Analysis .............................................................. 32 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............ 33 Annex 5. Beneficiary Survey Results ....................................................................... 36 Annex 6. Stakeholder Workshop Report and Results ................................................ 38 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR .................... 39 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ...................... 45 Annex 9. List of Supporting Documents................................................................... 47 A. Basic Information HIV/AIDS Response Country: Mozambique Project Name: Project Project ID: P078053 L/C/TF Number(s): IDA-H0300 ICR Date: 12/23/2011 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: APL Borrower: MOZAMBIQUE Original Total XDR 41.60M Disbursed Amount: XDR 41.20M Commitment: Revised Amount: XDR 41.60M Environmental Category: B Implementing Agencies: Ministry of Health CNCS Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 07/25/2002 Effectiveness: 08/15/2003 08/15/2003 06/04/2004 11/01/2005 Appraisal: 10/30/2002 Restructuring(s): 03/14/2006 06/27/2007 09/23/2009 Approval: 03/28/2003 Mid-term Review: 06/30/2005 02/17/2006 Closing: 12/31/2008 06/30/2011 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Substantial Bank Performance: Moderately Satisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Satisfactory Government: Moderately Satisfactory Quality of Supervision: Moderately Satisfactory Implementing Satisfactory i Agency/Agencies: Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance: Performance: C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry Yes Satisfactory at any time (Yes/No): (QEA): Problem Project at any Quality of Yes None time (Yes/No): Supervision (QSA): DO rating before Moderately Closing/Inactive status: Satisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 20 15 Health 40 50 Other social services 35 30 Sub-national government administration 5 5 Theme Code (as % of total Bank financing) Gender 14 10 HIV/AIDS 29 50 Participation and civic engagement 14 15 Population and reproductive health 29 15 Social safety nets 14 10 E. Bank Staff Positions At ICR At Approval Vice President: Obiageli Katryn Ezekwesili Callisto E. Madavo Country Director: Laurence C. Clarke Darius Mans Sector Manager: Jean J. De St Antoine Karen Mcconnell Brooks Project Team Leader: Humberto Albino Cossa James Herbert Coates ICR Team Leader: Cornelis P. Kostermans ICR Primary Author: Cornelis P. Kostermans ii F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) To slow the spread of HIV/AIDS in Mozambique and mitigate the effects of the epidemic, through prevention and care activities Revised Project Development Objectives (as approved by original approving authority) To contribute to slow the spread of HIV/AIDS in Mozambique and mitigate the effects of the epidemic, through prevention and care activities (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years By 2011, the percentage of 15-24 year olds who had sex before age 15 has Indicator 1 : declined to less than 20%. Value Female: 25% quantitative or 27.7% (UNAIDS 2006) <20% Male: 24.8% Qualitative) Date achieved 06/30/2006 10/14/2009 06/30/2011 Comments TARGET WITHIN REACH, as among adolescents (12-14 yrs) only 9.2% of (incl. % girls and 15.3% of boys say to have had sexual relations (INSIDA(2009)). MICS achievement) (2008) indicated that 29% of women aged15-24 had sex before the age of 15. By 2011, the percentage of young men and women aged 15-24 who have had sex Indicator 2 : with a non-marital, non-cohabitating partner in the last 12 months has declined by 10% from the 2002 level. Value 60% (UNGASS 2003- Female: 4.2% quantitative or 50% 2005) Male: 16.4% Qualitative) Date achieved 12/31/2002 10/14/2009 06/30/2011 TARGET ACHIEVED. INSIDA(2009) indicates the percentage of men and Comments women 15-24 yr with two or more sexual partners. MICS (2008) showed 4.7% of (incl. % women 15-24 yr had sex with more than one partner in the last 12 months, i.e. achievement) great improvement over baseline. By 2011 the percentage of young women and men aged 15-24 who used a Indicator 3 : condom at the last time they had sex with a non-regular partner, has increased by 20% from 2002 levels. Value 31% (UNGASS, 2003- Female; 33.1% quantitative or 51% 2005) Male: 37.2% Qualitative) Date achieved 12/31/2002 10/14/2009 06/30/2011 Comments TARGET ACHIEVED for men. Target set in the 4th amendment was 51%, (incl. % which is incorrect as it required an increase from the baseline by 20 percentage achievement) points, rather than 20 percent. Result MICS (2008) for women: 44.4%. iii (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years By 2011, there are at least 1000 HIV/AIDS civil society initiatives funded in Indicator 1 : each province each year. CNCS: 880; RRF: Value 1000/province/ 11 national; and (quantitative 127 nation-wide 1000/province/yr year CNCS 2004-2008: or Qualitative) 3771 in total. Date achieved 12/31/2002 12/31/2008 10/14/2009 06/30/2011 Comments TARGET ACHIEVED. From 2007 CNCS was discouraged to contract sub- (incl. % projects. RRF supported much larger subprojects than CNCS. Many other DP achievement) supported CSO. PAD target was 600 projects/prov./yr. Indicator 2 : By 2011 at least 250 initiatives with the private sector have been funded Value (quantitative 0 250 (2008) 250 557 or Qualitative) Date achieved 03/10/2003 12/31/2008 10/14/2009 06/30/2011 Comments TARGET ACHIEVED. Text in PAD: ―By 2008 at least 450 initiatives with the (incl. % private sector have been funded‖, but DGA mentioned ‖250‖. There were 80 achievement) initiatives in 2006 according to CNCS. By 2011 there are 500,000 clients reached annually by all of the key HIV/AIDS Indicator 3 : related activities delivered by civil society. 98.4% of Value N.A. population heard (quantitative 500,000 (2008) 500,000 445,000 (CNCS, 2006) about HIV or Qualitative) (INSIDA, 2009) Date achieved 12/31/2006 12/31/2008 10/14/2009 06/30/2011 TARGET ACHIEVED. Indicator is poorly defined, but knowledge about HIV is Comments nearly universal amongst 15-49 year olds. Thus, target can be considered as (incl. % achieved. 33.4% of the population aged 15-49 has extensive knowledge of HIV achievement) (INSIDA, 2009). By 2011 at least US$7 million per year are channeled to communities and Indicator 4 : implementing agencies. Value (quantitative N.A. US$ 7 million US$ 7 million US$ 4.7 million or Qualitative) Date achieved 03/31/2003 12/31/2008 10/14/2009 06/30/2011 Comments TARGET NOT ACHIEVED. In 2006, US$4.5 million was channeled to CSO. In (incl. % 2010, US$4.7 million was disbursed to RRF through UNDP for CSO. Many achievement) other DP also supported CSO. By 2011 there will be at least two project proposals financed for each of the Indicator 5 : following areas: (a) Multiple concurrent partnerships; (b) male circumcision. Value 2X2 none 2 x 2 proposals (quantitative proposals iv or Qualitative) Date achieved 10/14/2009 10/14/2009 06/30/2011 Comments TARGET ACHIEVED. Through the RRF, a total of four proposals were funded (incl. % with thematic area (a) multiple partnerships, (b) male circumcision. achievement) By 2011 the number of BCC sessions done in each province will be as follows: 1 Indicator 6 : through radio, 1 through TV, 1 through theatre. 1 through Value radio, 1 many sessions on (quantitative through TV, 1 radio, TV and or Qualitative) through community theatre. theatre. Date achieved 10/14/2009 06/30/2011 Comments TARGET ACHIEVED. Indicator is vague. HIV/AIDS is a regular topic on local (incl. % and national radio, TV and in all provinces CSO activities involving theatre have achievement) taken place. By 2011 at least 500 national leaders and 5000 provincial leaders Indicator 7 : trained/involved in HIV/AIDS campaign. 500 national 10,000 community Value N.A. (2003) leaders leades trained; (quantitative 5000 provincial 9,162 religious or Qualitative) N.A. (2007) leaders leaders trained. Date achieved 03/01/2003 12/31/2008 06/30/2011 Comments Comments: TARGET ACHIEVED. In 2010 alone, the program trained 9,162 (incl. % religious leaders and over 10,000 community leaders were reached with achievement) HIV/AIDS awareness interventions. Percentage of indicators from the national M&E framework that CNCS reports Indicator 8 : annually. Value (quantitative none 75% N.A. or Qualitative) Date achieved 03/01/2003 09/14/2009 06/30/2011 Comments PARTIALLY ACHIEVED. CNCS annual report does not systematically report (incl. % on all indicators of PEN. This indicator is very similar to indicator 21. achievement) Indicator 12: By 2011, 75% of annual action plan (HIV/AIDS activities) in each Indicator 9 : participating Ministry is funded and implemented. Value (quantitative N.A. 75% 18 Ministries No value or Qualitative) Date achieved 03/01/2003 12/31/2008 09/14/2009 06/30/2011 NO VALUE. Meaning of indicator not clear. It does not make sense to set 75% Comments financing of a plan as a target. Interpretation during implementation: % or (incl. % number of participating ministries implementing AIDS plans. Text of PAD and achievement) DGA differ. Indicator 10 : By 2011, 75% of all civil servants have been reached by HIV/AIDS IEC. Value N.A. (2003) 75% (2008) 75% (2009) N.A. (quantitative N.A. (2007) v or Qualitative) Date achieved 03/01/2003 12/31/2008 10/14/2009 06/30/2011 Comments TARGET ACHIEVED. As well over 95% of the general population has heard of (incl. % HIV/AIDS, one may assume that civil servants, who are on average better achievement) educated than the general population, reach at least the same coverage. By 2011 the aggregate annual number of condoms regularly distributed by all Indicator 11 : public institutions and agencies reaches 90,000 annually. Value (quantitative N.A. 30,000,000 90,000 66,535,117 or Qualitative) Date achieved 03/01/2003 12/31/2008 10/14/2009 06/30/2011 Comments ACHIEVED. Indicator in PAD: By 2008 10 million condoms distributed through (incl. % public sector. DGA mentioned 30 million. The 90K target in 5th Amendment is achievement) a mistake. In 2006 CNCS distributed 15M and in 2009 90M condoms. By 2011, 100 voluntary counseling and testing (VCT) centers in place with an Indicator 12 : equitable geographical distribution and coverage. Value (quantitative 36 100 100 359 or Qualitative) Date achieved 03/01/2003 12/31/2008 10/14/2011 06/30/2011 Comments ACHIEVED. PAD target was 50 VCT centers. The 4th Amendment maintained (incl. % original target, although actual 2006 number had already been surpassed. In achievement) 2004, 113 VCT had been established and in 2006 the number was 150. Indicator 13 : Indicator 16: By 2008, ensure that 100% of all blood for transfusion is screened. Value (quantitative 100% 100% 100% 100% or Qualitative) Date achieved 03/01/2003 12/31/2008 10/14/2009 06/30/2011 Comments ACHIEVED. The Mozambique UNGASS report, 2008, stated that 100% of (incl. % blood for transfusion was screened for HIV, but also noted: only 35.5% of blood achievement) units were tested under conditions of adequate quality control. Indicator 17: By 2008, 100% of secondary and tertiary health facilities have the Indicator 14 : capacity to diagnose and test STIs Value (quantitative N.A. 100% 100% 100% or Qualitative) Date achieved 03/01/2003 12/31/2008 10/14/2009 06/30/2011 TARGET ACHIEVED. Already by 2007 the target was achieved. The text in Comments PAD refers to: public health center facilities have adequate drugs and trained (incl. % personnel to treat sexually transmitted infections. Health centers are primary achievement) care facilities. Indicator 15 : Indicator 18: By 2011 100% of HIV+ TB patients receive cotrimoxazole. Value (quantitative N.A. 100% 96.2% or Qualitative) Date achieved 10/14/2009 10/14/2009 06/30/2011 Comments TARGET ALMOST ACHIEVED. In 2007, 92% of HIV positive TB patients (incl. % were receiving cotrimoxazole. vi achievement) Indicator 19: By 2011 40% of HIV+ people receive isoniazide preventive Indicator 16 : therapy. Value (quantitative N.A. 40% 10% or Qualitative) Date achieved 10/14/2009 10/14/2009 06/30/2011 Comments (incl. % TARGET NOT ACHIEVED. Baseline in 2008 was 3%. achievement) Indicator 20: By 2011 20% of men aged 15-49 in priority provinces will have Indicator 17 : received male circumcision. Value (quantitative 20% No data or Qualitative) Date achieved 10/14/2009 06/30/2011 Comments NOT ACHIEVED. No national data. Program just started. Likely 20% (incl. % circumcision prevalence only in provinces with traditionally a high circumcision achievement) rate. RRF supported circumcision for max. 52,200 males. Indicator 21: By 2011, CNCS can report annually on at least 75% of the Indicator 18 : indicators in its national HIV M&E Framework and that a report is disseminated to national leaders. Value (quantitative 75% N.A. or Qualitative) Date achieved 10/14/2009 06/30/2011 PARTIALLY ACHIEVED. CNCS produces an annual report, which is widely Comments disseminated and put on its website. However, the report does not systematically (incl. % report on all indicators of the national strategy and deals mainly with CNCS own achievement) activities. G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 06/25/2003 Satisfactory Satisfactory 0.00 2 11/02/2003 Satisfactory Satisfactory 2.63 3 05/20/2004 Satisfactory Unsatisfactory 3.15 4 11/18/2004 Satisfactory Unsatisfactory 4.49 5 06/29/2005 Satisfactory Unsatisfactory 10.06 Moderately 6 02/28/2006 Satisfactory 14.05 Unsatisfactory Moderately 7 09/07/2006 Satisfactory 22.23 Unsatisfactory 8 04/11/2007 Satisfactory Satisfactory 29.45 9 06/28/2007 Satisfactory Satisfactory 31.47 vii 10 01/17/2008 Satisfactory Moderately Satisfactory 33.47 Moderately 11 04/08/2008 Satisfactory 38.13 Unsatisfactory Moderately 12 07/11/2008 Satisfactory 38.13 Unsatisfactory Moderately 13 02/23/2009 Unsatisfactory 42.09 Unsatisfactory Moderately 14 10/28/2009 Unsatisfactory 45.57 Unsatisfactory 15 02/09/2010 Moderately Satisfactory Moderately Satisfactory 55.69 16 02/22/2011 Satisfactory Satisfactory 60.39 17 06/29/2011 Moderately Satisfactory Moderately Satisfactory 60.39 H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions Definitions of facilitating agent 06/04/2004 S U 3.15 and operating costs were broadened. Definition of facilitating agent further broadened to include 11/01/2005 S U 11.70 public sector agencies and institutions. Proceeds of Grant could now be 03/14/2006 S MU 14.86 used for payment of applicable taxes. 06/27/2007 Y S S 31.47 Established new performance monitoring indicators. 09/23/2009 N MU U 45.57 Establsihed Grant Management Agency in UNDP. If PDO and/or Key Outcome Targets were formally revised (approved by the original approving body) enter ratings below: Outcome Ratings Against Original PDO/Targets Moderately Satisfactory Against Formally Revised PDO/Targets Moderately Satisfactory Overall (weighted) rating Moderately Satisfactory viii I. Disbursement Profile ix 1. Project Context, Development Objectives and Design 1. As Mozambique entered the new millennium it was estimated that the country had one of the fastest growing AIDS epidemics in Sub-Saharan Africa with an adult HIV prevalence rate of roughly 12 percent (2000). 1 Multiple factors converged to create an environment of high vulnerability, including: a long and protracted civil war, large population movements, proximity to high prevalence countries which hosted its refugees (e.g. Zimbabwe, Zambia, Swaziland, Malawi and South Africa), and a deterioration of the physical infrastructure and social fabric. The highest prevalence rates were found along the major trade corridors and in areas with migrant miners, while the North had the lowest rates. 2. HIV/AIDS was prioritized as a major development issue in both Mozambique‘s Five Year Program, 2000-2005 and its Poverty Reduction Strategy. With a per capita GNP of only US$210 there was broad-based recognition that tackling HIV/AIDS would require a multi- sectoral approach which dealt with poverty and vulnerability, the underlying causes of the epidemic. 3. The Bank‘s 2001-2003 Country Assistance Strategy (Document 20521) cited HIV/AIDS as the single largest threat to Mozambique‘s development prospects and proposed this US$55 million operation to establish the foundation for the national response when few partners were on the ground. The project was part of the second phase of the Multi-country AIDS Program for the Africa Region (MAP), launched in 2000, which aimed to ―dramatically increase access to HIV/AIDS prevention, care and treatment programs, with emphasis on vulnerable groups‖. 1.1 Context at Appraisal 4. After the first case of AIDS was detected in 1986, the Ministry of Health (MOH) established the National AIDS Control Program in 1988 and developed three medium term plans over the following decade. The Council of Ministers adopted the first Strategic Plan to Combat HIV/AIDS and STDs, 2000-2002, (PEN) which was developed in a consultative and participatory fashion. A National Council to Combat HIV/AIDS (CNCS) was established in May 2000 with the following strategic directions: (i) to focus on priority vulnerable groups (e.g. young girls; and highly mobile populations including soldiers, miners, prisoners, truck drivers and commercial sex workers); (ii) to ensure relevance (i.e. interventions should be efficacious); (iii) to empower people living with HIV/AIDS (PLHIV) and contribute to reducing stigma; and (iv) to target the major trade corridors (Maputo, Beira and Nacala) and the surrounding communities. 5. The campaign against HIV/AIDS was originally led by the MOH, but with the approval of the PEN the leadership shifted to the CNCS. Professional staff was assigned to each province and national coordination fora were held annually. Thematic groups were formed on the main strategic areas (e.g. prevention, home-based care, VCT and orphans). 6. The timing of the HIV/AIDS Response Project (HARP) was opportune as Mozambique had few partners and large funding gaps. 2 The operation provided critical support following up to 1 Adult prevalence estimate was based on the first national surveillance data (MOH, 2000). For Maputo city, which has most data available, adult prevalence increased from 1 percent in 1988, to 9.9 percent in 1998,and further to 13.0 percent in 2000. 2 While donors had originally pledged around US$120 million for the PEN, this funding was not forthcoming and HARP was the first large contribution towards its implementation. 1 the Health Sector Recovery Project, which supported the country‘s health sector rebuilding efforts. During implementation many DPs came on board to support the national response, of which the largest included PEPFAR and GFATM, which brought with it the country coordinating mechanism (CCM). 1.2 Original Project Development Objectives (PDO) and Key Indicators 7. The Project Development Objective, as stated in the Project Appraisal Document (PAD) is to slow the spread of HIV/AIDS in Mozambique and mitigate the effects of the epidemic through prevention and care activities. While the Development Grant Agreement (DGA) articulated the PDO in a slightly different way 3, the version in the PAD is the one which was used for tracking implementation progress and project performance. 8. The original key outcome/impact indicators are as follows:  By 2008 HIV prevalence amongst pregnant women in the 15-24 age group has been reduced by 25% from 2000 levels.  By 2008 the age of sexual debut has increased by one year over the 2002 age.  By 2008 the number of sexual partners outside of primary union in last 12 months for persons 15-49 has declined by 25% from 2002 level.  By 2008 the rate of condom use in sexual encounters outside of the primary relationship in last 12 months by persons 15-19 has declined 25% from 2002 levels. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 9. The project was restructured in June 2007 to take explicitly into account that IDA was only one player in the fight against HIV/AIDS. The revised PDO is: ―to contribute� to slow the spread of HIV/AIDS in Mozambique and mitigate the effects of the epidemic through prevention and care activities. 10. There were also changes made to key outcome indicators. The indicator concerning the drop in HIV prevalence was dropped, as the Bank project could not be held accountable for such an outcome. To better align outcome indicators to the standard United Nations General Assembly Special Session on AIDS (UNGASS) indicators that the national program adopted, several other adjustments were made as discussed below under the Monitoring and Evaluation (M&E) (Section 2.3). 1.4 Main Beneficiaries 11. Given that Mozambique had a generalized epidemic it was expected that the entire population would benefit from information education communications (IEC) and behavioral change communications and other advocacy and awareness creation interventions, as well as 3 In the Development Grant Agreement the PDO is to ―assist the Government of Mozambique to finance the implementation of its National Strategic Plan to Combat HIV/AIDS and STDs. The strategy is designed to slow the spread of HIV/AIDS infection and mitigate the effects of the epidemic, through prevention, care, treatment and mitigation activities.‖ 2 targeted voluntary counseling and testing (VCT), promotion of condoms, and improved access to HIV health services. The project would benefit high risk groups, including young people especially girls; PLWHIV, orphans, highly mobile workers, teachers, soldiers, miners, prisoners and commercial sex workers. It would also benefit populations in high transmission geographic areas. 1.5 Original Components (as approved) 12. The project had five components: Component 1: Community and Civil Society Initiatives (US$ 28 million). This component would empower communities to respond effectively to the HIV/AIDS epidemic. A central activity would be mobilizing communities, promoting local initiatives and strengthening the capacity of local actors. A Community and Civil Society Facility held by CNCS would be created to finance appropriate HIV/AIDS related activities carried out by eligible applicants, such as community and faith based organizations, NGOs, the private sector, associations and other organizations. The management of the Facility would be decentralized to the provincial level. Component 2: Capacity Building for the Civil Society HIV/AIDS Response (US$5.5 million) would support efforts to create AIDS competent communities by improving skills and increasing implementation capacity of implementers, supporters, and target groups active under the Community and Civil Society Initiatives. Activities to be funded would include hiring and training ―Facilitating Agents‖ in each province, courses and activities to build awareness and capacity for leadership in the public sector, civil society and private sector, and the formulation and distribution of training modules and packages of IEC materials. Component 3: Government Multi-sector Response (US$ 7 million): Ministries and subordinate institutions would be supported in the development and implementation of HIV/AIDS related programs directed toward their own personnel (particularly high risk staff such as soldiers, police, extension agents, teachers, health workers) and their families, as well as programs directed towards their clients. The program would concentrate on non-health ministries and other key public sector institutions to effectively respond to the epidemic, with emphasis on prevention and care for infected and affected families. In each ministry a Focal Point would be appointed. With guidance from provincial Government and Ministry of State Administration, District Administrators would be engaged to foster development of District HIV/AIDS plans. Component 4: Strengthening and Scaling Up Health Sector Services for HIV/AIDS (US$17.5 million) aimed to scale up the response of the health sector to the HIV/AIDS epidemic and to provide technical leadership on treatment and care for people living with HIV/AIDS. The component would support (i) strengthening of the Integrated Health Network providing voluntary counseling and testing and HIV/AIDS related services in four central provinces; (ii) increasing the supply of HIV related drugs and supplies, condoms, drugs for opportunistic infections, anti- retroviral for the prevention of parent to child transmission (PPTCT) of HIV and post-exposure prophylaxis; (iii) strengthening clinical laboratory capacity for CD4 monitoring and diagnosis of opportunistic infections; (iv) measures to enhance bio-security; (v) measures to enhance blood safety; (vi) training and anti-retroviral drugs for post-exposure prophylaxis for health workers; (vii) improved management of opportunistic infections; and (viii) monitoring and evaluation of the component and the overall progress of the epidemic. 3 Component 5: Institutional Development for Program Management (US$6 million) supported the Executive Secretariat of the CNCS, strengthening its capacity to lead the country in the campaign. 1.5 Revised Components 13. The project components were not revised. 1.7 Other significant changes 14. Three significant changes are noteworthy over the life of the project. The first set of changes was made in the initial years (2004-2006) of the project to fine-tune the design, to better adapt to evolving country needs, to improve implementation and performance. This involved several minor amendments to the DGA. The second set of changes which took place in the final years of the project (2007-2009) was probably the single most important as it involved modifications to the PDO, outcome indicators, and financing modalities. The August 2007 restructuring required Board approval as it involved a modification to the PDO. 4 It also introduced an innovative financing modality whereby IDA funds were pooled in a Common Fund with those from other development partners, reducing the transaction cost for the government. 15. The final set of changes resulted in modifications to the institutional and governance structures for the project and national program. The October 2009 amendment took into account the new Grant Management Agency (GMA) established in MOH to channel funds and facilitate implementation of project activities in participating ministries, CSOs and sub-grant beneficiaries. This resulted in a new governance structure as the GMA would take over some of the roles and responsibilities of the CNCS which would now focus on coordination, facilitation and oversight of the national response. 5 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 16. The project was prepared in an efficient manner to respond quickly to the government‘s request for support. Many of the early lessons from MAP II 6 were effectively incorporated into the design of this operation. Project preparation focused on implementation arrangements, engaging civil society groups, piloting of community based programs, and strengthening fiduciary capacity of the CNCS. PEN I had been prepared by Government and development 4 The August 2007 restructuring consisted of: (i) a modification of the PDO; (ii) a change in the Results Framework to ensure that indicators were better linked to project activities; (iii) a change in the funding mechanism allowing IDA funds to channel funds to the Common Fund of the CNCS, pooling funds with bilateral and other multilateral organizations; (iv) a reallocation of funds between categories; and (v) extension of the closing date to December 31, 2009. 5 This final amendment also aimed to strengthen the links between reproductive health and HIV with most of the funds to be used to procure commodities. It also extended the closing date by one year. 6 Key lessons from MAP I included: (a) lack of project readiness slowed down implementation, (b) exceptional procedures were needed to engage CSOs, (c) national AIDS council should focus on coordination rather than control or implementation, (d) monitoring and evaluation is critical, (e) there was need for funds for knowledge management, and (f) intensive Bank supervision was needed. 4 partners (especially UNAIDS) in a highly participatory way and commitment and enthusiasm for the Plan were strong which provided a conducive environment for the Bank project. Bank financing would be the first substantial contribution to the CNCS. By introducing firmer fiduciary arrangements it provided confidence to other donors to channel funds through the public sector. While the project included the standard MAP components the flexibility of the MAP instrument, and the ―learn by doing approach‖, proved to be a good mechanism to use in the Mozambique context. In spite of these positive aspects, the speed of preparation (i.e. 10 months between identification and approval) may have compromised readiness, as reflected in subsequent delays. Furthermore, as was the case with other MAPs, insufficient attention was paid to M&E which was reflected in weaknesses in the Results Framework (discussed below). 2.2 Implementation 17. The project suffered from numerous implementation delays which stemmed from several factors, including insufficient readiness; weak institutional capacity in relation to the ambitious agenda; high staff turnover, and cumbersome procedures. As a result, the project was often in problem status during the implementation phase. Although Mozambique was politically stable and top leadership changed infrequently, just below the top managerial level high staff turnover and lengthy vacancies were the order of the day. Low capacity of virtually all agencies adversely affected the implementation of numerous activities. Nevertheless, in spite of numerous problems and delays, the government with the Bank team identified innovative solutions throughout the project cycle and repeatedly explored options for addressing these shortcomings. As a result, by project completion implementation was considered largely successful with most targets achieved. Several notable examples are cited below. 18. The effective implementation of the main component of the project, which aimed to empower communities to respond to the HIV/AIDS epidemic, was impeded by weak implementation capacity and cumbersome procedures which surpassed the capacity of most CSOs, for many till the end of the project. CNCS did not have the capacity to manage thousands of sub-projects, which caused delays in approvals and replenishments. 7 Moreover, much capacity building needed to be done for civil society organizations which had limited experience in project and financial management. Fiduciary requirements were heavy for rural communities where banking was virtually inexistent and three quotations hard to obtain. The 20 percent contribution originally requested from the NGO own resources also caused difficulties for many. The concept of sub-project was new and initially the Mozambique Tribunal Administrativo got involved in approval of each. Finally, reporting requirements were too heavy for most CSOs. 19. Many of these issues were successfully rectified during the life of the project. UNAIDS and others, including private firms, were brought in to provide capacity building to CSOs for proposal writing and sub-project implementation; and to provide technical assistance to CNCS. As Mozambique received large contributions from PEPFAR 8 from 2006 onwards there was a 7 CNCS would contract CSOs for sub-projects, which could vary from as little as less than US$1,000 to over US$20,000 (Category A: below US$1000, Category B: between US$1000 and 20,000, and Category C: over US$20,000). In total HARP supported 3771 sub -projects (Category A 179 sub-projects, Category B 776, and Category C 2816 sub-projects). 8 According to the PEPFAR website, Mozambique received $835.6 million to support comprehensive HIV/AIDS prevention, treatment and care programs from FY 2004 to FY 2009. 5 shift towards supporting stronger NGOs with higher levels of capacity, and an effort to increase the average size of the sub-projects. Following the 2007 restructuring a Grant Management Agency was contracted to handle the selection of CSO sub-projects. While this was good idea in principle, it did not work out as expected given that the CNCS management could not agree to relinquish these responsibilities; as a result the GMA resigned after one year and the CNCS continued to handle these functions until the end of 2009, diverting attention from its main responsibility of coordinating the national response. 20. Activities funded under the public sector multi-sectoral response component, which reached close to 40 institutions, also faced similar capacity challenges which were reflected in implementation delays and disbursement lags. Focal points, at central and provincial level, were part-timers and did not receive any additional payment in the spirit of mainstreaming. CNCS provided training at central and provincial level which partly improved the situation. With the creation in 2007 of a new Ministry of Civil Service, an HIV/AIDS strategy for civil servants was developed and funds were received to target prevention activities in 12 priority institutions. 21. While the health sector component was implemented relatively well as the MOH had experience with implementing Bank projects through a strong project unit (GACOPI) that also managed the Bank-funded Treatment Acceleration Project (TAP) 9 some implementation challenges arose with changes in the leadership of the MOH. Given the constrained implementation, especially of the contracting of CSOs a consultation was held with the Prime Minister and agreement was reached to: (i) procure commodities through the United Nations Population Fund (UNFPA) which had a strong track record and would be in a position to efficiently procure the additional US$12 million allocated to the MOH to scale up HIV and related supplies; and (ii) establish a Rapid Results Fund (RRF) to be managed by UNDP to focus attention on prevention, reproductive health, and TB/HIV integrated care. Taking the lessons learned by CNCS into account, UNDP contracted CSOs with more capacity and the average size of contracts for sub-projects would more than double. Smaller CSOs and NGOs could only participate through joint ventures with larger organizations. Following these changes, the implementation of the health component became more strategic but some delays persisted, highlighting the challenges of channeling funds to civil society groups. Finally, it should be noted that the drop in ARV prices led the Bank to allow procurement of these drugs for life saving AIDS treatment. 10 9 The regional Treatment Acceleration Project (TAP, P082613), which was a first for Bank-financing of ART, had a Mozambique component. HARP‘s design included only procurement of drugs for opportunistic infections and ART for prevention of parent to child transmission (PPTCT) and post -exposure prophylaxis (PEP). TAP showed that NGOs could play an important role in rapid expansion of ART and provided MOH with a framework for the contracting of NGOs with clear objectives for service delivery and measurable targets. 10 The dramatic price drop for ARV over the project‘s life from over US$10,000 to only US$100-400 per year had a significant impact on the treatment policy of GOM and WB support. While in the beginning of the project the Bank would only finance treatment for opportunistic infections and ART only for PPTCT, towards the end first line ARV were also allowed to be procured for general purpose although this did not happen on any large scale, as other sources of funding were used for it. It did however fundamentally change the way civil society organizations could provide care to PLHIV. 6 22. To address cross cutting capacity constraints, HARP provided substantial support for capacity building and staffing of the CNCS in management and administration; fiduciary arrangements; technical, communications and M&E. In turn, CNCS provided considerable training to implementing agencies, journalists, youth groups, and the CNCS sub-structures at the provincial level. Bank support was instrumental in the development of PEN II and III, which were inspired by the AIDS Campaign Team for Africa (ACTAfrica) new Agenda for Action, 2007-2011 as well as support from the Bank‘s AIDS Strategy and Action Planning service (ASAP) that helped to prioritize national interventions to be supported under the plans. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 23. Design. While implementation of a multi-sectoral response for a ‗learning by doing‘ operation would require a well-developed monitoring and evaluation system, the M&E architecture and the institutional arrangements of the national response and of the Bank project were still largely under development at the inception stage. CNCS had a small M&E unit and all provincial nuclei had only one Database Agent. The weak start was exacerbated by the fact that during implementation the planned international TA was never hired. This hampered especially CNCS‘s role to collect and analyze periodic reports of implementers, and prepare annual implementation reports on the national response in the effort to combat AIDS. 24. There were discrepancies between the articulation of the large number of indicators in the DGA and in the PAD. 11 Several indicators from the PAD were not in the DGA and were never measured and were subsequently dropped. Other indicators were also dropped during the various amendments as knowledge about the selection of indicators improved and there was a greater emphasis on outputs rather than process indicators (e.g. approval of a national strategic plan); a move away from indicators for which an individual project could not be held accountable (e.g. reduction in HIV prevalence). 25. During project preparation, the M&E unit of CNCS received Bank support on budgeting/costing, financial planning and program monitoring. Guidelines were developed and indicators defined. With Bank support, data bases were established at national and provincial levels. The heavy focus on fiduciary matters, especially for grant management purposes, was to be expected for a project which needed to set up systems, but it may have taken attention away from collection of core epidemiological and behavioral data at baseline. 26. Implementation. With external support from the UN, a National M&E Framework was developed and published in 2006 in the context of the preparation of the second phase of the National HIV/AIDS Strategic Plan. CNCS received support from WB Global AIDS Monitoring and Evaluation Team (GAMET) in 2005-2006. However, in 2007 CNCS still lacked dedicated information technology experts for programming, database design and maintenance and computer-based communications. Hiring of international technical assistance was recommended and eventually funded by USAID. For its provincial nuclei CNCS recruited M&E officers from 2007 onwards, which would allow for capacity building in decentralized program planning and implementation. 11 For example, the legal agreement stated: By 2008, the percentage of young men and women aged 15-24 who have had sex with a non marital, non-cohabitating partner in the last 12 months has declined by 10% from the 2002 level‖, but the text in the PAD was as follows: ―By 2008 the number of sexual partners outside of primary union in the last 12 months for persons 15-49 has declined by 25% from the 2002 level‖. The indicator was dropped at the fourth amendment. 7 27. Since 2007, CNCS was ―beginning to be able to fulfill its mandate‖ regarding M&E. CNCS was able to report reliably on about two-thirds of the M&E indicators in the National Framework. CNCS and its provincial nuclei were able to report on grants made and amounts disbursed for the sub-projects, and on most of the output indicators reported by a sizable number of the civil society implementers. Information about activities of the public and private sector s and data on activities of entities not receiving funding through CNCS were incomplete. 28. A national M&E framework was developed with an implementation plan for the second phase of the National Strategic HIV/AIDS Plan (PEN II). The plan outlined the key indicators, defined data sources, responsibilities of various partners, and information flows. Implementation required a lot of training and the cooperation of all development partners. Efforts to develop an integrated work plan proved challenging. As a result, a functioning M&E system in accordance with the Three Ones principles 12 is still not fully in place. At present CNCS‘ data sets are still incomplete as many donors do not report their assistance systematically and neither do the implementing agencies. Only organizations executing sub-projects financed by CNCS reported regularly. CNCS still faces enormous challenges to report on all activities in the country. 29. One of the important contributions of the Bank, in collaboration with other development partners was to strengthen knowledge of the HIV/AIDS epidemic in Mozambique. Some large surveys were conducted which provided good information on levels, trends and patterns in HIV/AIDS as well as knowledge, attitudes and sexual practices 13 . Likewise, with GAMET support considerable analytical work was done on the drivers of the epidemic. 14 30. Data Use. CNCS was able to publish annual reports since 2001 on its activities and these reports have gained in comprehensiveness over time. A technical working group on M&E, jointly chaired by the MOH and CNCS and with wide participation from many stakeholders including the Bank, helps with analysis and drawing conclusions from the data. The group was especially active during Annual Reviews and contributed to preparations of various reports, including the UNGASS report. The task team and other units of the Bank have provided much technical assistance over the years to strengthen the strategic focus of the national response through better M&E. The discussions helped to redirect attention towards prevention, including the need to reduce concurrent sexual partnerships and unsafe sex and boost condom use, all activities supported under the project. Data sharing by DP is, however, still an issue and CNCS is reviewing options to encourage data sharing. 31. The private sector sub-component was thoroughly evaluated after the two-year contract with the managing firm ended. The lessons learned through this evaluation could not be immediately applied as a second generation of sub-projects with the private sector did not take place. 2.4 Safeguard and Fiduciary Compliance 12 UNAIDS‘ Three Ones principles refer to: (a) One agreed HIV/AIDS Action Framework that provides the basis for coordinating the work of all partners, (b) One National AIDS Coordinating Authority, with a broad based multi-sector mandate, and c) One agreed country level Monitoring and Evaluation System 13 DHS (2004), INSIDA (2009) 14 See for example Aide Mémoire June, 2007 mission. 8 32. The project triggered the safeguard policy regarding Environment Assessment (OP/BP/GP 4.01) and was categorized as Category B. The assessment revealed serious deficiencies such as: lack of efficient management of health care waste, carelessness of clinical staff, insufficient waste containers, lack of security for collectors, and lack of an efficient treatment system (e.g., 90% of the hospitals disposed of their wastes within the hospital compounds through burial or incineration). A comprehensive environmental action plan (which went beyond this project) was developed in a participatory fashion and included infection control and waste management guidelines. The national plan, estimated at US$ 42.5 million, had two components: (a) strengthening of the regulatory framework; and (b) provision of equipment for selection, sorting and treatment of health care waste of public health care facilities. Medical supplies and waste treatment equipment, such as incinerators, were procured for some hospitals, but there was limited evidence of any systematic follow-up by the Bank of the implementation of the environmental action plan. 33. Financial Management (FM): Mostly with USAID support an accounting system was put in place to respond to the needs of both Government and donors, recording donations and use of funds. The system was installed and staffed at the center and in all provinces. The project had three Special Accounts, two of which were managed by CNCS: one for operating costs and one for sub-projects. Due to the volume of funds to be managed under component 4, a separate special account was established for MOH, administered by GACOPI which would also carry out all procurement for MOH, and for CNCS in case of ICB, under the project. 34. The FM procedures put in place proved to be overly complicated: two special accounts resulted in 33 project accounts at provincial level. Payment procedures were too complex for the provinces and the small community organizations which sometimes were unable to open a bank account. This resulted in delays in implementation of sub-projects under component 1. In some cases, the provincial nuclei of CNCS would manage the finances for the CSOs on their behalf, presenting a potential of conflict of interest. The 20% contribution required from Government hampered implementation and added to the complexity of the procedures, as it required special bank accounts. The 20% contribution was abandoned in 2007. Over the years the number of accounts managed by CNCS decreased from 27 to just 7. 35. At the beginning of the project the Bank was the main financier of the PEN. However, this dramatically changed over the life of the project. This provided good opportunities for pooling resources. After the fourth amendment became effective, the Bank joined the Common Fund, a pooling arrangement of various bilateral donors and GFATM. This Fund financed CNCS and with the Bank‘s involvement the financing became better linked to CNCS‘ operational pla ns. In 2007 the Bank signed a memorandum of understanding with several bilateral partners to join the ―Common Fund‖ which would support the public components of the project, especially CNCS. The RRF was also financed together with other donors. 36. Procurement: The CNCS Secretariat was charged with oversight for the whole program and hired a procurement officer for this task. Larger packages requiring international competitive bidding and all packages pertaining directly to component 4 would be handled by MOH which had contracting experience in this area. The provincial nuclei of CNCS carried out procurement of small value including recruitment of local staff with prior approval of the central Unit. Procurement plans of good quality were regularly sent to the Bank. Record keeping, however, was not always in good order. 37. The contracting of CSOs under component 1 would be done by CNCS, using the Simplified Procurement and Disbursement for Community-based Investments, under the title ―sub-projects‖. The sheer number of subprojects to be contracted through CNCS was clearly beyond the capacity of the young institution. It also pulled it away from its main area of focus: 9 coordination of the national response against HIV/AIDS. Procurement by MOH was relatively smooth in the beginning of the project, but later suffered as well from weakening capacity in GACOPI. 2.5 Post-completion Operation/Next Phase 38. The Bank considered a follow-on operation, named “HIV/AIDS Response II‖, but this plan was abandoned, as many development partners had entered the AIDS arena, injecting substantial resources into the national HIV/AIDS program. The Bank therefore decided to focus its follow up support on health sector reform which would address systemic issues in the sector, complementing support for HIV/AIDS from other DPs. 39. Given that the Bank‘s contributions towards the end of the project were largely flowing through pooling arrangements with other development partners and since these arrangements will continue to exist, the project activities will not come to a halt with the end of HARP although some activities may be scaled down. There was concern amongst donors that with the Bank out of the pooling arrangement fiduciary oversight might suffer in the future. The DPs hoped that WB will use its influence with the Ministry of Finance to keep HIV as a priority at macroeconomic level which has in fact occurred. The private sector initiatives of component 1 evolved into a self-standing initiative led by the private sector itself. CNCS continued to meet through ECoSida 15 with the private sector, proving sustainability beyond the project support. 16 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation 40. With a rapidly rising and generalized epidemic, the project‘s objective was highly relevant to Mozambique‘s development agenda and continues to be so, as the impact of HIV/AIDS on human development and poverty is well documented. The project‘s design to rapidly scale up through the involvement of CSOs and to adopt a multi-sectoral approach were also appropriate, as AIDS had become a national emergency that required broad based participation. The decision to implement the project largely through CSOs was an appropriate strategy, given that these institutions have a solid knowledge of their beneficiaries. Moreover, CSOs were well placed to support the design of interventions to address the precarious situation and vulnerability of their constituencies. Although the project was originally maybe overly ambitious and implementation arrangements not always well adapted to the local conditions, a series of amendments to the DGA and adjustments in implementation arrangements responded to these shortcomings so that relevance could be maintained or enhanced. Following the 2007 restructuring there was a gradual shift in main focus from mitigation to prevention with the establishment of the RRF. The relevance of the public sector interventions could have been further enhanced by a more targeted approach both in terms of the number of institutions involved and the type of activities that were supported. Finally, it should be noted that the analysis of the HIV/AIDS situation in Mozambique in the appraisal document correctly flagged 15 ECoSIDA is a national Mozambican private sector initiative in the fight against HIV/AIDS, mainly financed by the private sector itself. It has around 50 active members. 16 In 2010 it attended 88 meetings to mobilize managers and 21 firms started work place interventions. 10 the urgency and gravity of addressing the epidemic, particularly as AIDS was almost always deadly, as ART was still prohibitively expensive. 17 41. Relevance is rated high as the project reflected a proper diagnosis of AIDS as both a major public health problem and a development priority, and over the life of the project the Government and Bank team made adjustments in the implementation arrangements to maintain or further enhance its relevance. 3.2 Achievement of Project Development Objectives 42. The achievement of the PDO is rated as substantial. The HIV/AIDS epidemic in Mozambique appears to have stabilized over the past decade. While data from two comparable data resources are not available the HIV estimates appear to suggest that the epidemic has been contained, given the sizable resources injected into the national response, including the contribution from IDA through this project. According to the first and only nationally representative survey (INSIDA (2009)) of HIV prevalence and knowledge, attitudes and practices, Mozambique has an HIV adult infection rate of 11.5 percent with slightly higher rates of female infection (females: 13.1 vs. males: 9.2 percent). As noted in the PAD, the baseline figures which came from the 2000 surveillance data showed a national adult prevalence of 12.2 percent, with the South Region 13.2, the Center 16.5 and the North 5.7 percent. For INSIDA these figures are 17.8, 12.5 and 5.6 percent respectively. While it is not possible to establish a counterfactual, there is no reason to think why Mozambique would not have reached the very high prevalence rates of all its neighboring countries in the absence of the PEN, which the Bank supported. 43. It should be noted that not only did the project contribute to slow the spread of HIV/AIDS but it also assisted to put in place the systems and capacity for the implementation of the PEN, as it was its first large financing and was instrumental in attracting sizeable financing from other multi- and bilateral donors. 44. As summarized below, there was generally good progress on many of the PDO indicators. The percentage of 15-25 year-olds who had sex before the age of 15 declined from 60 percent at baseline to 25 percent in 2009. The end of project target (i.e. less than 20 percent) was therefore well within reach and may have been achieved by project closing. Unfortunately, more recent data are not available. Achievement of PDO indicators Indicator Baseline Achievement Comments 1: By 2011, the percentage of 15-24 year 27.7% ♀: 25% Target within olds who had sex before age 15 has (UNAIDS ♂: 24.8% reach declined to less than 20%. 2006) 2: By 2011, the percentage of young men 60% ♀: 4.2% Target Achieved and women aged 15-24 who have had sex (UNGASS ♂: 16.4% 17 The subsequent steep decline in prices of ARV changed this picture completely and due to new interventions AIDS became for many a chronic disease. 11 with a non marital, non-cohabitating 2003-2005) partner in the last 12 months has declined by 10% from the 2002 level. 3: By 2011 the percentage of young 31% ♀: 33.1% Target achieved women and men aged 15-24 who used a (UNGASS, ♂: 37.2% for men. condom at the last time they had sex with 2003-2005) a non-regular partner, has increased by 20% from 2002 levels. 45. The percentage of young men and women aged 15-24 who had sex with a non-marital, non-cohabitating partner in the last 12 months declined dramatically from 60 percent at baseline to 4.2 percent for women and 16.4 percent for men, according to INSIDA (2009). The MICS (2008) found similar results with only 4.7 percent of 15-24 year old females reporting having sex with more than one partner in the last 12 months. While there may be some issues with the comparability and reliability of this self reported data, nevertheless, it appears that messages about HIV behavior change and the need to reduce multiple concurrent sexual partners, one of the main drivers of the epidemic, were paying off. 46. The percentage of young women and men who used a condom when they last had sex with a non-regular partner increased from 31 at baseline to 35.1 percent (women 33.1 percent and men 37.2 percent). The MICS (2008) found a similar but even higher level of condom use with some 44.4% of women aged 15-24 reporting condom use during their last sexual relation with a non-marital or non-cohabitating partner. The target for condom use appears to have been achieved for males. 18 47. In summary, for the three PDO indicators the targets were largely achieved. For the 17 intermediary indicators 11(65 percent) were achieved, 2 (12 percent) were partially achieved, and 4 (23 percent) were not. The achievements demonstrate the substantial impact of the project‘s initiatives. The numerous activities of the project in communities, firms and institutions raised HIV awareness enormously (by project end over 98 percent of both men and women had heard of HIV, according to INSIDA (2009)) in an effective manner, leading to the intended behavioral changes. The qualitative results of a 2006 beneficiary assessment (see annex 5) also showed how well the project was rooted in the communities. 3.3 Efficiency 48. Many of the sub-projects under the civil society component of the project aimed at alleviating the difficult conditions of people living with HIV/AIDS, and showed how a small amount of money can go a long way to provide relief to needy individuals and communities. From a generally accepted strict public health perspective it would have been more efficient, at least in the short term, to have the majority of projects in areas with highest risk of transmission (hot spots) and to focus more attention on prevention, and on groups most at risk. This occurred but only in later years of the project, when the national program improved its targeting strategies and focused on specific groups (―the window of hope‖ for the HIV-negative youth). However, some CSO projects which had started with impact mitigation interventions later included prevention measures. Indeed, the civil society projects which focused on impact mitigation 18 According to the project documents, the end of project target for this indicator was 51 percent. However, the target value appears to have been incorrectly stated as it confuses percentage with percentage points. The correct target (i.e. a 20 percent increase over 2002 levels) implies an target of 37-38 percent which was indeed achieved for men. 12 enabled a discussion in communities on HIV transmission and related risks, and therefore often could provide the platform for prevention efforts with broad community ownership. 49. Under the RRF a strategic approach, delivering high impact proven specific prevention interventions was gradually scaled up with a specific focus on larger organizations with greater potential which fostered efficiency. The private sector component of the project was successful but ran only for a two-year period, diminishing its potential efficiency. Efficiency of the public sector component may have been diluted by the large number of ministries that were involved. When the RRF was established, it was decided that only priority ministries could equally apply for ―sub-projects‖. This was especially important for the newly established Ministry of Public Service. 50. The project provided urgently needed resources to address the dramatic impact of HIV on the health sector. It provided substantial resources for staff training for both preventive and palliative care. The training programs were well implemented and largely according to plan. The MOH was also very efficient in mainstreaming the HIV/AIDS interventions. As a result, VCT centers were set up in much larger numbers than originally planned, as the services were integrated in the general health services. Similarly, the PPTCT program grew quicker than anticipated, although further improvements are needed as coverage of ante-natal care services is still too low and characterized by high drop-out rates. The sharply declining prices for ARVs during the project‘s life made it possible that about 28 percent (UNGASS, 2008) of patients who needed ART could receive treatment, thus positively impacting on the efficiency of the project although it financed few ARV itself. Integration of HIV and TB services during the latter half of the project led to further gains in efficiency in fighting both TB and HIV. A reallocation of Grant funds towards MOH during the 2007 and 2009 restructurings made it possible that the Grant was almost fully disbursed. 51. In conclusion, the project‘s activities were largely appropriate, as was the allocation of funds, but efficiency of the project could have been further enhanced if: (a) delays had been reduced, (b) the public sector interventions had concentrated on a limited number of ministries and institutions with the largest potential impact on slowing down the epidemic, and (c) the private sector initiatives had been implemented for more than just two years. Therefore efficiency is rated only as modest. 3.4 Justification of Overall Outcome Rating Area Rating Relevance High Efficacy Substantial Efficiency Modest Overall Moderately satisfactory 52. The project‘s objectives were highly relevant to the country‘s development and two out of three PDO targets have been achieved while the target for the third is within reach. The results for several intermediate indicators have exceeded the targets. All these results can be attributed to the PEN, within which framework the national response is being implemented and to which the Bank support has contributed in a critical manner as its first large external financier; HARP helped to craft the national structures and implementation arrangements. The overall rating for the project could be satisfactory. However, given moderate shortcomings in efficacy and the modest rating for efficiency, due to delays and maybe the project‘s too broad of a focus, the overall rating is moderately satisfactory. 13 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 53. It is generally recognized that activities under component 1 made a considerable impact on the alleviation of poverty, as many sub-projects dealt with mitigating the effects of the epidemic, including income generating projects. Social impacts are also considerable, as for many communities, the sub-projects represented the very first time that communities organized themselves around a certain theme and had to take decisions together. Moreover, in the context of HARP several impact mitigation projects were later developed to include prevention efforts. Beneficiaries also recognized that the success of the VCT or ART services delivered by the MOH was mostly due to the awareness creation and stigma reduction through the community projects. 54. Gender aspects were well addressed for some (sub) components. For example, A grandmother lives with her eight orphaned over 20 percent of the civil society sub- grandchildren close to a truck stop. She lets projects dealt with orphans. The care for her 12 year old granddaughter work there as these vulnerable children is a task which a prostitute. When the health worker rests mainly on the shoulders of women. In explains her the dangers of HIV, the the ―window of hope campaign‖ the needs grandmother responds: ―There is no of adolescents were given a gender-specific alternative. We may die of hunger or we may focus. The income generating projects for die of AIDS. It‘s the same thing. ― women also contributed to female empowerment in the communities. However, the special needs for men having sex with men were never considered under the project. 55. Male circumcision, since a few years well known to have a highly protective effect on HIV transmission, is traditionally common in the North and coastal areas of Mozambique, but no concerted effort has taken place by Government to take the intervention to scale at national level. So far the RRF supports two sub-projects. A campaign to promote circumcision has just started under the title ―Tira chapeu‖ (take off your hat). (b) Institutional Change/Strengthening 56. The institutional development impact of this project and its long-term impact on institutional capacity are substantial for the CNCS, but also for other organizations. The project was very helpful in establishing sound fiduciary processes in the Council and to set-up of planning and programming functions, and M&E systems, all of which are now beneficial to national and international partners. CNCS‘ staff carried out numerous training sessions and capacity building efforts in project management for the NPCS, CSOs, NGOs, private sector, and public sector agencies. (c) Other Unintended Outcomes and Impacts (positive or negative) 57. Probably the most important and very positive unintended outcome of the project was capacity building at community level. In many places, the project presented the first occasion that funds were channeled directly to communities, thereby causing them to organize themselves. This very grass roots level capacity building may not always have been the most efficient way of dealing with the AIDS epidemic, but was groundbreaking and meant an enormous step forward in community organization. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 58. During the ICR mission various groups of PLHIV who benefitted from the sub -projects were interviewed. People were tremendously grateful, especially for livelihood support provided. 14 Several noted that it was easier to listen to IEC messages regarding HIV transmission if the activities were part of a comprehensive set of services that went beyond the narrowly defined window of HIV only. As the number of very small projects had been drastically reduced over the last year when UNDP took over the management of sub-projects, many people showed great concern with the phasing out of their specific project. 59. Health workers interviewed showed a very positive impact of the bio-security support provided by the project, through which they received training and supplies, such as gloves, needles, syringes, condoms, but also larger equipment for incineration, blood transfusion equipment, and vehicles. 4. Assessment of Risk to Development Outcome 60. The risk to development outcome is rated substantial. The structures which were set up by GOM to fight the epidemic are now well established and there is considerable continued support for HIV/AIDS from the donor community. 19 However, due to the present international financial crisis the development assistance for AIDS from bilateral and multilateral partners may be declining, while the needs for resources, especially for ART are increasing. Mozambique is heavily dependent on external aid. The Bank‘s continued support for the health sector‘s systems will somewhat mitigate this risk. Health systems strengthening efforts will be needed to provide ARV for the masses in need of treatment and to continue prevention efforts and scale up efforts to link especially TB and HIV programs. 61. The Government‘s continued commitment to fight HIV will somewhat mitigate the development risk by maintaining HIV as one of the priority areas in the subsequent phases of its poverty reduction strategy. The battle has been led from the beginning by politically high profile figures. The previous and present president both provided strong leadership, speaking out on many occasions on the importance of fighting HIV. 62. Importantly the epidemic seems to be contained in the center and northern parts of the country, where HIV prevalence rates have stabilized or even moderately declined over the last few years. In the South this is not yet the case, and the sub-region warrants additional and focused interventions to further reduce the spread of the epidemic. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: moderately satisfactory 63. Recognizing the multi-sectoral character of the operation, the Bank‘s first three team leaders for the project were not health specialists. The preparation team was truly multi-sector and included staff from the education, health, agriculture, and rural development sectors. The preparation team was well organized and one of the first in the Bank to set up a web site http://afr/aftr1/mozambiqueaids/ making all relevant documents easily available to all team members and stakeholders. 19 So far, Mozambique has received US$244,980,748 in approved funding for HIV/AIDS from GFATM, according to the official website of GFATM. Through PEPFAR, Mozambique received $835.6 million to support comprehensive HIV/AIDS prevention, treatment and care programs from FY 2004 to FY 2009. 15 64. The fact that the design of the project, even after five amendments to the DGA and three National Strategic Plan (PEN I, II and III), remained basically the same, is testimony to its robustness and its good fit with the national structures. At the same time, the MAP instrument with its learning by doing approach proved to be highly flexible and allow modifications to the country context. 65. The operation was prepared on a fast-track basis, and as a result some unresolved actions which were pushed forward to implementation. The numerous conditions of effectiveness and the delay in effectiveness are an illustration of this. 66. The Bank preparation team may have been too optimistic in its institutional assessment of the CNCS. Not only did it overestimate the capacity to perform the management of the sub- projects, but these responsibilities also took away from its major responsibility to monitor, coordinate and guide the national response – a role for which it also was not yet fully equipped. 67. The project was selected for the Bank‘s Sixth Quality at Entry Assessment (2003). The review took place in the period between Board Approval and project Effectiveness. The panel gave the project generally high ratings. It liked especially the good fit of the national and MAP approach and the strategic relevance of the operation. However, it was concerned about the implementation readiness, as the planned pretesting of fiduciary processes had not proceeded as planned, and about possible conflicts between CNCS and participating ministries, which would feel disempowered. The number of conditions for effectiveness and readiness were concerns which in fact materialized. (b) Quality of Supervision Rating: moderately satisfactory 68. Aide-mémoire often started with a review of the status of the epidemic, thereby keeping a strategic focus on the main drivers of the epidemic and the eventual development impact. 69. The various restructurings and amendments to the FA are evidence of responsiveness of the Bank team to address implementation issues. Allowing for the closing date to be extended twice made it possible that most of the project targets were achieved. In order to get or keep the project out of problem status, intense and frequent supervision efforts took place. This was possible since experienced FM staff and the TTL were field based and headquarter staff travelled frequently to Mozambique. ACTAfrica also supported the task team and the implementers on various occasions. 70. But there were also delays in finalizing several arrangements. After the Bank signed on March 4, 2006, a Memorandum of Understanding with Government and the Common Fund Partners (Canada, DFID, Denmark, GFATM, Ireland and Sweden) it still took almost 16 months for it to process the agreement internally. After the approval of the fifth restructuring package, it still took several months for the Bank to approve the sole-source contract with UNFPA. 71. M&E did receive extra attention during supervision and the Bank, through its Global AIDS Monitoring and Evaluation Team (GAMET), provided considerable assistance to the program. Several missions were joined by an M&E expert and the Bank held many conceptual and strategic discussions with CNCS on capacity for M&E. However, reporting was not well done, and only the last supervision report contained a table with the relevant indicators. 72. From the very first Implementation Status and Results Report the formulation of the PDO was raised as an issue, as the team realized that it would take more than the 5-year project cycle to change behavior and consequently slow down the spread of HIV/AIDS. However, when the PDO was reformulated in 2006, the change was modest (i.e. focusing on the Bank‘s contribution 16 to the slowing down of the spread of HIV) but did not shift the focus towards measurable outputs (e.g. coverage and utilization of core HIV services and interventions). Fiduciary: 73. During implementation financial management arrangements and compliance were well supervised. Without exception, missions were accompanied by a WB FM specialist. The issues raised in previous missions and audits were followed-up systematically. While there were numerous problems (e.g. need for extra efforts to improve accounting systems, providing supporting documentation by implementing agencies, formulating the budget and execution of component 4 of the project, and managing the designated accounts) they were followed up rigorously. World Bank FM specialists put in considerable extra effort in training staff in the CNCS and MOH to strengthen procedures and internal controls. When the Bank joined that the pooling arrangements with other development partners for the Common Fund (CF) and RRF additional FM design efforts were needed to make the Bank a participant in pooling arrangements. After the Bank joined these pooling arrangements partners soon started relying on the Bank for fiduciary oversight. 74. Procurement for such an operation with multiple implementers and large scale civil society involvement was challenging, both on the Mozambican and the Bank‘s side. The Bank‘s guidelines had a steep learning curve. The institutional procedures and the urgency of taking programmatic action were not always in line, and there were numerous implementation delays. 75. While it is commendable that during preparation a comprehensive Environmental Plan was developed to deal with the public sector, beyond what would be strictly needed for an HIV/AIDS project, a truly effective plan would also need to look at handling of medical waste in the private sector clinics. Moreover, the ambitious plan was not systematically implemented. 76. During implementation monitoring of environmental arrangements and compliance was virtually nonexistent as not a single mission was joined by an environmental specialist, and none of the supervision reports paid any significant attention to these aspects. At the same time compliance with the environmental requirements was consistently rated as ―satisfactory‖. (c) Justification of Rating for Overall Bank Performance 77. All five Bank task team leaders were familiar with Mozambique and were fluent in Portuguese. Five amendments and two extensions to achieve the PDO demonstrated the perseverance and extra effort made by the task team. GAMET support to strengthen M&E for the national program showed that the Bank‘s team went the extra mile. The appreciation of other DP s for the Bank‘s fiduciary oversight also speaks volumes about the design and the actual implementation of the arrangements. The in country presence of the TTL for most of the project‘s life made support to the government counterpart team possible in a much more intense format than would have been the case from headquarters. All these considerations would lead to a fully satisfactory rating. However, since: (a) there was not much apparent focus during supervision on the results framework or (b) safeguards issues, and (c) delays on the Bank‘s side in processing of some restructurings, the rating is lowered to moderately satisfactory. 5.2 Borrower Performance (a) Government Performance Rating: moderately satisfactory 78. Political commitment to fight HIV has been high in Mozambique from the start of the project. In public speeches Presidents often spoke about the epidemic and ways to fight it. A national multi-sectoral council was set up early on. The National AIDS Council under chairmanship of the prime minister met regularly to discuss the general strategy and follow up on 17 international commitments. HIV is a constant theme in Government‘s subsequent poverty reduction strategies. 79. From the start of the project, the CNCS was overwhelmed by its many duties. The task of management of the CSO sub-projects was huge and took away from its main duty of coordinating the national response. Donor attention for HIV/AIDS was rapidly increasing and the Council had to attend to many missions and carry out the necessary preparation work in order to receive large amounts of funding from other DPs. CNCS often suffered from high staff turnover as a result of which important positions remained vacant for a considerable time. Especially towards the end of the project when Component 1 was managed by the UNDP the provincial nuclei of CNCS would sometimes hardly monitor what interventions were supported by the RRF, let alone the projects supported directly by development partners. 80. MOH performed relatively well, although there were some FM issues. The health sector component achieved well beyond its targets. Thanks to the use of UN agencies towards the end of the project, major procurements for the MOH could be largely completed on time. The Ministry took advantage of other support available to it and as a result targets for VCT and ART were achieved. 81. The performance of other ministries and public institutions varied greatly. The lack of true mainstreaming in most ministries beyond health shows that the multisectoral character of HIV has not taken hold in the mindset of most ministries. (b) Implementing Agency or Agencies Performance Rating: satisfactory. 82. CSOs and NGOs performed well during the project. The organizations may have been very small sometimes, with little capacity, but they were highly motivated and eager to learn how to comply with HARP‘s sub-project requirements. The capacity building efforts by the CNCS were well appreciated. The resilience of these organizations implied that they continued to implement the sub-projects even when confronted with challenging administrative procedures. 83. The agency hired to implement the private sector component also performed well. In total 557 vs. a goal of 250 sub-projects were contracted, which targeted workers in general, workers living with HIV, their family members, the neighborhood of the firm, and orphans of deceased workers. According to the evaluation, the large majority of the projects made a significant impact on workers knowledge, safe sexual behavior and reduction of stigma. These awareness raising activities also contributed to HIV testing seeking behavior. With the benefit of hindsight more efforts could have been made to involve trade unions and local authorities. 84. After the fifth amendment two UN agencies implemented important sub-components of the project which enhanced performance and accelerated implementation. UNDP was contracted by MOH to manage the contracting of CSO and NGO to implement sub-projects in civil society, which became increasingly focused on both prevention and mitigation. Public agencies and firms were also allowed to submit proposals for funding. UNFPA was contracted by the MOH for a very large procurement (US$9.4 million) of medical and reproductive health equipment and supplies–almost all delivered on time. (c) Justification of Rating for Overall Borrower Performance 85. The overall borrower performance is rated as moderately satisfactory reflecting the enormous efforts on the side on CNCS, private sector, some Ministries and especially the small CSOs to fight HIV in accordance with their capacity and responsibility, in spite of shortcomings and implementation delays. 18 6. Lessons Learned 86. The main lessons learned from this project can be summarized as follows:  It makes sense for the Bank to develop framework operations such as the MAP. The flexibility of the MAP instrument and the “learn by doing approach� proved to be effective in the Mozambique context. The Bank responded swiftly and appropriately to addressing the HIV/AIDS situation in Mozambique when there were few development partners on the ground and limited financing. The fact that the basic design of the project could be maintained through five amendments to the DGA proves that the MAP instrument was used in a flexible way and could respond to the country‘s evolving needs as the national strategy developed and new partners came on board.  The project development objective needs to state clearly for what an individual project can be held accountable even if it is part of a series in Adaptable Program Lending (APL). While the MAP was originally designed as a 12-15 year APL, in practice HARP ended up as a self-standing operation and now needs to be assessed in that light. The QAG (2003) panel in its assessment warned against setting HARP‘s (and MAP‘s in general) expectations too high, when it declared: ―While this well crafted and broadly supported ―process‖ operation is likely to achieve its important institutional development/capacity building objectives, achieving the stated Human Development outcomes will require critical behavioral changes and sets of interventions that are outside the main scope of this operation and for which other actors are responsible. These outcomes may or may not be achieved, but success or failure cannot and should not be ascribed to this single operation.‖ Fortunately HARP instigated behavioral changes, as the results for the PDO indicators show, but one may wonder if one hereby holds the project to a fair standard.  Pooling of Bank resources with resources of other partners has multiple benefits. HARP was truly unique in that it involved pooled financing with other donors through the Common Fund and it also used a joint financing modality (RRF) to channel funds to NGOs and CSOs. Such pooling of resources reduces transaction costs for Government. It also enhances the prospects for sustainability. The Bank‘s participation in the pooled funding mechanism enhanced the government‘s ability to mobilize substantial donor resources. Bank participation in these joint funding mechanisms implies that even after a project closes many of these activities can continue to be funded although probably not always at the same level. Finally, the Bank can play a critical role in bolstering fiduciary systems. In the case of HARP, the Bank assisted the government to put in place fiduciary systems and procedures which contributed to better management of the national HIV response. This support was well appreciated by other development partners who subsequently used these reinforced systems to channel their own financial support. When the Bank entered the RRF other DP soon appreciated much the fiduciary responsibilities which the Bank took for the management of the fund, a role which it continues to play after project closing.  Projects need to be well tailored to institutional capacities and training needs to be timely and extensive if a discrepancy exists between the projects demands and the implementer’s capacity. The fact that the HARP was often in problem status during the implementation phase reflects a combination of factors, including lack of readiness, difficulties in mastering IDA procedures, but above all insufficient capacity building up- front and during implementation. Especially the implementation of the civil society component was hampered by difficulties of smaller organizations to adhere to Bank 19 procurement and financial management procedures. It is hard to underestimate the training needs of civil society in projects as HARP.  Governments need to focus more attention and resources on understanding their HIV epidemics and adopting more targeted approaches. The Bank and other development partners assisted Mozambique to build M&E capacity and conduct surveys which enhanced understanding of the drivers of the epidemic, and allowed for prioritizing interventions in high transmission areas and high risk groups. HARP shows that holding strategic technical discussions during supervision missions and provision of Bank assistance for M&E may lead to a more evidence-based national response.  While a strictly technical public health approach would direct resources from the start towards targeting interventions to geographic areas and population groups with the highest HIV prevalence, that direction may from a longer term perspective not always be the right one. Money came to the Mozambican communities because of AIDS. As a result, communities organized themselves. They were empowered to begin to take control of their lives. The income generating activities assisted vulnerable groups to better cope with their vulnerability, and demonstrated that a small amount of money could go a long way. They focused on the enormous and immediate needs of orphans and vulnerable children. These impact mitigation efforts opened the door for an open dialogue in the communities on the risks for HIV infection and laid the foundation for a community-based approach to prevention. Active involvement of civil society means that one must be prepared to finance its priorities, even if these are not immediately and evidently perceived as the most cost-effective to curb the epidemic. Nevertheless, more should probably have been done in capacity building to make these investments results- oriented and to adopt from the start a more integrated approach, providing a comprehensive package of prevention, mitigation and care and treatment to communities.  HIV/AIDS financing can also be used for broader health systems strengthening. As Mozambique was part of MAP II, there was a concerted effort to incorporate this lesson of the first series of MAP project into the project design. The MOH was one of the key beneficiaries of the operation and this helped to further bolster its capacity and avoid the kind of problems which occurred in other countries where national HIV/AIDS councils undermined health ministries. HARP funded a wide range of activities which bolstered overall capacities, including undergraduate training (e.g. nurses, laboratory technicians, and radiologists), procurement of pharmaceuticals, and technical assistance to strengthen planning. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies Comments Government of Mozambique Action taken We would first of all like to express our gratitude for the Accepted with gratitude. quality of the evaluation undertaken. Indeed, the evaluation reflects a thorough analysis of the Project accomplishments over the period of the Project implementation. It also raises critical issues in a deep and comprehensive understanding of the course of HIV and Aids epidemic in Mozambique, the national response as well as the general international trends and approaches. Overall the evaluation is balanced as it indicates areas of positive outcomes as well as those of poor 20 accomplishment and does also address what failed to be considered in order for the response to be different. Delays in replenishment of accounts had also to do with Text adjusted lack of absorption capacity from the implementing agencies, as well as delays in replenishment of CNCS‘s accounts by the World Bank. Under point 19, the evaluation does not mention KPMG KPMG and E&Y are now mentioned and Ernest and Young contracted in by CNCS to not by name but in generic terms. strengthen project planning and management capacity of both CNCS and the implementing organizations. Under We continue to mention UNAIDS the same point, the mention to UNAIDS as an and other partners. organization that has given technical assistance to CNCS is, somehow, surprising. A mention to one partner leaving out others is, indeed, unfair. In the same point, the evaluation has found reluctance of Reference to CNCS reluctance is CNCS in contracting a GMA. This, from our point of taken out of the paragraph. The view, is just a simple judgment with no fundaments. withdrawal of GMA was a business Some of the issues proposed in the Inception Report, dispute with CNCS. GOM policy taking them on board, would change the scope of the was that contracting of sub-projects ToR for contracting a GMA with all implications for an should not be done by CNCS. international competitive bidding that should be transparent and give equal consideration and treatment to all competing agencies from the onset of the contract throughout its implementation up to the end. Under point 20, it should be clarified that there was no Text adjusted to reflect this spirit of any kind of agreement for provision to Ministerial Focal mainstreaming. Points of additional payment, beyond their monthly salaries, due to their involvement in HIV and AIDS coordination in the Ministries. The idea of mainstreaming HIV in the sectoral plans was put in place in order to create capacity and sustain it for years to come. The point 21 does not report fairly the facts. One must UNDP continues as contract manager separate the treatment of the Health Sector from that of beyond the closing of HARP with the contracting of UNDP to act transitorily as grant funding from other partners in the making agency. The contracting of UNDP and the pool. NGO projects are not yet procurement of medicines and other commodities (by concluded. Bank was part of this UNFPA) deserve a more critical assessment. By the time pool and polo continues to exist. the evaluation UNDP had less than 30% disbursements to UNFPA contract was largely project implementing agencies. implemented. The evaluation seems to praise the way UNDP has Text adjusted to reflect this. entered into the scene and the strategies adopted to finance the Civil Society organizations. However, it fails to mention that CNCS‘s experience was crucial in the effort. The evaluation of the Private Sector outsourcing needs to We stand by our views expressed in be deepened. The project evaluation seems to give a the ICR. positive impression of the performance of the component, which it is not the position CNCS shares. The point made by the assessment on point 71 was a The failure of the RRI is common ground throughout the implementation of the reintroduced in the text. 21 project. One should add contradicting positions coming from the World Bank Project Team (management and procurement from the WB) which in most of the cases delayed project implementation. An example is the 100 days Rapid Results Initiative (RRI). (b) Cofinanciers HARP did not have co-financiers. (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) See Annex 8. 22 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Estimate (USD Components (USD millions) Appraisal millions) Community and Civil Society 28 26.6 95% Initiatives Capacity Building for Civil 5.5 3.8 69% Society HIV/AIDS Government Multi-sectoral 7 5.1 72% Response Strengthening and Scaling Up 17.5 19.5 111% Health Sector HIV/AIDS Institutional Development for 6 4.2 70% Program Management Total 55.00 59.2 107% (b) Financing Appraisal Actual/Latest Type of Estimate Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 9.00 N/A N/A International Development 55.00 59.2 107% Association (IDA) 23 Annex 2. Outputs by Component Component 1: Community and Civil Society Initiatives (US$ 28 million). This component would empower communities to respond effectively to the HIV/AIDS epidemic. A central activity would be mobilizing communities, promoting local initiatives and strengthening the capacity of local actors. A Community and Civil Society Facility held by CNCS would be created to finance appropriate HIV/AIDS related activities carried out by eligible applicants, such as community and faith based organizations, NGOs, the private sector, associations and other or ganizations. The management of the Facility would be decentralized to the provincial level. Under component 1, the main component of HARP, CNCS contracted CSOs for sub-projects, which could vary from as little as less than US$1,000 to over US$20,000 (Category A: below US$1000, Category B: between US$1000 and 20,000, and Category C: over US$20,000). In total HARP supported 3771 sub-projects (Category A 179 sub-projects, Category B 776, and Category C 2816 sub-projects). The concept of sub-project was new and in the beginning the Mozambique Tribunal Administrativo was involved for approval of each, as if it concerned a World Bank supported project, causing undue delays. CNCS was overwhelmed by the hundreds of sub- projects, which caused further delays in sub-project approvals, monitoring and then in payments to implementing agencies. Besides this, much capacity building needed to be done on the side of small emerging NGOs applying for grants but which had little or no experience in project or financial management. Often the reporting requirements were heavy for CSOs which didn‘t even have a computer. Some sub-project fiduciary requirements seemed inappropriate for the Mozambican rural reality where e.g., a banking system was inexistent and three quotations impossible to obtain. The 20 percent contribution requested from the NGO‘s own resources also caused difficulties for many. Given the immediate needs of communities, the high level of poverty, and the demand-driven character of the sub-projects, many focused on mitigation of the impact of HIV infection and thus turned into income generating projects, limiting the focus on HIV prevention. There was some targeting on geographic areas with a high risk of HIV transmission such as border areas or truck stops, the so called ―hotspots‖. Number of sub-projects approved by CNCS under HARP Province Category 2004 2005 2006 2007 2008 Total A 2 2 1 2 7 B 9 25 3 18 55 1 Cabo Delgado C 78 46 6 46 176 A 6 4 2 4 16 B 21 16 4 15 56 2 Nampula C 184 31 7 35 257 A 2 3 2 2 9 B 7 10 7 7 31 3 Niasa C 62 53 11 14 140 A 3 10 4 6 8 28 B 35 16 35 75 161 4 Zambezia C 289 94 70 78 531 A 2 4 6 5 3 18 B 15 21 22 14 72 5 Tete C 98 81 42 39 260 A 3 0 5 0 8 6 Manica B 17 7 19 23 66 24 C 106 75 36 24 241 A 9 3 3 2 17 B 33 15 2 12 62 7 Sofala C 223 43 3 34 303 A 1 3 6 7 4 20 B 9 37 27 10 83 8 Inhambane C 98 101 46 23 268 A 4 9 5 5 23 B 15 21 17 9 62 9 Gaza C 92 38 29 31 190 A 4 4 6 2 2 14 B 3 15 37 7 10 69 10 Maputo Cidade C 95 53 14 26 188 A 4 7 3 5 19 B 13 24 4 18 59 11 Maputo Provincia C 94 109 14 45 262 National total 13 1659 1003 466 643 3771 In total 179 projects in Category A were approved. For category B and C the numbers were 776 and 2816 respectively. The table shows that the sub-projects were remarkably well distributed over the country. Their number peaked early on in 2005 after which they declined greatly, reflecting the capacity constraints at the center and the entry of other donor funding. The aim for national coverage was in line with the goal of the first phase of the MAP. From 2006 onwards Mozambique would receive large contributions from PEPFAR (US$70 million in 2006 and US$ 150 million in 2007) which transformed the national response. The support was given to larger, often international, NGOs with higher levels of capacity, thereby complementing but also sometimes sidelining CNCS-financed smaller local CSOs. The support also allowed MOH to dramatically increase its support to ART beyond PPTCT to include eligible PLHIV in general. CNCS encouraged small CSOs or NGOs to form joint ventures or liaise otherwise with larger NGOs. This would allow for economies of scale and smaller CSOs could spend more of their time on actual project implementation, and struggle less with project preparation and reporting of expenditures. The larger NGOs would receive project funding for capacity building of their smaller partners. After two years of discussions and a major restructuring of the project in 2007 a Grant Management Agency (GMA) was contracted, supported by various donors and the Bank (a so called Common Fund). It would take over the recruitment of CSO for civil society sub-projects, which had stretched CNCS‘ capacity. The hiring happened only after protracted negotiations, as CNCS leadership only reluctantly let this responsibility go. The difficult relationship between CNCS and GMA continued after the hiring and almost a year into the contract the GMA resigned, due to disagreement on the objective and the price of the contract, giving CNCS the opportunity to keep funding sub-projects of non-state actors until end of 2009. As a result managing sub- projects continued to take away from CNCS‘ main responsibility of coordinating the national response. Given the continued slow implementation progress, a consultation was held with the Prime Minister and as a result another major restructuring of the project took place by mid-2009. This restructuring (a) reallocated US$12 million to MOH for scaling up HIV and syphilis testing, expanding condom distribution, strengthening the links with reproductive health, and improving 25 bio-safety and nutrition. Procurement for related commodities would be done through the United Nations Population Fund (UNFPA); (b) established a new funding mechanism for non-state actors through the creation of a Rapid Results Fund (RRF) of US$3 million for prevention. The fund was supported by various donors and managed by UNDP which would call for proposals focusing on multiple concurrent sexual partnerships, reproductive health and HIV, TB and HIV, and male circumcision; (c) continued financing CNCS through the Common Fund; and (d) allocated US$0.7 million for evaluation of innovative programs, such as post-test counseling, call for proposals and male circumcision. For the RRF, UNDP was contracted by MOH, not by CNCS which had hired the GMA. When management of the sub-projects moved from CNCS to UNDP —and the funds moved from a Special Account administered by CNCS to an account with MOH— after the fifth amendment of the FA, the component became more strategic with greater focus on prevention but without increased focus on geographic areas with the highest HIV prevalence. UNDP would contract CSOs with more capacity and the average size of contracts for sub-projects would more than double. Smaller CSOs and NGOs could still participate through joint ventures with larger organizations but this did hardly happen. As coordinators, CNCS and its provincial nuclei (NPCS) would officially maintain a participatory role in the recruitment of the CSOs but in practice, especially at provincial level, the NPCS were little aware of the activities of UNDP- contracted agencies. Towards the end of the project MOH planned a nutrition intervention> A 6- month food basket would be provided to patients starting on ART through the World Food Program. The intervention never took place under the project as government took a long time to decide on the final target districts for the intervention. Government had requested a third project extension (in a letter dated June 17, 2011) for the implementation of just this contract. The Bank did not agree and instead it was decided that the intervention would be financed through another Bank operation, the HCSP. By project closing the following sub-projects had been contracted: Partner Budget # of Geographic Thematic area beneficiaries area World Food 903,000.00 3,500 National Antiretroviral Therapy, TB- Program HIV collaboration Ministry of 903,000.00 50,000 National Public Service TB-HIV collaboration, male circumcision, Reproductive Health-HIV collaboration ECOSIDA 449,832.00 7,500 Maputo, Gaza, TB-HIV collaboration, Sofala, Multiple concurrent Manica, Tete, partners Nampula PATHFINDER 27,832.00 30,000 Maputo, Gaza Reproductive Health-HIV collaboration Right To Play 304,814.25 Maputo City, TB-HIV collaboration, 25,000 Gaza, Reproductive Health-HIV Quelimane collaboration ACJM 42,385.20 12,000 Maputo City Reproductive Health-HIV collaboration CISLAMO 73,842.13 2,200 Nampula, Male circumcision Maputo City ADPP 90,137.00 20,000 Reproductive Health-HIV 26 Maputo collaboration Province CCM 192,645.00 3,000 Inhambane Reproductive Health-HIV collaboration, Multiple concurrent partners MATRAM 128,615.35 4,000 Maputo City, TB-HIV collaboration Gaza, Inhambane HANDICAP 283,250.00 15,000 Maputo Reproductive Health- HIV INTERN. Province, collaboration The program supported initiatives in private firms under the project Avante. One agency was hired to manage sub-projects in companies nation-wide. The companies ranged from multi- nationals with over 200 workers to small enterprises with less than 20 employees. HARP supported 557 sub-projects against a target of 250, thereby reaching an estimated 45,000 workers and family members. At the end of the two-year contract an evaluation, which reviewed 30 percent of 102 sub-projects, showed that most sub-projects dealt with information, education and communication (IEC), peer education, distribution of condoms, VCT. Little attention was given to legal support, evaluation, or care and treatment. The appreciation for the sub-projects was great amongst the workers, who for the large majority considered the support very important for them; the sub-projects mostly did not involve managers both in the design and implementation, nor spouses of workers or surrounding communities. VCT services were not fully utilized in this component because of stigma and fear of lack of confidentiality. Sub-projects often suffered from delays due to weak financial management. The two-year duration of the contract of Austral hampered the implementation of subcontracts towards the end of the period as the remaining implementation time became simply too short. Sub-projects tended to be more successful where the management of the company provided active support and allowed e.g. for activities during working hours. The evaluation could have led to a better designed second generation of sub- projects with the private sector, but such a second phase was never implemented as CNCS did not renew the contract with the originally hired firm or sign a new contract with another agency. However, CNCS continued to meet through ECoSida 20 with the private sector, proving sustainability beyond the project support. In 2010 it attended 88 meetings to mobilize managers and 21 firms started work place interventions. Component 2: Capacity Building for the Civil Society HIV/AIDS Response (US$5.5 million) would support efforts to create AIDS competent communities by improving skills and increasing implementation capacity of implementers, supporters, and target groups active under the Community and Civil Society Initiatives. Activities to be funded would include hiring and training ―Facilitating Agents‖ in each province, courses and activities to build awareness and capacity for leadership in the public sector, civil society and private sector, and the formulation and distribution of training modules and packages of IEC materials. The implementation of component 1 required much capacity building amongst CSOs and NGOs. Facilitating Agents were hired for this purpose and the implementation of component 1 started as a pilot in 3 provinces: Sofala, Manica and Tete. UNAIDS especially was helpful in providing capacity building to CSOs for proposal writing and sub-project implementation; it also provided 20 ECoSIDA is a national Mozambican private sector initiative in the fight against HIV/AIDS, mainly financed by the private sector itself. It has around 50 active members. 27 technical assistance to CNCS. Three United Nations Volunteers helped in the coordination of civil society initiatives The MOH had a role in the technical supervision of NGOs involved in the implementation of health activities. Technical assistance was hired to assist CNCS in the development of a standard contract in which it is set out how an NGO interested in carrying out health sector related activities with HARP financing had to behave: (a) in obeying the MOH rules and regulations – also with regards to medical waste management; and (b) in reporting relationships with both CNCS and MOH. Component 3: Government Multi-sector Response (US$ 7 million): Ministries and subordinate institutions would be supported in the development and implementation of HIV/AIDS related programs directed toward their own personnel (particularly high risk staff such as soldiers, police, extension agents, teachers, health workers) and their families, as well as programs directed towards their clients. The program would concentrate on non-health ministries and other key public sector institutions to effectively respond to the epidemic, with emphasis on prevention and care for infected and affected families. In each ministry a Focal Point would be appointed. With guidance from provincial Government and Ministry of State Administration, District Administrators would be engaged to foster development of District HIV/AIDS plans. Under the Government Multisector Response component focal points were selected in 18 Ministries and also public or para-statal organizations such as the National Institute of Health, or Linhas Aéreas de Moçambique, the national airline, were financed for a truly multisectoral public response. Some claim that too many agencies were involved, thereby diluting the efforts. In total 38 agencies were supported. Disbursements were often far below expectations, showing lack of commitment of their leadership. Focal points in the agencies, at central and provincial level, were part-timers and did not receive any additional payment for their HIV-related work. CNCS provided training at central and provincial level; it considered the hiring of extra technical assistance to jumpstart some of the programs but this never happened. Often the component did not result in activities beyond AIDS in the workplace activities or handling of personnel issues. It did not consider the strategic role a ministry could play in the national response; exceptions being the incorporation of HIV/AIDS in the curriculum by the Ministry of Education (with support from other development partners), or the development of anti-discriminatory legislation by the Ministry of Justice –Lei 5/2002. As Government is the largest employer in the country, and civil servants are amongst the most highly educated citizens, workplace activities were certainly justified to safeguard the human capital. CNCS could have brought the better performing ministries, such as the Ministry of the Interior or Defense, together with poorer performing ones to improve implementation. The Ministry of Women and Coordination of Social Action was one of the largest grant recipients of the public sector component. The Ministry‘s HIV/AIDS programs included support for orphans and vulnerable children and income generating projects for women affected by HIV. A new opportunity arose with the creation in 2007 of a new Ministry of Civil Service, which also received support from a Bank financed Public Sector Reform Project. An HIV/AIDS strategy for civil servants was developed and the Ministry received US$ 1 million to finance prevention activities in 12 priority ministries and public institutions. At project closing, efforts were underway to mainstream these activities in the agencies‘ regular annual budgets and work plans. Ministries would now receive support from the Civil Service Ministry rather than from CNCS. Examples of specific sectoral activities or initiatives supported under this component are:  For Higher Education: creation of a HIV/AIDS Group for students of the Eduardo Mondlane University; short term training for some members of the group; production of a biannual 28 newsletter; support for training by this group of local training institutions in the provinces to enable them to set up their own HIV/AIDS Groups; awareness workshops; set up of a local clinic (mainly for VCT services); supply of equipment and IEC materials.  The Ministry of Education received project support for a program called ―Direct Support to Schools‖, which helped decentralization efforts so that schools could plan themselves in planning of HIV/AIDS activities. The Ministry received a lot of support from other agencies as well, including UNICEF, UNFPA and DANIDA.  Ministry of Agriculture: thanks to mainstreaming this ministry had its own funding to finance its HIV efforts, which included an economic impact assessment, a study into ―Adult mortality in rural households in Mozambique and Implications for Policy‖. CNCS helped in training of educators and the development of strategic plans for the sector. The Ministry used its role in food production to encourage diversification of crops and diets aimed at boosting the immune system. Programs to build home farms and school gardens were carried out. The large PROAGRI program included HIV/AIDS activities in the work program of extension workers.  Ministry of Transport and Communication incorporated HIV/AIDS activities for construction workers as a requirement in contracts for its infrastructure projects. Such activities included theatre, workshops, counseling, and condom distribution. Originally the component was not included under the grants component of the DGA, and its activities could not be considered as sub-projects. This meant that regular Bank procurement and FM requirements would apply. CNCS procedures sometimes also formed a barrier since it only approved quarterly plans and only disbursed per activity, thereby making it impossible to apply for any interventions other than isolated events or equipment. The laborious procurement process brought unnecessary centralization and was responsible for the majority of disbursement delay, although other factors also played a role, such as the lack of capacity, the 20 percent requirement in co-financing, and the part-time character of the focal points in each ministry (except for Education). After the 4th restructuring the public sector interventions of component 3 would also be considered as sub-projects and could be funded just as interventions under component 1. Component 4: Strengthening and Scaling Up Health Sector Services for HIV/AIDS (US$17.5 million) aimed to scale up the response of the health sector to the HIV/AIDS epidemic and to provide technical leadership on treatment and care for people living with HIV/AIDS. The component would support (i) strengthening of the Integrated Health Network providing voluntary counseling and testing and HIV/AIDS related services in four central provinces; (ii) increasing the supply of HIV related drugs and supplies, condoms, drugs for opportunistic infections, anti- retroviral for PPTCT of HIV and post-exposure prophylaxis; (iii) strengthening clinical laboratory capacity for CD4 monitoring and diagnosis of opportunistic infections; (iv) measures to enhance bio-security; (v) measures to enhance blood safety; (vi) training and anti-retroviral drugs for post-exposure prophylaxis for health workers; (vii) improved management of opportunistic infections; and (viii) monitoring and evaluation of the component and the overall progress of the epidemic. For the fourth component MOH was the implementing agency. The component was implemented relatively well --especially during the first half of the project-- as the ministry had experience with implementing Bank projects and used a project unit (GACOPI), which also manage d the TAP. However, with changes in MOH leadership and resulting shift in responsibilities from GACOPI to various departments in the ministry, a lack of coordination resulted in disjointed activities and delays, as the regular departments were not acquainted with donor requirements. The program focused on bio-security. 29 Human resources development for MOH received much attention. First a training of trainers course was developed, together with didactic materials, before training programs for health workers in safe practices could be implemented largely as scheduled. The training programs were implemented with only minor delays. MOH had the facilities and the experience of contracting these institutes for project-financed training and it was able to pre-finance some of the training if project funding was not immediately forthcoming. The courses included training for doctors, nurses, laboratory staff, dentistry, etc. The MOH also had a role in the technical supervision of NGOs involved in the implementation of health activities (see component 2). Standards were developed in five areas: VCT, home-based care, day hospitals, treatment of sexually transmitted infections, and prevention of parent to child transmission. Blood banks received training in bio-safety. Materials for IEC campaigns were developed and campaigns for blood donations took place in cooperation with schools and churches. MOH performed especially well in the set-up of VCT. Originally it was foreseen that separate VCT services would be set up, but by mainstreaming these services the number of locations could be dramatically expanded to 359, much beyond the original target of 100. Provision of ART was expanded much beyond the original target of PPTCT because of reasons mentioned above and ART is now available in all districts. Minor civil works for the rehabilitation of VCT centers, laboratories, blood banks. Given the general status of much of the health infrastructure demand for this sub-component far outstretched resources. This and the need to make sure that the rehabilitations would benefit the health services in general and not just HIV services, required very careful planning. Most of the works were related to the proper installation of equipment. Procurement of goods, such as vehicles, laboratory equipment, medical supplies such as disposable syringes, test kits, reagents, and pharmaceuticals, condoms and medical and sterilization equipment was done with some delays and Bank staff provided extra procurement training. HARP disbursements declined in 2005-2007 as the Common Fund, supported by other donors, became available and its procedures were easier to follow for MOH than the Bank‘s. During the second half of the project implementation slowed down. In order to fully utilize the project resources still available to MOH, UNFPA would be contracted on sole-source basis for major procurement of medical supplies and goods, as was possible after the 2008 restructuring,. The dramatic price drop for ARV over the project‘s life from over US$10,000 to only US$100- 400 per year had a significant impact on the treatment policy of GOM and WB support to it. While in the beginning of the project the Bank would only finance treatment for opportunistic infections and ART only for PPTCT, towards the end first line ARV were also allowed to be procured for general purpose although this did not happen on any large scale, as other sources of funding were used for it. It did however fundamentally change the way civil society organizations could provide care to PLHIV. Component 5: Institutional Development for Program Management (US$6 million) supported the Executive Secretariat of the CNCS, strengthening its capacity to lead the country in the campaign. HARP provided much support to capacity building and staffing of CNCS in management, administration, FM, procurement, but also in technical areas such as communication and M&E. ACTAfrica supported such training by bringing MAP implementers together from various African countries. The training provided practical guidance and an opportunity to share 30 experiences in (a) implementation challenges, (b) procurement management, (c) FM, (d) administration and documentation, and (e) country action planning. The community-driven development approach of component 1 required extra fiduciary training. HARP financed staffing at central level and for all provincial nuclei of CNCS. Technical assistance for organizational development was hired to work with the provincial coordinators on the budgetary planning process, and focused on the demand-driven concept of funding civil society and community initiatives. CNCS and development partners provided also considerable support for monitoring efforts in the provinces. Technical assistance was provided to CNCS to develop a communications strategy for central level and also for the provinces. A Technical Working Group comprising communication specialists from 12 public sector institutions and a Communications Working Group of development partners provided assistance. Such strategy had been deemed so important that it had been made a legal covenant in the Grant Agreement. It included a range of methodologies, messages and materials. The Communications Strategy focused on awareness and knowledge of prevention, treatment and care for HIV, behavior change, recognition of vulnerability and stigma, and HIV and cultural norms. As a result of the capacity building efforts, CNCS was able to organize a lot of training programs in the various areas for implementers, journalists of radio, TV and written press, community leaders, religious leaders, public and private companies, youth groups, its provincial nuclei etc. M&E received much support from development partners. Besides M&E policy development the project financed M&E staff, equipment and operational costs at central and provincial level. Although planned from the start, the project was never able to recruit international TA for M&E. The CNCS officers would make regular supportive visits to the provinces. Three M&E working groups were established: (a) one led by CNCS to facilitate dialogue around strategic planning, monitoring, evaluation, research policy and technical issues; (b) one led by MOH to focus on epidemiology and health service statistics; and (c) a multi-sectoral working group including the M&E officers of most ministries, focusing on the multi-sectoral response. At provincial level M&E officers were hired to monitor especially the interventions of the CSOs and NGOs. Bank support was further used for the development of PEN II and III, to which the Bank‘s AIDS Strategy and Action Planning service (ASAP) also contributed. MOH‘ GACOPI –its staff being paid for by the project-- provided FM and procurement services for the all project components, as it was well-versed in World Bank procedures. During implementation the unit suffered major blows in its capacity when as a result of leadership changes at the top of the ministry contractual staff was replaced with regular civil servants without a good transition plan. GACOPI was handling many other donor projects in the health sector and therefore required a separate HARP coordinator. For the provincial directorates it was agreed that CNCS would do financial and procurement management, as they lacked the capacity as well to fulfill the DP requirements. 31 Annex 3. Economic and Financial Analysis A fresh economic analysis was not carried out for this core ICR. 32 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending James Coates Sr. Agricultural Economist TTL Institutional Development Kate Kuper Specialist Mary Mulusa Public Health Specialist Jean Rutabanzibwa-Ngaiza Social Scientist Noel Kulemeka Education Specialist LeadFiinancial Management Iraj Talai Specialist Slaheddine Ben-Halima Sr. Procurement Specialist Joao Tinga Financial Management Officer Procurement and Disbursement Isabel Nhassengo Assistant Christian Hurtado Consultant, Operations Costing Health Waste Management Mbaye Mbengue Faye Specialist Kees Kostermans Sr. Public Health Specialist Aniceto Bila Operations Officer Adelia Chebeia Task Assistant Community Development Rima Al-Azar Specialist Private Sector Development Elisabeth Ashbourne Specialist Manuel Sumbana Procurement Specialist Bachir Souhlal Lead Operations Advisor Joao Paulo Kastrup Legal Counsel Mohammad Ali Pate Public Health Specialist Institutional Development Katy Backes Specialist Maria Nita Dengo Agricultural Economist Supervision James Coates Sr. Agricultural Economist TTL Jacomina de Regt Sr. Social Development Specialist TTL Jean-Jaques De St Antoine Lead Operations Offcer AFTHE TTL Humberto Cossa Sr. Public Health Specialist AFTHE TTL Kate Kuper Urban Specialist 33 Caroline Forkin HIV/AIDS consultant Aniceto Bila Operations Officer AFTAR Manuel Sumbana Procurement Officer AFCS2 João Tinga FM Analyst AFTFM Isabel Nhassengo Procurement Assistant AFCS2 Subhash Dhingra Sr. Procurement Specialist TWIAF Bert Voetberg Lead Public Health Specialist SARHN Noel Kulemeka Sr. Economist Daniel de Sousa Sr. Agricultural Services Specialist AFTAR Sasha Clifton Intern AFTHE Slaheddine Ben-Halima Sr. Procurement Specialist MNAPR Antonio Chamuço Procurement Specialist AFTPC Bina Valaydon HIV/AIDS Consultant AFTHE Donald Whitson M&E Specialist, Consultant AFTHE Anne-Marie Bodo Pharmaceutical Consultant AFTHE Clarisse Nhabangue Team Assistant AFCS2 Jonathan Nyamukapa Sr. FM Specialist AFTPC Amos Malate Procurement Analyst AFTPC Luc Lapointe Sr. Procurement Specialist AFTPC Khovete Panguene Operations Analyst AFCS2 Elvis Langa FM Analyst AFTFM Maria Micaela Team Assistant AFCS2 Kees Kostermans Lead Public Health Specialist SARHN Main Author ICR (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) USD Thousands Stage of Project Cycle No. of staff weeks (including travel and consultant costs) Lending FY02 21.52 78,461.61 FY03 81.94 234,878.90 Total LEN: 103.46 313,340.51 Supervision/ICR FY03 31.10 74,355.53 FY04 29.96 94,548.62 FY05 31.01 72,659.99 FY06 38.19 110,588.30 FY07 33.17 87,046.38 FY08 30.16 73,676.83 FY09 43.83 113,784.00 FY10 21.08 56,105.62 34 FY11 11.58 39,788.94 FY12 6.11 39,856.77 Total SPN: 276.19 762410.98 Total LEN + SPN 379.65 1,075,751.49 35 Annex 5. Beneficiary Survey Results A special beneficiary survey was not carried out for this core ICR. During the ICR mission‘s interviews, a picture emerged in which beneficiaries were very grateful for the support, but also expressed major concern about the short duration and limited scope of many of the civil society projects. CNCS expressed its great appreciation for the groundbreaking support from the World Bank, which had been crucial in getting the national response going. In 2006, a study was carried out of the sub-projects with civil society organizations, titled ―Evaluation of the Efficacy of the sub-projects financed by CNCS under the National Response against HIV/ AIDS in the Republic of Mozambique‖. We summarize here the opinions of the beneficiaries of the sub-projects, as reflected in the report. For the study, 217 projects from categories between US$1,000 to over 20,000 per project had been selected from all provinces. By far the greatest focus of the projects was on mitigation, with prevention a distant second. A small number of projects focused on advocacy, stigma, treatment, research of community involvement. The beneficiaries were evaluated through group interviews. They had been selected by the members of the implementing agencies. The beneficiaries were a varied group in terms of gender, age, HIV status or affiliations with a religious group. In general the beneficiaries knew well what they had received from the project and were mostly satisfied and happy with the material support, as they had no work and the project gave them something to eat. Besides recognizing that the projects improved their lives, it was very often mentioned that more assistance was needed and that many people in need were not benefitting from this or a similar project. Beneficiaries were afraid that the project would be interrupted and asked for more projects to reach more people. Such requests were especially frequent for projects with children, as many districts have large numbers of orphans in need. Often the numbers of children in need were greater than the project could cover and the most vulnerable had to prioritized, such as double orphans. Exclusion of some children was not always well accepted by the communities. The majority of beneficiaries had gotten in touch with the project through the community leader. It was him who presented the project to the community and who indicated which people should participate in the project or which should receive a home visit. The leader‘s involvement showed how well the project was rooted in the community – an aspect which is very important for its sustainability. Questions about the needs of PLHIV provoked various reactions in the different groups of beneficiaries. In some groups people said that one should not know the status of beneficiaries but should only look at the needs of people. It was clearly difficult to discuss this theme in public. In other communities the beneficiaries were surprised as they had not realized that the project was linked to the national AIDS strategy. Again in other groups the discussion about HIV became very lively and people had many questions about treatment and the protective effect of condoms. In projects where PLHIV were involved there was a greater openness to discuss the problems of infected and affected people. The most frequently mentioned issues were poverty, lack of jobs, discrimination of PLHIV, lack of access to health services and insufficient nutrition for people on ART and treatment for opportunistic infections. 36 Several problems mentioned by the beneficiaries were linked to delays in disbursements which delayed the beginning of the projects or had a very negative impact in case of late arrival of critical elements such as seeds or school materials. Other issues were: the premature closure of projects because CNCS funding dried up; the lack of markets to sell the products produced by the projects; the lack of transport to bring patients to the hospital; the low number of projects in the communities with many needs; the quantity of products or supplies distributed was almost always insufficient. 37 Annex 6. Stakeholder Workshop Report and Results (if any) 38 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR The MAP21 Evaluation summary by CNCS, Implementer of the project 1. General context Mozambique was one of the African countries that benefited from World Bank funding under the Multi-country AIDS Program, to strengthen its national response against HIV/AIDS. The objective of the MAP project was to assist the country reducing the rate of new HIV infections and mitigating the adverse effects of infection, through prevention, care and treatment. The budget for this project was estimated at US$55.00 million, to be invested in five main areas:  Community and civil society initiatives - US$ 28.0 million;  Capacity building of civil society - US$ 5.5 million;  Public sectoral response (Ministries and Provincial Directorates) - US$ 7.0  Strengthening and expansion of the Health Sector response (managed by MOH) -US $ 17.5 million;  Institutional development for the implementation of the program - US$ 6.0 million. The design and negotiations for the Project took place between mid-2002 till early 2003. It was required that from the beginning, the minimal institutional capacity was in place to guarantee the funding. Therefore CNCS, as host organization, recruited qualified technical staff for important areas, such as financial management, procurement, planning and monitoring as well as communication. The 2003 functional analysis and the readjustment affected interventions, professional posts, profiles and functions. These changes were reflected in revisions and exemptions to some articles of Decree 10/2000 (March). These actions were much appreciated by the Bank, as they reassured it to finance the national response against HIV/AIDS. After WB financing was secured, the project approved and Grant Agreement signed, the CNCS worked diligently to meet the following Grant Conditions:  Finalization of a Operations Manual (including the components of monitoring and the Health sector, as well as the respective annexes); 21 In Mozambique HARP was known by the acronym MAP. 39  Establishment of a financial management system in line with the Operations Manual, as well as the design and execution of a training program in program ming and financial management for the pilot provinces - Manica, Tete and Sofala - to test and validate the system;  Filling of key positions, especially for procurement, monitoring and evaluation and fund management at provincial level;  Preparation and submission of a procurement plan covering all project activities, including contracting and procurement of goods and services;  Contracting of an independent auditor for the Project;  Opening of a Special Account for the MAP Project. After having satisfied the above conditions in a timely fashion, the Project was declared effective in August, 2003. 2. Challenges in the Implementation The implementation of the MAP experienced several challenges, some more critical than others, as described below:  The start-up of implementation, above all the scale-up to national coverage, would have to be assured by facilitating agents. Their hiring was a slow process, because the contracts needed to be reviewed by the Administrative Tribunal and because not all had the required skills (many overstated their potential on paper, but reality on the ground was different);  The prior review by the Administrative Tribunal formed often a bottleneck to the accelerated implementation, in line with the emergency character of the operation;  The banking network and financial systems in the country were not sufficiently developed to satisfy the project‘s requirements (financial transactions only through banks, i.e. checks and bank transfers);  Lack of experience and poor interpretation of bidding documents on the side of implementers and facilitating agents;  Misrepresentation by implementers;  Inexperience of some provincial nuclei in managing contracts and the demand for services that these generate;  Weak technical qualifications of the project staff both in civil society and in the Public Sector to deal with the strict requirements of the Map project;  Inappropriateness of the strict procurement requirements imposed by the World Bank, through the MAP, in view of the rudimentary state of the market for goods and services in the rural and peri-urban areas (strongly dominated by the informal sector, without accounting record systems and without issuing of invoices, pro-formas, receipts, etc.).  Difficulties of the implementers in satisfying the requirement to contribute up to 20% of the total value of a subproject as demanded for MAP financing;  Management overload because of the existence of two parallel financial systems for the same target groups and sometimes the same beneficiaries of the same organizations (Common Fund and MAP); 40  Deficient and unclear communication by the Bank about the criteria for the rating of performance of the Bank‘s portfolio. This made it difficult to take adequate measures to change the poor ratings‘ flags (red or yellow) which were often given to some aspects of the subprojects;  Lack of flexibility for disbursement, which caused delays in the installments to the organizations and caused on their part some discouragement;  Weak financial absorption capacity of many civil society and public sector organizations. 3. Decisions and its effectiveness During the implementation of the project, the World Bank tried different approaches that it thought could galvanize the performance of the project, but which proved not very productive. In early 2004, one tried out financing through the Rapid Results Initiative. In theory it seemed that the World Bank would be willing to relax some of its austere procurement requirements, e.g., by increasing the thresholds for purchase categories without invoice (direct procurement or shopping) and so forth. This turned out not to be the case. The management of these projects soon deplored that the World Bank itself was not prepared to show results in 100 days. Just the deliberations in Maputo and the requirement of no-objections to proceed took more than two months. Already in the final years of the MAP funding, when it was fully incorporated in the Common Fund and CNCS showed a noteworthy performance, an unsuccessful attempt was made to hire a grant management agency. Later, the World Bank decided to withdraw part of the financing and reallocate it to a new Rapid Results, Initiative, this time managed by UNDP. This administrative measure substantially modified the management processes for subprojects and warrants in time an objective evaluation. 4. Lessons  The MAP Project truly leveraged the energy for a dynamic real multisectoral response and for the capacity building of civil society to address the adversities imposed by the HIV/AIDS pandemic.  The MAP will have been, while it lasted, a learning exercise for the CNCS and for all stakeholders that benefited from its financing, because of its rigorous rules of financial management, scrupulous observance of procurement processes and its orientation towards results. It erred by exaggerating with its zealous austerity and requirement of banking transactions even where a bank didn‘t exist, and by failing to adapt to the objective development conditions of the country. 41  For CNCS the Project inspired the confidence of other international partners. The fact that the World Bank assumed the risk to initiate financing the response has been instrumental in mobilizing many other partners.  To some extent, the mandatory aide memoire, the rigorous requirements, the imposition of its functional models and operational systems and all the pressure that came with it being exerted on the national structures, contradicted part of the principles of the Paris Declaration, namely National Ownership and Alignment.  HIV prevalence at the start of the MAP was estimated at 15%, with an epidemic causing 500 new infections daily. ART was embryonic (with experience limited to the contribution of the Comunità di Sant'Egidio), and prevention of parent to child transmission --not to mention pediatric ART—was barely incipient. The community- based approach against the disease faced widespread taboos and prejudices. The massive scale-up of the response that the MAP provided allowed to demystify many issues and to take the message of prevention to many homes and families, schools and clubs. While it is true that the rate continues to rise, the project‘s contribution towards this pandemic, which actually demands a cultural revolution, is not to be underestimated. CNCS Comments of ICR We would first of all like to express our gratitude for the quality of the evaluation undertaken. Indeed, the evaluation reflects a thorough analysis of the Project accomplishments over the period of the Project implementation. It also raises critical issues in a deep and comprehensive understanding of the course of HIV and Aids epidemic in Mozambique, the national response as well as the general international trends and approaches. Overall the evaluation is balanced as it indicates areas of positive outcomes as well as those of poor accomplishment and does also address what failed to be considered in order for the response to be different. Notwithstanding the above, there are aspects of the project development that deserve a more in depth clarification. Also there are actors that have played an instrumental role over the course of project implementation which have been left out in this evaluation. In some parts, credits are given to actors that have not performed what the evaluation seems to have perceived on the ground. In order for it to also bring on board issues we consider important for a clear understanding of the contours the project has described over its lifespan, as well as a fair judgment of some events that have shaped up the implementation, below are comments we deem worth to consider: 1. Under point 18, it should be mentioned that delays in replenishment of accounts had also to do with lack of absorption capacity from the implementing agencies – which has not been mentioned in any point of the evaluation – as well as delays in replenishment of CNCS‘s accounts by the World Bank (the evaluation has pointed out 42 the Bank delays in reacting to issues that needed urgency as the project dealt with emergency situation under point 71). 2. Under point 19, it should be pointed out that the evaluation has been unfair in leaving out a mention to KPMG and Ernest and Young, two consult organizations that were contracted in by CNCS to strengthen project planning and management capacity of both CNCS and the implementing organizations. Under the same point, the mention to UNAIDS as an organization that has given technical assistance to CNCS is, somehow, surprising. In fact we have worked with UNAIDS as a valid partner within the broader context of partnership under which collaboration from other entities is always welcome. It is a broader collaborative effort in place that boosts up the national response. A mention to one partner leaving out others is, indeed, unfair. In the same point, the evaluation has found reluctance of CNCS in contracting a GMA. This, from our point of view, is just a simple judgment with no fundaments. It should be recalled the various negotiations‘ meetings - all with minutes given no objection by the World Bank - to settle an agreement with the GMA which ended up being difficult to come to an agreement. Critical issues most of which of fiduciary nature did not allow to consider what the GMA had proposed from their Inception Report which is available for the evaluation to consult, should the consultant deem it worth to do so. Some of t he issues proposed in the Inception Report, taking them on board, would change the scope of the ToR for contracting a GMA with all implications for an international competitive bidding that should be transparent and give equal consideration and treatment to all competing agencies from the onset of the contract throughout its implementation up to the end. 3. Under point 20, it should be clarified that there was no any king of agreement for provision to Ministerial Focal Points of additional payment, beyond their monthly salaries, due to their involvement in HIV and AIDS coordination in the Ministries. The idea of mainstreaming HIV in the sectoral plans was put in place in order to create capacity and sustain it for years to come. 4. The point 21 does not report fairly the facts. It would be more understandable to separate the treatment given to the Health Sector from that of the changes in the project direction which ended up with the contracting of UNDP to act transitorily as grant making agency. We believe that the contracting of UNDP and the whole procurement of medicines and other commodities deserve a fair more critical assessment which should be more impartial and better documented by all supporting documentation. By the time the evaluation team was undertaking the project assessment UNDP had less than 30% disbursements to project implementing agencies and that was even beyond the time period agreed upon for this agency to close the project. The evaluation seems to praise the way UNDP has entered into the scene and the strategies adopted to finance the Civil Society organizations. However, it fails to mention that CNCS‘s experience was crucial in the effort. The selection of organizations and their endorsement in the provinces was coordinated by the CNCS Provincial Nucleus, which 43 the evaluation does not mention at all. The Steering Committee of the Program spearheaded by the Minister of Health counted variably on the experience and expertise of CNCS‘s technical staff that had experienced working with all kind of organizations. The minutes of the committee are available and can be consulted any time. It was CNCS who suggested that for the lifespan of the initiative it would be good to strategize and bet on organizations that have some stability, and this needs to be clearly recorded. That was based on CNCS‘s experience with the so called Facilitating Agencies. 5. The evaluation of the Private Sector outsourcing needs to be deepened. The project evaluation seems to give a positive impression of the perform ance of the component, which it is not the position CNCS shares. 6. The point made by the assessment on point 71 was a common ground throughout the implementation of the project. One should add contradicting positions coming from the World Bank Project Team (management and procurement from the WB) which in most of the cases delayed project implementation. An example is the Result Based Management project, which after launching in 2003 had no objection three months later, despite being considered a 100 days result experience. We hope the above mentioned comments will be taken on board in this evaluation of the MAP Project. 44 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Comments Luisa Brumana (HIV Vice- Focal Partner, UNICEF, Mozambique) Comment Action taken ICR states: ―the heavy focus on fiduciary matters ICR text has not been changed. may have taken attention away from collection of core epidemiological and behavioral data‖. This was indeed the case. Please highlight the role of the Bank, at least Bank did not contribute as a UN mentioning the surveys in the body of the text, if the technical agency. ICR text was Bank was a main contributor to their not changed. implementation. The role of the Bank in fiduciary oversight was This is well reflected in the ICR. indeed highly appreciated by the HIV partners, and in particular Common Fund partners and the UN. While there are a number of considerations about the Since GOM made a similar RRF and its performance, there is little reflection of comment. Text now reflects this the temporary nature of this funding mechanism and lack of ownership of the RRI. on the absence, since then, of a real engagement in finding a nationally owned, longer term mechanism to fund civil society interventions. Comments by Guenter Dietz (GIZ, Mozambqiue) Comment Action taken Reading trough your report, it is evident the central These points are all reflected in role the MAP project has played in shaping the aid the ICR. architecture and building institutional structure around the multisectoral response for HIV in Mozambique. Especially to the CNCS, the MAP program and the WB has been the key donor leading the way in early stages of his mandate. Also the bold decision in 2008, to re- allocate resources away from CNCS, have triggered important subsequent positive changes for the further development of the organization and coordination structure. It is also evident that the MISAU has been a valuable Point already reflected in ICR. partner and absorbed MAP funds in a satisfactory way. The main shortcoming is the limited impact of the Text added that CSO capacity MAP on capacity building of Civil Society. There continued to be an issue till the are no improvements or systemic changes in CS end of HARP. Need for capacity Initiatives or CSO capacity. CSO performance building of NGOs is included in should have been monitored and evaluated more Lessons Learned section. closely. After the MAP closed, the situation of the unprepared and helpless CSO deteriorated. 45 Some donors are continuing their support for in the ICR mentions donors appreciation multisectoral response and we really miss the strong for Bank‘s fiduciary oversight arm and leverage from the World Bank MAP role. program in our daily activities. 46 Annex 9. List of Supporting Documents Coughlin, P., F. Bukali and J. Chalufo. ―Avaliação do Projecto Avante, Intervenções no Sector Privado contra o HIV/SIDA.‖ 2008. CNCS, Maputo. Martin, G. ―Portfolio Review of World Bank Lending for Communicable Disease Control.‖ IEG working paper 2010/3. Independent Evaluation Group. Washington, DC: World Bank. Durao, J. ―MEGAS 2007-2008, Despesas realizadas com HIV e SIDA, Mozambique.‖ 2010.Maputo World Bank. 1997. Confronting AIDS: Public Priorities in a Global Epidemic. Washington, DC: World Bank. World Bank. 2000. Intensifying Action against HIV/AIDS in Africa: Responding to a Development Crisis. Washington, DC: World Bank. World Bank. 2005. The World Bank’s Global HIV/AIDS Program of Action. Washington, DC: World Bank. World Bank. 2005. Committing to Results: Improving Effectiveness of HIV/AIDS Assistance: An OED Evaluation of the World Bank’s Assistance for HIV/AIDS Control. World Bank Operations Evaluation Department. Washington, DC: World Bank. World Bank, 2006. Mozambique, Better Health Spending to Reach the Millennium Development Goals. Report 33003. Washington, DC: World Bank. World Bank. 2008. The World Bank’s Commitment to HIV/AIDS in Africa: Our Agenda for Action, 2007-2011. Washington, DC: World Bank. World Bank. 2009. The Changing HIV/AIDS Landscape: Selected papers for the World Bank’s Agenda for Action in Africa, 2007-20011. Washington, DC: World Bank. World Health Organization Positive Synergies Collaborative Group, ―An assessment of interactions between global health initiatives and country health systems.‖, The Lancet, Volume 373, 2009. 47 IBRD 33451R1 30° E 35° E 40° E 10° S 10° S Lake La ke TA N Z A N I A To Mtwara Malawi Mocimboa MOZAMBIQUE Mueda da Praia a end Lug Metangula CABO ssa lo DELGADO Pemba NIASSA Me Lichinga Lichinga ue Montepuez M A LAWI ALAWI Marrupa q Catur bi io To Chipata To m au a e Lúr ZAMBIA Lilongwe oz t Mualadzi To M Pla Nacala To Petauke Mangoche Cuamba NAMPULA Furancungo Ribáu Ribáuè 15° S To 15° S To Zomba Lusaka Fíngo Fíngoè Montes Namule Nampula Moçambique Lago de TETE Zam (2,419 m) Cahora Bassa be Gurué Guru Zumbo Songo ze To Alto Molócue Blantyre Ligo Moatize Milange n ha Tete Angoche un ZAMBÉZIA Lic go Changara Mocuba To Mutoko Namacurra Pebane Sena Za mb Quelimane ZIMBABWE Catandica ez e Gorogosa Inhaminga To Harare SOFALA Chimoio INDIA N O CE AN in To Masvingo MANICA a Pl Monte Binga (2,438 m) (2,436 Beira 20° S 20° S u e i To Bu z Masvingo i q Espungabera m b Nova Mambone e z a Sav Inhassôro To Rutenga M o Vilanculos Chicualacuala 0 50 100 150 200 Kilometers Machaíla To Messina INHAMBANE 0 50 100 150 Miles Chigubo Mapai GAZA Ch a Lim po M O Z A M B I QUE ngane op SOUTH o Massingir Inhambane AFRICA Panda SELECTED CITIES AND TOWNS Guija Inharrime Chibito PROVINCE CAPITALS 25° S MAPUTO NATIONAL CAPITAL 25S To Xai-Xai Nelspruit This map was produced by RIVERS Manhica the Map Design Unit of The World Bank. The boundaries, MAIN ROADS Moamba colors, denominations and Matela MAPUTO any other information shown RAILROADS on this map do not imply, on To the part of The World Bank Mbabane Group, any judgment on the PROVINCE BOUNDARIES legal status of any territory, SWAZILAND Zitundo or any endorsement or acceptance of such INTERNATIONAL BOUNDARIES boundaries. 30° E 35° E JANUARY 2007