Document of The World Bank Report No: ICR00001688 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-33350 IDA-43910 TF-24755 TF-53570) ON A CREDIT IN THE AMOUNT OF SDR 22.4 MILLION (US$ 30.0 MILLION EQUIVALENT) TO THE REPUBLIC OF YEMEN FOR A CIVIL SERVICE MODERNIZATION PROJECT December 2010 Social and Economic Development Department Egypt, Yemen, Djibouti Country Office Middle East and North Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective October: 31, 2010) Currency Unit = SDR 0.63666 SDR = US$1 1.57078 US$ = SDR 1 FISCAL YEAR January 1- December 31 ABBREVIATIONS AND ACRONYMS AFMIS Accounting and Financial Management Information System ARADO Arab Administrative Development Organization BIS Biometric Information System CAS Country Assistance Strategy COCA Central Organization for Control and Auditing COFOG Classifications of Functions of Government CPIA Country Policy and Institutional Assessment CSM Civil Service Modernization CSMP Civil Service Modernization Project CSF Civil Service Fund CSR Civil Service Reform DFID Department For International Development DPG Development Policy Grant EDB Employee Database ESAF Extended Structural Adjustment Facility EU European Union GDP Gross Domestic Product GFSM Government Finance Statistics Manual GOY Government of Yemen HR Human Resources IDA International Development Association IDF Institutional Development Fund IEG Independent Evaluation Group IF Innovation Fund ILO International Labor Organization ISR Implementation Status and Results report KPI Key Performance Indicators M&E Monitoring and Evaluation MENA Middle East and North Africa MIS Management Information System MOCSAI Ministry of Civil Service and Insurance MOF Ministry of Finance MOPIC Ministry of Planning and International Cooperation N/A or N.A. Not Applicable PIU Project Implementation Unit PDO Project Development Objectives PAD Project Appraisal Document PIP Project Implementation Plan PSMAC Public Sector Management Adjustment Credit QALP Quality Assessment of Lending Portfolio RICs Regional Information Centers SDR Special Drawing Rights SOE Statement of Expenditure TA Technical Assistance TS Technical Secretariat UNDP United Nation Development Program USD United States Dollar YR Yemeni Rial Vice President: Shamshad Akhtar Country Director: David Craig Sector Manager: Guenter Heidenhof Project Team Leader: Robert Beschel ICR Team Leader: Fabian Seiderer REPUBLIC OF YEMEN Civil Service Modernization Project 1. Project Context, Development Objectives and Design 1 1.1 Context at Appraisal .................................................................................................. 1 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved)2 1.3 Revised PDO (as approved) and Key Indicators, and reasons/justification.............. 3 1.4 Main Beneficiaries .................................................................................................... 4 1.5 Original Components (as approved in the PAD and legal agreement) ..................... 4 1.6 Revised Components ................................................................................................. 5 1.7 Other significant changes .......................................................................................... 5 2. Key Factors Affecting Implementation and Outcomes 5 2.1 Project Preparation, Design and Quality at Entry ..................................................... 5 2.2 Implementation ......................................................................................................... 7 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization ........ 10 2.4 Safeguard and Fiduciary Compliance ..................................................................... 11 2.5 Post-completion Operation/Next Phase .................................................................. 11 3. Assessment of Outcomes 12 3.1 Relevance of Objectives, Design and Implementation ........................................... 12 3.2 Achievement of Project Development Objectives .................................................. 15 3.3 Efficiency ................................................................................................................ 22 3.4 Justification of Overall Outcome Rating................................................................. 22 3.5 Overarching Themes, Other Outcomes and Impacts .............................................. 22 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops: N/A23 4. Assessment of Risk to Development Outcome 23 5. Assessment of Bank and Borrower Performance 24 5.1 Bank Performance ................................................................................................... 24 5.2 Borrower Performance ............................................................................................ 27 6. Lessons Learned 28 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners 29 Annex 1. Project Costs and Financing .......................................................................... 31 Annex 2. Outputs by Component .................................................................................. 32 Annex 3. Economic and Financial Analysis ................................................................. 41 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............. 42 Annex 5. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 44 Annex 6. Comments of Cofinanciers and Other Partners/Stakeholders ....................... 66 Annex 7. List of Supporting Documents....................................................................... 67 MAP 68 A. Basic Information RY-CIVIL SERVICE Country: Yemen, Republic of Project Name: MODERN IDA-33350,IDA- Project ID: P050706 L/C/TF Number(s): 43910,TF-24755,TF- 53570 ICR Date: 12/23/2010 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: YEMEN Original Total XDR 22.4M Disbursed Amount: XDR 21.5M Commitment: Revised Amount: XDR 21.5M Environmental Category: C Implementing Agencies: AFMIS PIU Ministry of Civil Service & Insurance Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 02/05/1998 Effectiveness: 11/29/2000 11/29/2000 Appraisal: 06/18/1999 Restructuring(s): Approval: 04/20/2000 Mid-term Review: 10/07/2003 Closing: 12/31/2005 06/30/2010 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Unsatisfactory Risk to Development Outcome: Substantial Bank Performance: Unsatisfactory Borrower Performance: Moderately Unsatisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Moderately Moderately Quality at Entry: Government: Unsatisfactory Unsatisfactory Implementing Moderately Quality of Supervision: Unsatisfactory Agency/Agencies: Unsatisfactory Overall Bank Overall Borrower Moderately Unsatisfactory Performance: Performance: Unsatisfactory i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry Yes None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes Moderately Satisfactory 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 100 100 Theme Code (as % of total Bank financing) Administrative and civil service reform 100 100 E. Bank Staff Positions At ICR At Approval Vice President: Shamshad Akhtar Kemal Dervis Country Director: Sidi Mohamed Boubacar Inder K. Sud Sector Manager: Guenter Heidenhof Mustapha Kamel Nabli Project Team Leader: Fabian Seiderer Linda Van Gelder ICR Team Leader: Fabian Seiderer ICR Primary Author: Fabian Seiderer F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The Project Development Objective is to assist government in the creation of capacity, institutions and systems for improved and sustained utilization of human and financial resources of the Yemen civil service. To this end, the project would help put in place core personnel and financial management systems, establish a mechanism to reduce the number of unqualified civil servants, and initiate a restructuring process in individual ministries. To make this foundation as solid as possible, another objective of the project is to upgrade the management capacity in the civil service. Revised Project Development Objectives (as approved by original approving authority) None ii (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 Indicator 1 : Core systems improved Considering the succesful development of Financial or Human AFMIS, the overall Resource management Value Fully functional achievement of this are developed. Procedures quantitative or HR and Financial component/ are streamlined and Qualitative) system. indicator can be process is estimated at 50% institutionalized. and is therefore rated moderately satisfactory. Date achieved 03/23/2000 06/30/2010 06/30/2010 Comments (incl. % achievement) Indicator 2 : Workforce streamlined and becoming more efficient CSF was fully CSF fully operational but is operational and suffering budget over 34000 constraints. 17 753 redundant workers were employees processed mainly Value Civil service is bloated removed following from defunct State quantitative or and lacking in crucial transparent enterprises. 3792 Qualitative) skills. procedures. double dippers Double dippers were removed, of and ghost workers which 25 through fully removed the incomplete from all public biometrics system. employment. No ghost worker was removed Date achieved 03/23/2000 06/30/2010 06/30/2010 Comments (incl. % It is rated unsatisfactory achievement) Indicator 3 : Operating procedures within selected agencies streamlined. Pilot ministriy Most reengineering Antiquated organizational Value organizational studies have been structures and mandates; quantitative or structures finalized and functional duplication and Qualitative) streamlined; key approved, with excess staffing business processes substantial delays, iii simplified; basic but there is little information evidence of structures implementation or installed. results. Date achieved 03/23/2003 06/30/2010 06/30/2010 Comments (incl. % It is rated unsatisfactory achievement) Indicator 4 : Capacity to sustain reforms and administer the public administration enhanced. After substantial delays, up to 27 training programs Effective mid and have been delivered high level training to 1,500 civil Limited skills for conducted on an servants, including advancing a long-term ongoing basis; Value 5 programs for 600 reform program and training on core quantitative or senior Government managing and computer systems Qualitative) managers. The administering the civil completed. absence of service. Training strategy qualitative developed for information and whole civil service sustainability are an issue. No overall training plan. Date achieved 03/23/2000 06/30/2010 06/30/2010 Comments (incl. % It is rated moderately unsatisfactory achievement) (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Operating manuals written and approved by cabinet resolution. Computerized Indicator 1 : systems (AFMIS & Biometric) in place where appropriate. HR management The HR HR and financial system allowing information management procedures government to systems (EDB and are outdated and flawed. track employees BIS)are incomplete Value Systems to be fully and their impact and not up to date. (quantitative computerized. HR system on the wage bill is AFMIS has been or Qualitative) enhanced by the in place. AFMIS developped and implementation of a system procedures successfully pilot biometric information are developed and tested in the 4 main system. computerized Ministries. system is in place. Date achieved 03/23/2000 06/30/2010 06/30/2010 iv Comments (incl. % It is rated moderately satisfactory achievement) Indicator 2 : Compensation of civil servants is improved in a budget neutral fashion. Wages have been Individual wages increased twice but are competitive to accross the board The civil service is attract and retained and not linked to bloated and underpaid. qualified the streamlining Wage bill for the public personnel. efforts. The number Value service is very high (15 Number of civil of civil servants (quantitative percent) while individual servants is and the wage bill or Qualitative) wages of civil servants reduced, were not reduced. are well below poverty individual wages Pay progression has line. Compression ratio are improved, slightly improved 2:1. compression ratio (1:3) but salary is improved. attractiveness is questionable. Date achieved 03/23/2000 06/30/2010 06/30/2010 Comments (incl. % the rating is unsatisfactory achievement) Diagnostic Reengineering studies are completed and implementation of programs Indicator 3 : is prepared. Studies have been Customs, Tax, Civil Implementation of completed after Service and COCA are the studies is Value substantial delays, largely ineffective and completed. (quantitative but there is little overstaffed agencies Agencies have or Qualitative) evidence of requiring high level been fully implementation, reengineering. reengineered. apart for MOCSAI. Date achieved 03/23/2000 12/31/2007 06/30/2010 Comments (incl. % the rating is unsatisfactory achievement) Effective program plan for follow-on reforms developed. Civil service training Indicator 4 : program implemented with agreed number of civil servants trained in management of public administration Up to 1500 civil A core group of Civil Service has no servants and civil servants is effective training mangers have been Value able to act as function. No civil trained. There is (quantitative change agents to servants able to lead little qualitative or Qualitative) push change management information and administration program. impact on the civil reform in Yemen service reform. Date achieved 03/23/2000 06/30/2010 06/30/2010 Comments the rating is moderatly unsatisfactory (incl. % v achievement) G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 06/17/2000 Satisfactory Satisfactory 0.00 2 02/28/2001 Satisfactory Satisfactory 0.00 3 08/02/2001 Satisfactory Satisfactory 1.08 4 08/03/2001 Satisfactory Satisfactory 1.08 5 04/10/2002 Satisfactory Satisfactory 1.58 6 07/17/2002 Satisfactory Satisfactory 1.88 7 12/11/2002 Satisfactory Satisfactory 2.04 8 12/19/2002 Satisfactory Unsatisfactory 2.04 9 06/16/2003 Satisfactory Unsatisfactory 2.51 10 12/12/2003 Satisfactory Unsatisfactory 4.74 11 06/30/2004 Satisfactory Unsatisfactory 7.57 12 12/27/2004 Satisfactory Unsatisfactory 8.53 13 05/04/2005 Satisfactory Moderately Satisfactory 10.14 14 06/26/2006 Satisfactory Moderately Satisfactory 17.69 15 06/22/2007 Satisfactory Satisfactory 20.87 16 12/03/2007 Satisfactory Satisfactory 22.02 17 09/23/2008 Satisfactory Satisfactory 25.50 18 02/02/2010 Moderately Satisfactory Moderately Satisfactory 27.71 Moderately 19 12/18/2010 Unsatisfactory 31.67 Unsatisfactory H. Restructuring (if any) Not Applicable vi I. Disbursement Profile vii 1. Project Context, Development Objectives and Design (this section is descriptive, mainly taken from the CAS, PAD, it is not evaluative) 1.1 Context at Appraisal The country background at appraisal is characterized by a fragile environment and daunting challenges, but a strong political will for reforms. As one of the poorest countries in the region with a GDP per capita of US$ 320, and a poverty rate of 25% of the population in 1998, Yemen faced enormous development challenges, including very low levels of social indicators such as for poverty, illiteracy, morbidity, malnutrition; continued depletion of scarce water resources; combined with a high growth rate of the population, much of which was underserved by modern infrastructure and by public services; and a high dependence on a single and rapidly depleting commodity, oil. The country faced further challenges, as the Republic of Yemen was a relatively new and fragile State in the process of nation building since the reunification of North and socialist South Yemen in 1990 and the end of the civil war in the South in 1994. The government's actions and authority were further challenged by political tensions, insurgencies in the North-East province, secessionist sentiments in the South and prevalent security threats, as evidenced by a spate of riots, bombings, and kidnappings in 1998. Despite these circumstances, the political will to address these daunting challenges and the Government's commitment to reform were high and yielded encouraging results, according to the Bank's Country Assistance Strategy (CAS 99-02). The Government completed successfully a first round of macro- economic and structural reforms ranging from cuts in fuel and food subsidies to measures of trade liberalization. It succeeded in stabilizing its economy, hard hit by the economic and social consequences from the civil war and the Gulf war, and secured a substantial debt cancellation from the Paris Club, reducing its debt to GDP ratio from 176% to 68.5% in 1997. Despite a short set back due to a sharp decline in oil prices in 1998 (which halved the growth rate and reduced government revenues by 20%), the Government engaged in a second wave of reforms to enhance economic growth, private sector development, privatization, and revenue mobilization, with support from the IMF. However, further progress on the ambitious reform agenda and development efforts were severely constrained by the public administration's limited capacity and skills, as well as by a bloated, poorly paid and managed civil service. Acknowledging these limitations and the unsustainable wage bill, the Government committed to thoroughly reform and reengineer its administration and civil service. The World Bank responded favorably and made improving governance and public administration modernization its overarching goal and top priority in the 1999-2002 CAS. The Bank committed to increase its engagement both in financial terms (with a 25% increase in lending per annum. compared to the previous CAS) and in the scope of interventions, combining a programmatic series of Public Sector Management Adjustment Credits (PSMACs) with an Civil Service Modernization Project (CSMP), a Legal and Justice Development project and support to the creation of an Independent Commission against Corruption. The CSMP was thus part of a comprehensive response strategy to the Government's challenges and reform program. The range and scope of this assistance strategy reflected the reform momentum and optimism at the time. At appraisal, the public administration and civil service was plagued by a number of weaknesses which rendered the system largely inefficient and impeded the government's ability to deliver essential public services. The size of the civil service was bloated with extremely low pay scales, low capacity, opaque hiring and promotion practices, and virtually absent enforcement mechanisms. The public administration was further characterized as an environment where nearly all the core systems were inadequate and paper-based (ranging from personnel and financial management, to budgeting, procurement, and property management), organizational structures and work flows were duplicative and 1 unnecessarily complex, and human resource-management was virtually absent. Human resources data was fragmented, incomplete, and not up to date. The personnel data creation, management and reporting processes were unclear and uneven across spending units. Many of the main weaknesses stemmed notably from the fact that during the 1990s, Yemen's civil service was a central element of the reunification process, began to act as an employer of first resort, and served as a central component in the government's social safety net. Consequently, the civil service doubled in the 1990s. By 1999, civil servants represented about two percent of the population (17.6 million) or eight percent of the labor force active, which in 1999 was estimated at 4.4 million. Additionally, employment in public enterprises was over 83,100. In total, public sector employment accounted for four percent of the population or 14 percent of the labor force. Due to the size of the civil service, the wage bill represented 14 percent of GDP and 35 percent of total government expenditures in 1999. This is high by any international standard. The decline in oil revenues further reinforced the need to reduce the wage bill to stimulate growth by ensuring that scarce resources could be dedicated to investment or operational costs, that government does not unwarrantedly and excessively interference in private sector economy through its inflated civil service, and ensure that government spending is in a social equitable manner. Past efforts to contain the financial impact of the civil service increase led to a dramatic reduction in wages and had dire consequences. By 1996, the average real wage was only 15 percent of the 1990 real wage level. Wages were also seriously compressed, with senior managers earning only 3 times the lowest level civil servant (against an international compression rate of 6-7) and only about 11% of the wages of their private sector contemporaries. These low wages resulted in the inability to attract high quality workers as well as in lack of motivation and effort of existing staff, which translated in short working hours, high absenteeism, and low productivity. To compensate for the loss of revenue, many civil servants took up parallel jobs in the public sector (double dippers) or in the private sector (ghost workers), and relied increasingly on petty corruption to top-up salaries, and further reinforced the notion that wages paid to a civil servant are nothing more than a safety-net. The public administration's organizational structure was complex and heterogeneous. It comprised 28 ministries and over 80 "departments" with a great difference in size and capacity, ranging from a ministry such as education (over 150,000 staff) to the Social Security Fund (8 staff). The organizational structures of many of these departments was a spider- web of complexity. And, despite the intricate structures, the chain of command generally passed upward with little ability of lower level managers to take decisions or be held accountable for activities even as mundane as approval of staff's annual leave. This shared assessment, guided the Government's comprehensive reform agenda and the Bank's intervention rationale. The Government adopted in 1998 a Strategic Framework for Civil Service Modernization, to: (i) streamline government entities and increase the transparent and consistent application of procedures, (ii) improve budgeting and financial management systems, (iii) restructure public employment and establish a transparent personnel management system, and (iv) establish effective controls on staffing levels and reduce the number of redundant workers. This program was expected to take ten years, although slippage was anticipated. The Bank committed to support the implementation of this holistic strategy, through the above mentioned comprehensive assistance strategy, of which the CSMP was the flagship project. As stipulated in the PAD, this very ambitious civil service reform agenda was to be pursued in phases, with CSMP as the first phase to establish the foundation for the Government's long- term reform strategy for modernizing the civil service. 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) The Project Development Objective is to assist government in the creation of capacity, institutions and systems for improved and sustained utilization of human and financial resources of the Yemen civil service. To this end, the project would help put in place core personnel and financial management systems, 2 establish a mechanism to reduce the number of unqualified civil servants, and initiate a restructuring process in individual ministries. To make this foundation as solid as possible, another objective of the project is to upgrade the management capacity in the civil service. The Project Key performance indicators are: (i) the production of consolidated financial statements in a timely manner, and (ii) employment based on positions and merit with transparent application of civil service law. The Component objectives and key performance indicators are as follows: Component output objectives Output indicators 1. Core systems improved Operating manuals written and approved by cabinet resolution. Computerized systems in place where appropriate. 2. Workforce streamlined and Number of civil servants on the payroll reduced. High quality becoming more efficient workers retained. 3. Operating procedures within Workflow delivery is improved against service standards in selected agencies streamlined restructured agencies. 4. Capacity to sustain reforms and Effective program plan for follow-on reforms developed. administer the public Civil service training program implemented with agreed administration enhanced number of civil service servants trained in management of public administration and specialist work skills for operating core systems. Source: PAD Annex 1 Even though the project's logical framework does not include outcome indicators and targets at component level, the Government's Policy letter requesting the credit does. It foresees the following expected results, which shall be considered for the present assessment: 1. Removal of 12 000 overdue pensioners by December 2000; 2. Removal from the payroll of all double dippers and ghost workers within 4 month of their identification through the biometrics system; 3. Transfer to the Civil Service Fund of workers identified through the restructuring process; 4. Significant overall reduction in the number of civil servants over the project period; 5. Adoption of a Wage and employment policy by June 2001, that is fiscally affordable and that permits the gradual decompression of wages; 6. Restructuring of 4 agencies by December 2000 and 4 more by September 2001; 7. Introduction of revised, computerized personnel management, payroll and AFMIS systems in Sana'a and gradual expansion to all administrative units; 8. Design and carry out of training programs for personnel in the restructured departments. Likewise, the logical framework does not include a specific baseline but the project document details the situation of the civil service at project start, as mentioned under section 1.1 above. 1.3 Revised PDO (as approved) and Key Indicators, and reasons/justification The project's development objectives and key performance indicators were not revised during its two restructuring, respectively in 2005 and 2007. In March 2008, the KPI were specified and enhanced in the context of an additional financing credit (4391-YEM). However, as the latter did not reach effectiveness, the initial PDO and KPI remain the basis for this evaluation. 3 1.4 Main Beneficiaries The project's main direct beneficiary is the Government of Yemen, through a better and more sustainable utilization of human and financial resources in public administration. Eventually, further expected beneficiaries are: (i) the qualified civil servants, to be rewarded for good performance with higher salaries, and (ii) the population, which can expect higher quality, more efficient delivery of public services. 1.5 Original Components (as approved in the PAD and legal agreement) The original project had four thematic components and one for project management and administration. 1. The first component ­ Improve Core Systems (US$ 14.1m1)- aimed to improve the Government's financial and human resources management systems. It supported: (i) the modernization of the human resources management framework (policies and practices), (ii) the establishment of personnel information systems, through a census and verification of personnel records, a computerized employee database (EDB) and a Biometric Information System (BIS), the introduction of transparent hiring and job descriptions, (iii) the standardization and streamlining of payroll procedures and processes as well as the planning of payroll computerization, (iv) the development of an accounting and financial management information system (AFMIS). 2. The second component ­ Civil Service Fund (CSF) (US$ 4.9m), aimed at supporting the rationalization and strengthening of the Civil service, through the: (i) enforcement of retirement rules, (ii) removal from payroll of ghost workers and double dippers, through the improved core systems, (iii) the identification and transfer of surplus workers to a newly established CSF, and (iv) the design, implementation and monitoring of assistance packages to redundant workers, through the CSF. 3. The third component ­ Streamline and Rationalize (US$ 1.5m) aimed at (i) reviewing the mandates, responsibilities and objectives of pilot agencies, based on a functional analysis (ii) removing duplications and overlap, realign agencies' missions and functions and develop new policies, including decentralization initiatives, and (iii) analyze staff requirements and introduce job descriptions. This was essentially done through horizontal studies (Government mandate, institutional performance indicators) and agency specific studies for the Ministry of Civil Service and Insurance, the Ministry of Social Affairs, the Tax Authority and the Central Office for Control and Audit (COCA). In tandem re-engineering of other Ministries (such as the Ministry of Health, Prime Minister's Office, Ministry of Labor) would be supported through other donors' involvement (EU, UNDP, ILO). This component included also the set up of an innovation fund to support, in a flexible manner, the testing of new ideas and "quick fixes" to support the reform momentum. 4. The forth component- Capacity building (US$ 3.9m) supported two dimensions of capacity building: (i) skills and activities to sustain the reform in the long term and (ii) skill development for the management and administration of a modern public service. Under the first dimension, the project will provide support for managing, monitoring and evaluating the overall reform program. This also includes support for public information and internal communication programs. Under the second dimension, training would be provided for mid and high level management levels to start developing an effective management framework, users of new core systems (component 1), and project management teams within ministries. 1 The figures in parentheses represent the IDA funding per component as per the PAD. 4 5. The fifth component was dedicated to program management and administration (US$ 2.5m) to set-up and operate a fully functioning program management structure, with a Project Implementation Unit (PIU) both within the Ministry of Finance (to support AFMIS) and the Ministry of Civil Service to cover the other components. 1.6 Revised Components A project restructuring in December 2005 led to a formal revision of project components, with a reallocation of 27.6% of the total budget from the capacity building and the civil service components to the AFMIS component. The reallocation financed essentially additional IT equipment for AFMIS as well as technical assistance to the project management and the functional design of AFMIS. In order to compensate for the substantial budget reduction of the capacity building component, from US$ 3.9m to US$ 620,000, the training activities were transferred to a larger Dutch Trust Fund. The financial reallocation increased substantially the resources allocated to IT systems. 1.7 Other significant changes The project's initial closing date of 31 December 2005 was extended twice, respectively to December 31, 2007 and to June 30, 2010, thereby doubling the implementation period of the project from 5 to 10 years. The Bank financing was complemented by two Dutch Trust Funds (TF). The first TF (nb: 24755), was approved in August 2001, with a budget of US$16 m intended to finance the IT and office equipment of MOCSAIs Regional Information Centers. However, due to delays in implementation, the funds could not be disbursed and were canceled after expiry of the Trust Fund on December 31, 2003. The second TF (nb: 53570), for training and capacity building in Yemen, was approved in February 2005, with a budget of US$ 6 m. It notably financed the innovation fund as well as the project's capacity building activities after the 2005 restructuring. It closed end of December 2007 with a disbursement rate of 65%. Initially the biometrics data capturing was limited to the core civil service. However, following a presidential decree in 2004, the Military, Security and Judiciary sectors were included, thereby doubling the size of civil servants to be captured in the project and delaying the completion of the BIS. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Positive key factors: A timely and long preparation phase. The project preparation was timely as it coincided with the FY 99-05 Country Assistance Strategy (CAS) preparation, ensuring cross-fertilization and strong coherence between both. The CSMP was considered the flagship project of the CAS's governance pillar. Project preparation lasted nearly 3 years, building on an IDF grant which supported consultations on the Government's "Strategic Framework for Civil Service Modernization", the initial set up of the EDB, based on the consolidation of personnel files in 1998. An important project preparation facility, of US$ 2m, supported analytical work such as a study on the public sector management building blocks in Yemen (January 2000). It also financed the set up of the project management units and implementation framework, prior to project start. The project also built on the preparation of the Public Sector Management Adjustment Credit (PSMAC), approved by the Board a year earlier in March 1999. The CSMP was thus part of a comprehensive assistance strategy. The long preparation phase enabled to establish strong relations between the reform champions on both sides, and the initiation of a dialogue based on data from the civil service census. 5 The scope and depth of stakeholder consultation aimed to build understanding and consensus on the civil service reform and the supporting project. According to the CAS 99, the proposed governance and civil service projects were endorsed during the two rounds of stakeholder consultations organized in March and September 1998, both in Sana'a and in governorates. These consultations included government services, civil society, academe, donors, parliament, and tribal leaders. The supported Strategic Framework for Civil Service Modernization, was discussed in a high level national conference held in 1998, under the auspices of the President. The project preparation was based on a social and institutional assessment including qualitative focus group discussions in four Governorates and quantitative household surveys covering 600 households, which underlined the priority public services expected as well as the perceived problems in the public/civil service. IEG's 2006 country strategy evaluation confirmed the project's participatory process of consensus building, in line with best practices at the time. However, as detailed under sections 3 and 4, the impact of these positive factors on project implementation and outcomes was less than expected and/ or offset by the following negative factors: Negative key factors : The scope and ambitions of the CAS 99 and the project reflected the high optimism at the time and were based on the assumptions that the positive outlook and reform momentum persists. The appointment as Minister of Civil Service, of the former Minister of Finance, who successfully implemented the first structural adjustment plan, was seen as an indicator of the Government's strong commitment to the civil service reform. However, the degree of optimism and commitment diminished, as evidenced by the less ambitious CAS 02, but the CSMP remained unchanged. The overambitious and complex project design both in terms of scope (5 components and 15 subcomponents) and sophistication (development of complex human resources and financial management IT systems, such as AFMIS, EDB, BIS). The project was seeking to address at the same time major legislative and regulatory reforms, civil service downsizing, institutional and organizational restructuring, new core management information systems, as well as intensive capacity building. This strained limited Government management and implementation capacity and increased resistance from major constituencies affected by these numerous and simultaneous reforms. The bleak findings of the World Bank's evaluation of two decades of support to Civil Service Reforms (CSR), published on April 27, 19992, might have warranted a stronger consideration of the operation's risks and a more prudent and gradual approach in order to mitigate these. The evaluation found "that Bank supported CSRs were largely ineffective in achieving sustainable results in downsizing, capacity building and institutional reform, in part due to significant political difficulties in implementing CSRs". These were also the objectives and results of the Yemen CSMP. The absence of an initial public finance management reform strategy and MoF reengineering underpinning AFMIS, affected the system's relevance, as it automated old processes and was lacking important functions. A functional design, undertaken after the project's extension, enabled to review core processes but the MoF and AFMIS are still lacking cash management function. A public finance management reform strategy was finally adopted on August 9, 2005, through Cabinet Resolution No. 253 of 2005. Insufficient integration of in-country lessons learned in the project design. The PAD included a review of the main lessons learned from: (i) project implementation in Yemen, suffering from weak ownership, project management and M&E; (ii) TA projects, relying too much on supply driven, international expertise and suffering from unrealistic timeframes; (iii) public sector reform projects, 2 WB ODE report No. 19211 6 advocating for component specific and adaptable work programs; (iv) labor force adjustment programs, suggesting excessive severance packages, adverse selection problems, rehiring and inefficient training programs; (v) the CSMP project preparation, highlighting the need to put in place implementation capacity. However, the project design did not fully reflect these recommendations, as evidenced by the weaknesses in the project's implementation framework and M&E, the lack of specific and adaptable work programs and the unrealistic timeframes. The project design and implementation framework overstretched weak inter-ministerial coordination capacity. The holistic project approach and the horizontal nature of the civil service and financial management reforms required a strong coordination mechanism between the numerous ministries and stakeholders involved and affected by these reforms. The situation was worsened by the numerous donors supporting notably the reengineering and capacity building reforms (UNDP, DFID, EU, Dutch cooperation, ...). Donor coordination was an issue, particularly in the field of government agency streamlining and capacity building. The PAD acknowledged it and the innovation fund was supposed to test the feasibility of establishing a trust fund to pool all donors resources for reengineering. However, the much delayed innovation fund did not achieve this objective and the absence of a donor coordination framework weighted negatively on the reform and project implementation. The absence of a full fledge socio political analysis, weighted negatively on the design of a project supporting a very sensitive and inherently political reform program, such as civil service and public administration restructuring and streamlining. Socio-political analysis was at its infancy at the time, and the underpinning analytical work focused mainly on the formal regulatory and institutional set up. Important risks were underestimated in the project and the mitigation measures turned out to be insufficient. The following risks materialized and affected the project most (i) the deterioration of the political and security situation which impacted the Government's reform program and authority in parts of the country; (ii) the prevalence of weak governance, with corruption and patronage undermining the civil service streamlining and reform; (iii) the discontinuation of the concomitant Civil Service policy loans and IMF program; (iv) the strong pressure from regional/ tribal constituencies on public sector employment, in the context of unfavorable labor market conditions and regional tensions; (v) the resistance and influence of strongly decentralized Governorates and spending units on the establishment of central civil service management systems and databases. 2.2 Implementation Since its approval in March 2000, and despite the important preparatory work, the project's start was very difficult, uneven and slow. Until the end of 2004, the project performance was very poor, with a disbursement rate of 25% after 4,5 years of implementation and 22 months in problem status, among the longest in the Bank's overall portfolio. The project also had the highest number of problem flags (5) including for procurement, project management, and financial performance. The implementation of the AFMIS component was particularly problematic, as it was lacking a conceptual design, considered mainly as a computerization project with little ownership and was suffering from serious project management issues. The civil service component was also experiencing important procurement delays (BIS contract signed only in the end of the year 2004). Given the bleak prospect of achieving the project development objectives in the remaining year, the Bank engaged the Government at high level to resolve these issues or cancel the credit. Following a request for the project's extension and a substantial budget reallocation to the non-performing AFMIS 7 component, the Bank substantially increased its supervision and involvement in the project in an attempt to turn it around. In 2004, 4 missions were carried out, including at director level, three management review meetings, chaired by the Vice President were held 3 and a high-level dialogue was engaged with the Yemeni authorities to save the project. By end of 2004, an agreement was reached on important corrective actions to improve the project's implementation and outcomes. These measures included the development of a budget classification consistent with GFS 2001, necessary for the development of AFMIS, the recruitment of an advisor for the development of AFMIS's conceptual design, the recruitment of an international project manager supporting the AFMIS PIU and an independent firm to test the AFMIS system before implementation. It further led to a revised AFMIS work program and budget reallocation. The Mid-term review, produced by the project in October 2003 failed to provide an accurate overview of the project's performance and output and did not mention the serious issues raised during the different supervision missions at the time. It focused on activities and inputs and provided an unrealistic and over optimistic outlook for future project implementation. The project was revised and extended two times: The first revision took place on December 15, 2005, on the background of a renewed Government commitment and initial improvements in project implementation. Following the intense dialogue with the Bank in 2004, the Government agreed on an action plan to achieve the project development objective, on the above mentioned corrective measures as well as on the set up of a high-level steering committee, to improve project guidance, appropriation and coordination. Project execution showed some signs of improvement with an increase in disbursement and the exit of the project from problem status. Given these positive developments, the Bank agreed to a project revision comprising: (i) a two-year extension of the project closing date, to December 31, 2007; (ii) an important budget reallocation (27.6% of total) from the non performing civil service and capacity building components to AFMIS, and (iii) the transfer of the capacity building component to a Dutch Trust Fund (TF-24755). As a result the project increased its focus on the improvement of core systems such as AFMIS and the Biometric Information System. The second revision took place on December 21, 2007, based on sustained efforts to improve project implementation and a request for additional financing. The project's closing date was extended to June 30, 2010 and a proposal for additional financing of US$14 million was approved by the Board on April 3, 2008. The additional financing (IDA 4391-YEM) was seeking to finance the costs associated with the scaling up of activities to enhance the project's impact. It was supposed to: (i) provide support for the continuation of the reengineering process in selected line departments; (ii) assist with the expansion of the Human Resource Database beyond the civil service to include employees in the armed forces, judiciary and the security services, the establishment of a comprehensive Human Resource Management System and the roll out of AFMIS to all ministries and agencies in Sana'a; and (iii) support the revision of all current relevant legislation to strengthen the rule of law, meritocracy and performance orientation in public employment. Eventually the additional financing failed to reach effectiveness. Consequently, the formal and long overdue revision and strengthening of the projects results framework, with the adoption of detailed performance indicators, which was part of the additional financing package, did not materialize. The revision was thus limited to the extension of the project closing date to June 30, 2010 and the project's component and logical framework remained unchanged. 3 Missions took place in March, September, November, and December 2004 as well as in June and September 2005. Management reviews were held in July, October and December 2004 8 Key factors affecting project implementation: Factors outside the control of Government or implementing agency: The deterioration of the security situation affected cooperation and project implementation since the beginning, with stronger disruptions by security and terrorist related incidents in 2000, 2002-2003, 2007-2009. The repeated wars in the north created further political and security tensions, affecting reform and project implementation. The strong fluctuation in oil prices and government revenues, with a surge in 2000 and 2008 adversely affected the acceptance of the civil service streamlining and restructuring reforms and fuelled resistance. Likewise, the deterioration of the economic and employment situation in the non- hydrocarbon sector increased pressure on the civil service. The important brain drain and high rotation of trained staff further strained the public administration's capacity, already suffering from a wage policy and structure not attractive for qualified staff. This also strongly affected the effectiveness of capacity building and training activities. Lack of appropriate technical expertise and qualified contractors in Yemen adversely affected contract and project implementation. This was particularly the major IT contract to develop and implement the AFMIS system. Factors subject to the control of Government or implementing agency: Positive factors The direct involvement of some senior government officials in project management and implementation, which compensated to a certain extent the lack of capacity and appropriation by the middle management and deficiencies at the PIU level. This was particularly the case after the 2005 project revision, as evidenced by the AFMIS component turn around and by the BIS expansion to the military and security sectors. Negative factors The absence of sustained commitment of the Government to the sensitive but crucial elements of the civil service and public administration modernization reforms, such as public service streamlining the removal of ghost workers, the revision of the Government mandate and agency reengineering. This was already considered a very high risk during project appraisal. It materialized sooner than expected on the background of a softening of the fiscal constraint with the increase in oil prices, the resumption of tensions and conflict in the northeast and southern provinces and mounting resistance from vested interests affected by the reform. The limited ownership of the proposed reform implementation modalities, such as the development of core information systems, such as the complex EDB, BIS and AFMIS, which at the time were proposed in many Bank funded public administration reforms, with very limited success, according to IEG's evaluation in 1999. As mentioned, these complex and sophisticated solutions appeared to be inappropriate for the country's situation and capacity. The high-tech solutions, such as the biometrics system, were seen as a solution to overcome the prevailing governance and patronage issues and strong resistance to the civil service streamlining. It raised very high expectations, did not succeed in countering resistance and current practices and overshadowed more basic and fundamental HR organizational and management deficiencies, such as lack of organigrams and clear reporting chains, uneven enforcement of HR policies, lack of documentation, deficient HR and payroll data recording, reporting and verification processes. Until 2004, and the consideration of its cancellation, the ownership of AFMIS by the Ministry of Finance was also very limited, as evidenced by the important delays and disconnect with the Ministry's reforms. 9 The size and scope of the biometric system and employee database implementation was expanded during project implementation to include the military, judiciary and security sectors, and thereby almost doubling the scope of work. These additional sectors were later added through a presidential order in 2004. The significant expansion of the system both from a biometrics and employee database perspective greatly complicated implementation. To date data capturing of military and security forces is still incomplete. The authority and mandate of MOCSAI regarding human resource management was challenged by some Governorates and spending units. This is evidenced by the very uneven cooperation with MOCSAI and the limited contribution of HR data to the compulsory central repository. The law 43/2005 on employment, wages and salaries, reasserted in art. 20 the obligation to report all HR related administrative and financial data to MOCSAI's EDB, but with limited effect. HR and payroll data exchange is also limited among central ministries and within MOCSAI. This situation strongly undermined the civil service reform and the feasibility of the spported central HR information systems. The low capacity of the Yemeni public administration with overall poorly paid and performing staff lacking skills and motivation had a strong adverse impact on the project implementation and disbursements. Most activities, foreseen in the initial procurement plan slipped 2 years, in a five year project. Capacity was particularly low in IT expertise and project management, both of critical importance for this broad and complex project, as shown by the delays and difficulties experienced with the AFMIS and BIS contracts. The absence of a detailed and time bound action plan outlining the Strategic Framework for Civil Service Modernization adopted in 1998 negatively impacted the project implementation. At project level, while the PAD outlined the project's main activities, there was no specific, time-bound, and measurable CSM implementation plan. The project included an initial procurement plan, which soon turned out to be unrealistic and suffered a 2-3 years delay. AFMIS's initial implementation plan, was mainly focusing on the software development and lacked of realism. It was revised during the project restructuring, with more detailed steps and time bound verifiable outputs, which could be verified and tested by an external company. This situation affected negatively reform and project implementation. Inadequate overall project coordination within the Government due to overlapping mandates, high information asymmetry and fragmented implementation arrangements proved detrimental for a project with such a broad scope and strong inter-linkages between the different components/ activities. This weighted particularly on the civil service component, and the development of core information systems, which required a close coordination with the Ministry of Finance and the cooperation from the Governorates, spending units and the security Ministries. The latter participated only at a very late stage in BIS, HR data exchange remained very limited, and the coordination of civil service streamlining efforts and data reconciliation between MOCSAI and MoF was formalized only very recently with the creation of a joint committee chaired by the two ministers in march 2009. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization The project's M&E design was limited by the absence of a detailed results framework with appropriate objectively verifiable performance indicators, related to a reform action plan. The project did not provide for component specific and time bound implementation plans, which was essential given the project's scope and complexity. This was to be developed during the project inception phase but did not materialize. Thus, the contract specific implementation schedules could not be linked to a component and project implementation plan. Furthermore, in the absence of a detailed Government action plan and a monitoring system, the project implementation could not be linked to reform milestones. The absence of this results chain from reform to contract implementation, made it extremely difficult to implement the PAD's two tier M&E system, which foresaw: (i) the set up of a computerized monitoring system, supposed to track the implementation of key indicators in the logical framework and (ii) semi- 10 annual external performance audits of all project activities supposed to evaluate the performance of project management as well as output delivery and quality. The M&E implementation and utilization are considered highly insufficient, as evidenced by (i) the failure to establish a functioning computerized monitoring system by the end of the project, (ii) the absence of regular progress reports produced by the PIUs (only financial reports were submitted), and (iii) the absence of external performance audits. Only one "performance audit" was carried out by an independent consultant, in the context of a supervision mission in July 2003. This report made a very candid and realistic assessment of the reform and the project's weaknesses in light of the CSMP's poor implementation so far and the lessons learned from many similar unsatisfactory projects. It made strong recommendations, such as scaling down the AFMIS and capacity building components, strengthening the underlying PFM reforms, adopting a more incremental and participatory approach to agency reengineering, enhancing project supervision and dialogue both within the Bank and with the counterparts, strengthening Government and donor coordination. This report informed the high-level dialogue and supervision in 2004 and the project revision that followed. No further performance audit was undertaken. The project's M&E system was insufficient given the scope of the project, its numerous stakeholders, as well as the sensitivity and complexity of its activities. Lack of M&E is part of the reason why the project failed to promptly detect and prevent unnecessary delays and diversions in the various components. 2.4 Safeguard and Fiduciary Compliance (focusing on issues and their resolution, as applicable) Weak procurement capacity was an issue both in the administration and in the PIUs. It adversely affected the pace of project implementation, despite the preparation of tender documents before the project's effectiveness. Tender and contract approval was protracted, reflecting the cumbersome decision- making processes and limited capacity and efficiency of the civil service. The initial project procurement plan was delayed by 2 years. The contract to develop the AFMIS software was signed only end of 2004, one year prior to project end. Important deficiencies in procurement filing as well as the absence of assets control increased risks. An independent procurement review carried out in 2010 confirmed these deficiencies and noted insufficient awareness and compliance with procurement rules and procedures by program management as well as by the Bank's task teams. The project's financial management system was stringent and closely monitored by external auditors and by the Bank. The PIUs FM capacity has been built throughout the project implementation, in a cooperative and iterative process with the Bank. The project produced quarterly financial monitoring reports (FMRs), automatically generated the project's accounting software and reviewed by an external auditor, it submitted annual audit reports, of acceptable quality with reasonable delays (5 months). The issues identified by the auditors were closely monitored and most of them followed up, including by COCA, when relating to Government expenditures. Even though, the project experienced delays in payments it managed to honor all outstanding claims during the project closure phase. 2.5 Post-completion Operation/Next Phase The AFMIS component will be taken forward by a new Bank financed Public Finance Modernization Project, scheduled for Board discussion in December 2010. The project notably supports the expansion of AFMIS's scope, by including the following functions: Cash Management, Commitment Controls, Arrears Management, and Final Accounts. It also supports the expansion of its coverage to cover all the spending units in the country up to the level of Governorates and Districts. This should contribute to consolidate the impact and sustainability of AFMIS component's outcomes under the CSMP project. 11 The other components of the project (including the reforms through MoCSAI such as the biometrics, HRM and re-engineering) are, at the moment, not scheduled to receive any follow-up support. Other international and bilateral donors are expected to take forward some of the reforms with agencies such as the Customs Authority and COCA. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Relevance of objectives The project's development objective is overall relevant but overambitious. The objective to "assist government in the creation of sustained utilization of human and financial resources in the administration, by putting in place core personnel and financial management systems, establishing a mechanism to reduce the number of unqualified civil servants, and initiating a restructuring process in individual ministries, while upgrading the management capacity in the civil service", is relevant to address the persistent acute capacity constraints of the public administration and the civil service's severe quality, effectiveness and affordability issues, hampering the Government's development and reform efforts. However, these objectives were too ambitious for a fragile State like Yemen, challenged by political tensions and security threats, and dire economic and social conditions. The PDO did not sufficiently take into account: (i) the high sensitivity and long term of the reforms pursued, (ii) the country's political economy and governance situation, (iii) the administration's significant capacity constraints, or (iv) the lessons learned from the Bank's unsatisfactory experience in similar projects. While the sustainable management of human and financial resources was a very relevant objective, the means to achieve this were less, particularly the set up of new HR and financial management systems. The gradual improvement and modernization of existing processes and systems might have been more appropriate. The PDO appears excessively ambitious for a single investment project and the limited time frame foreseen, as evidenced by its limited outcomes and the two project extensions. The objectives seem more appropriate for the CAS 99' Governance and public sector reform pillar than for the CSMP, which was only one element of this comprehensive assistance strategy. The project was in line with the Government's Civil Service Modernization Strategy. However, the AFMIS component was less relevant for this strategy, which required rather the consolidation and modernization of the payroll system, necessary for an integrated HR information system as well as for the implementation of the new wage structure. The support to the payroll reform was foreseen in the early stages of project preparation but eventually scaled back, but the elaborate EDB and BIS were maintained. While streamlining the project was necessary, this arbitrage can be questioned and has adversely affected the effectiveness of the EDB and BIS, as detailed below. The project was consistent with the Bank's strategy and focus on public sector reforms at the time. In FY 98, more than 72% of the Bank's Country Assistance Strategies had a component on public sector reform, many of which contained civil service reform operations. Yemen's CAS 99 had a pillar focusing on public sector modernization, of which the CSMP was the flagship project, complemented by a programmatic series of public sector management adjustment credits, aimed at supporting the policy dimension of the reforms. However, it did not fully take into account the lessons learned from twenty years of, largely unsuccessful, support to civil service reforms, as highlighted in IEG's thematic evaluation in 1999. 12 Relevance of design The project design was overall relevant and reflected the CAS's optimism and holistic approach to public sector modernization. However, it proved to be overambitious and complex considering the reform implementation context and capacity. As mentioned under section 2.1, the project's design was too ambitious and complex both in terms of scope (5 components and 15 subcomponents involving directly more than 12 agencies) and sophistication (development of complex human resources and financial management IT systems, such as AFMIS, EDB, BIS). The holistic and integrated approach towards civil service and public sector reform reflected the views at the time and aimed to mitigate risks and maximize positive inter-linkages between project component and activities. However, these inter-linkages played against the project when the commitment towards downsizing faded and institutional resistance grew. They significantly increased the project's risks. This was evidenced by the disconnect between the new wage policy and increases with the civil service streamlining and reorganisation, reinforced by the delays in the establishment of an operational employee database and biometric information system. The core systems' (including AFMIS) lack of integration with the payroll, itself fragmented is another example of failed inter-linkages. The overambitious design with inadequate sequencing and phasing strained limited Government management and implementation capacity, limited sustainable outcomes and increased resistance from major constituencies affected by these numerous and simultaneous reforms. The project's coherence and relevance suffered also from the fact that AFMIS, was de-facto a stand-alone project, with its separate project implementation unit (PIU) reporting to the Ministry of Finance. AFMIS proved very demanding in terms of technical and management skills both for the Government and the Bank as it covered the entire budget process and required strong inter-ministerial coordination. Despite a long preparation phase and extensive stakeholder consultation, the complex and politically sensitive reforms supported by the project seem to have paid insufficient attention to the Country's political, economical, and social specificities, and overestimated the political clout and technical capacity of the still young and weak central Government. The sophisticated IT solutions proposed to build core information systems did not sufficiently take into account the existing formal and informal institutional set up and organizational arrangements in place. For instance, the complex and largely informal management and reporting chains, the lack of consistency, documentation and enforcement of existing human resources and financial management policies and processes, the important autonomy and influence exerted by regional and tribal constituencies and by the security sector were not sufficiently taken into account in the project design. The latter was added much later through a presidential executive order but compliance is uneven and slow, and biometric data capturing is still incomplete. The very decentralized, fragmented and heterogeneous human resources and payroll management system across 1400 spending units with uneven reporting lines to central ministries was not adequately incorporated in the project design and ultimately negatively weighted on the planned central HR management information systems. The country's socio-political and governance situation was insufficiently reflected in the project design and affected adversely the reform process and the project supporting it. For instance, the politically and socially very sensitive, civil service streamlining and agency restructuring reforms did not sufficiently take into account the civil service's welfare, cohesion and patronage functions, which created strong resistance against downsizing. As a result, no ghost worker has officially been identified and removed and most redundant workers identified and referred to the Civil Service Fund came from defunct State enterprises and not from the administration. The Bank's Quality Assessment of Lending Portfolio (QALP), in 2009, considered the design arrangements at entry as moderately satisfactory considering "the sound institutional knowledge of the Yemeni institutional environment as well as on its adjustment to complex and evolving external environment". However, it also noted that the design was too ambitious and the addition of the AFMIS 13 component inappropriate for a fragile, post-conflict and low capacity state like Yemen. Furthermore, the risk assessment and mitigating measures were considered insufficient, notably with regard to the strong involvement of other donors and the project's high dependence on policy reforms supported by the parallel Public Sector Management Adjustment Credit. The programmatic PSMAC series was seen as a means to support the political decisions and policy measures of the civil service reform, thus complementing the more technical and capacity building role of the CSMP. However, the Bank's internal Quality of Supervision Assessment (QSA7) rated the project's quality at entry unsatisfactory, based on a similar assessment and questioning the relevance of going ahead with such an ambitious and complex project in the Yemeni context. It found that the Government's commitment and significant ownership were not sufficient conditions for success of such a complex and sensitive project. Relevance of implementation The relevance of the project implementation framework suffered from the ambitious and complex project design as well as from phasing and sequencing issues. It did not adequately address the very demanding coordination required by this holistic reform approach, the high number of stakeholders involved and affected by these reforms as well as by the numerous donors supporting these reforms. The implementation framework reflected the large scope and complexity of the project design and exacerbated already limited management and coordination capacity within the administration. The decentralization of implementation responsibility to up to 12 agencies and the absence of a program coordinator with overall responsibility posed important coordination, reporting, and accountability challenges. The project's logical framework was not sufficiently detailed and operational to facilitate project implementation and monitoring. It was not built on a Government results framework, nor complemented by effective PIU monitoring systems. The numerous project subcomponents and activities as well as their strong inter-linkages and potential synergy warranted a much stronger project monitoring and evaluation system. As mentioned under section 2.3, the project's M&E was weak and insufficiently implemented. The reform timing and phasing proved to be an issue, as shown by the implementation of the new wage structure and increases, without a consolidated and accurate human resources and payroll information system. Likewise, the timing of the wage increases was suboptimal. While the strong social pressure and expectations can explain why the Government went ahead with the first general wage increase it had announced in its strategy, the second increase should have been linked to the agencies' reengineering and cooperation with the central HR database. This would have permitted to leverage these two essential and difficult reforms, which in turn could have generated savings partly offsetting the wage increase. Further, the introduction of complex IT solutions without prior institutional and organizational clarification and consolidation of HR processes and management is questionable and partly explains the persistent difficulties encountered by the EDB. The phasing and linkage between the personnel information and management systems and payroll was particularly problematic. The project focused attention on the development of an autonomous EDB and BIS within the Ministry of Civil Service, on which eventually an integrated and computerized payroll would be based. Expected project output for the payroll were limited to the development of new payroll practices and a plan for its computerization. Given the dire state of personnel record keeping and the strong decentralization of personnel and payroll management, the lack of integration of the payroll with the EDB and BIS from the start, strongly affects the credibility and usefulness of these new databases for civil service streamlining and management. At present the discrepancies between the payroll (400 000), the BIS (489 000) and the EDB (540 000) will make reconciliation and integration extremely challenging. The ICR would rate the project's overall relevance, from today's perspective, as unsatisfactory, based on the project design and preparations' strength and weaknesses specified under Section 2.1 and 3.1. 14 3.2 Achievement of Project Development Objectives At project outcome level: The project is rated moderately unsatisfactory. There were significant shortcomings in the achievement of the project's very ambitious development outcome objectives as well as in its efficiency. The project's contribution to the broad objective to assist the government in the creation of capacity, institutions, and systems for improved and sustained utilization of human and financial resources in Yemen's civil service was very limited, as evidenced by: (i) the increase in the civil service and the wage bill, (ii) the administration's continued severe capacity constraints, (iii) the limited progress in the institutional reengineering reform, (iv) the absence of an operational new human and financial resource management systems. The project efficiency was also limited considering the two restructuring and extensions, stretching its implementation over 10 years and the project high preparation and management costs, exceeding 17% of the budget4. Apart from the AFMIS, the sustainability of project outputs and outcome is subject to significant risks, such as (i) the limited implementation of the new civil service legal framework, (ii) limited progress of labor redundancies through the Civil Service Fund, (iii) limited implementation of re-engineering of ministries, and (iv) the non-completion of the facilitating processes and systems (EDB and data verification) that make the BIS non-operational. The PAD two outcome indicators reflect only partially the PDO. The first project outcome indicator, "the production of consolidated financial statements for budget execution in a timely manner", fails to capture the objective to create operational HR and financial management systems, which go beyond the timely production of consolidated financial statements and also include the HR systems (EDB BIS) representing a central element of the reform and the project. The second outcome indicator "employment based on positions and a transparent application of civil service law", fails to address the sustainable utilization of human and financial resources, which was the main rationale for the reform and project and as such a central part of its objective. For instance, this indicator does not capture the civil service streamlining and agency reengineering efforts. The capacity building dimension is not covered either. In order to address these limitations, the project's outcome assessment will also look at the PDO and the component specific outcomes, notably based on the more comprehensive results indicators included in the Government's development policy letter annexed to the PAD. The project's two above mentioned outcome indicators are rated moderately unsatisfactory. The first objective can be considered partially achieved, based on the last Public Expenditures and Financial Accountability (PEFA) report, published in June 2008, confirming that the Government prepares consolidated financial statements, on average 10 months after the budget closure, which is considered timely. However, it also notes, under indicator PI 25, that these statements do not include payment or tax arrears and are not based on the international COFOG/GFS standards. This should be the case once the new accounting and financial management information system is finalized and rolled out. The project's second objective cannot be considered as achieved, given the fact that: (i) the integrated and standard system of employment description and evaluation, required by art. 29 of the law n°43/2005 on employment, wages and salaries, is not yet fully in place, and (ii) the core human resource information 4 Project operating costs represented US$ 2.74 m and WB management cost (BB) reached US$ 2.41 m. 15 systems, which would enable to correlate staff and positions and ensure a consistent and transparent implementation of civil service law are not yet operational. The civil service legal and regulatory framework has been substantially revised, with the adoption of the new national salary strategy, the law n°43/2005 and its numerous bylaws. This comprehensive law is a cornerstone of the civil service modernization reform, introducing notably a simplified and modernized wage structure, consolidated allowances, increased wages and a link to performance, some wage decompression, and providing for a more integrated and harmonized human resource management system, linked with the payroll and employee database. However, its full and effective implementation is very demanding and challenging, considering the following important institutional and organizational prerequisites. For instance, it requires the set up of specific technical units in all agencies and governorates to ensure the correct transfer of staff to the new wage structure, the preparation and adoption of integrated standard systems of job profiles and evaluation, directories and head counts for each of the 1400 spending units, integration of the payroll. Furthermore, the law obliges, in its art. 20, all spending units to regularly update EDB with accurate data, which is necessary for the effective and transparent implementation of this law. Compliance and data accuracy are still very uneven. In sum, most of these prerequisites are not yet in place, limiting greatly the effectiveness and the expected benefits of the reform. The holistic approach to civil service modernization adopted in the Strategy aimed at reducing the wage bill by compensating the wage increases with savings form streamlined and efficiency gains. However, compliance with the removal of all double dippers and ghost workers (art. 19) is partial. The restructuring process has largely remained limited to studies with little implementation, which was initially foreseen for the second phase. Most studies did not translate in restructuring plans and the second phase did not materialize. Thus, the expected savings and efficiency gains from the reengineering process remained very limited. Likewise, the biometrics system is not yet operational, further hampering the project's objective to identify and subsequently remove all double dippers and ghost workers. Finally, the envisioned linkage between the salary reform and restructuring has not been adhered to in practice and wage increases were not conditioned by savings accomplished through re-engineering. The former proceeded independently from the latter. Wage and allowances increases contributed to a near doubling of the wage bill, from YR 281 billion in 2005 to YR 597 billion in 2010i, with savings estimated at YR 100 billion over the same period. Thus, the original aim of the reform and project to address the unsustainable wage bill was not achieved. In order to assess the project's effectiveness and complement the limited outcome indicators at the component level, it seems important to also consider the Government's result indicators and targets mentioned in its policy letter, annexed to the PAD. Result indicators/ targets Achievements / Results 1. Removal of 12 000 overdue pensioners 28 097 employees were removed, between the tightening by December 2000; of retirement rules and the Wage strategy in 2006 and 2009. Official statistics do not differentiate between regular and past-due retirements. 2. Removal from the payroll of all double Out of the estimated 60 000 double dippers publicly dippers and ghost workers within 4 announced, only 3792 were removed, of which 25 month of their identification through the identified by the incomplete biometrics system. No ghost biometrics system; worker has been identified and removed in official records. 3. Transfer to the Civil Service Fund of End of 2009, 17 753 redundant workers were processed. workers identified through the The agency restructuring process has yielded limited restructuring process results. 4. Significant overall reduction in the The number of civil servants, excluding military and 16 number of civil servants over the project security personnel, increased from 352,000 in 1999 to period 490,000 in 2009. 5. Adoption of a Wage and employment The National Wage Strategy and the employment, wage policy by June 2001, that is fiscally and salary bill were adopted in 2005 and contributed to affordable and that permits the gradual the near doubling of the wage bill and a slightly decompression of wages; improved decompression (from 3 to 4) 6. Restructuring of 4 agencies by December Restructuring has started at MOCSAI and the Customs 2000 and 4 more by September 2001; authority. None is completed nor has referred workers to the Civil Service Fund. 7. Introduction of revised, computerized The EDB is incomplete and not up to date; BIS is personnel management, payroll and incomplete and data only partially verified. Payroll AFMIS systems in Sana'a and gradual remains fragmented and heterogeneous; AFMIS is used expansion to all administrative units for budget preparation since 2008 and for budget execution in four pilot ministries. 8. Design and carry out of training Numerous training have been provided, two have been programs for personnel in the institutionalized at the University of Sana'a. restructured departments. Out of the eight performance indicators above, six warrant negative ratings at project end. Only two objective were somewhat achieved and can be rated moderately satisfactory (AFMIS and training). This outcome is slightly higher than IEG's assessment in its 2006 Country assistance evaluation, prior to the substantial wage increases. The Governance pillar of the two CAS (99-05), of which this project was the cornerstone, was rated unsatisfactory, considering that "little progress has been achieved, public sector employment has continued to increase, notwithstanding overstaffing and an unaffordable wage bill; pay reform and restructuring of key ministries have been delayed; service delivery standard remain low and financial management has improved slowly". While most of these findings remain valid, progress has been made on the Accounting and Financial Management Information System (AFMIS). At component outcome level, based on outcome and outputs detailed in annex 2: 1. The Core Systems component is rated moderately unsatisfactory, considering the uneven and incomplete achievements of AFMIS, EDB and BIS. The component's objective was to have core systems improved, measured by operating manuals written and approved by cabinet resolution and computerized systems in place where appropriate. Unlike the civil service component, AFMIS did not benefit from an overall public finance management reform strategy or plan at project appraisal, and as such suffered, at least initially, from limited visibility and appropriation by the Ministry of Finance. Despite a late and difficult start, the AFMIS sub-component has made significant progress since 2004. Its operating manuals have been adopted. By project closure a computerized system with the core modules has been developed, tested, and rolled-out to the four envisioned pilot ministries (Finance, Education, Health Care and Population, and General Operations and Works). In addition, the project has started the roll-out to five additional institutions: Customs Authority, Tax Authority, Ministry of Water and Environment, Ministry of Oil and Minerals and the Ministry of Electricity. The project developed the following modules: (i) budget preparation, (ii) budget execution, (iii) account payable, (iv) fixed assets, (v) revenue, (vii) final accounts. The core modules (budget preparation, execution (safe commitment and procurement controls, and final accounts) have been finalized and are currently pilot tested in the Ministry of Finance and the above-mentioned three line ministries. The budget preparation module is used by the Ministry of Finance since 2007, for the 2008 budget, although based on the uploading of budget proposals from spending units in the form excel spreadsheets, as the module has not yet been rolled 17 out. The budget execution and the final accounts modules are pilot tested since 2008 in all four pilot ministries, representing a substantial part of the budget (60%). The other modules (accounts payable, fixed assets and revenue) are currently being finalized, based on the external verification done by KPMG and will then be tested in the four pilot ministries before being rolled out. The data exchange module with other systems, notably HR and payroll, needs further important work and cooperation with the other agencies. The cash management module has been postponed, in the absence of a treasury function at the Ministry of Finance, despite recommendations since the beginning of the project that this function, presently carried out by the Central Bank be transferred to the Ministry of Finance, in accordance with international practices. Likewise, the commitment control module has been postponed to the second phase and will be covered by the follow up project. The system does not currently support the verification of intended commitments against available resources. In sum, the three core modules have been successfully developed and are currently being implemented in pilot ministries. Three out of the nine modules are been further finalized before being pilot tested. And three modules have not been developed yet. The further development and rolling out of AFMIS will be supported by the upcoming public finance modernization project. The development of the civil service core information systems is still faced with concerns and challenges. Even though the operation manuals have been adopted, as evidenced by the resolution of council of Ministers n° 18/2008, the computerized systems are not yet operational. The Employee Database (EDB) has been developed during project preparation, based on the review and consolidation of personnel files in 1998 but its maintenance has not been directly supported by the project. To date it is not fully operational for the following reasons: (i) the database is incomplete and not up to date. Some personnel records are incomplete and missing basic information, such as employees' age, length of service, unit of employment or pension contributions. It has not been consistently up dated and many personnel changes (departure, transfer, retirement) and changes to personnel details (life events/ dependants,...) have not been recorded, as evidenced notably by the much high number of entries than the BIS and payroll databases. This limits greatly the accuracy and usefulness of data. (ii) EDB is not fully connected / integrated with the human resources databases for the Military, the Security sector and the insurance sector (retirement), nor with the fragmented and decentralized payroll applications, despite the provisions of the civil service law 43/2005. Notwithstanding important efforts to improve data collection and reporting, through the set up by MOCSAI of 22 Regional Information Centers (RICs), HR data ownership and reporting deficiencies from spending units remains an important issue. A clear, standardized and documented process for HR and payroll data creation, collection, reporting, updating and control still needs to be put in place and effectively and consistently enforced. This should have been a priority for the project. Likewise, the establishment of a biometric information System (BIS) is well advanced but not yet operational. A significant work program remains for the Government to operationalize and institutionalize the BIS within its HR operations. First of all, the capturing of biometrics data is not complete yet for the military and security forces, with both showing a completion ratio around 80%. Second, only 50% of the captured data has been uploaded into the biometrics system, and the rest remains stored on temporary files (computers, CDs,...). The main bottleneck in loading and printing ID cards is the linkage between BIS and the EDB. The EDB plays a critical role as it would provide demographic and administrative data (such as name, function and location of work) to the BIS to be printed on ID cards. However, as mentioned above, details of changes to employee data have not been systematically recorded in EDB. Data is submitted to the "central" HR database in paper form and most of the paper is simply stored, without the details being entered into the database. In essence, what this approach has resulted in is an obsolete data set, which compromises the integrity and accuracy of data in EDB. Therefore, before EDB be used and before any new fingerprint ID card can be printed from BIS, EDB must be up dated and its accuracy reviewed and ensured, to prevent printing cards for people that no longer work in the civil service or contains inaccurate information. 18 The project to date does not have in place adequate plans and has not secured funding to undertake this revision and completion. Nevertheless, the process of capturing biometric data had an incidental positive effect as many civil servants holding several positions came forward voluntarily and relinquished one position fearing of being caught by the new system. The project put little emphasis on the payroll system consolidation and modernization. Despite MOCSAI's efforts to develop a uniform MS Access based Payroll database, presently tested in 10 sites, the current Payroll system is still fragmented, not harmonized. It suffers from a wide heterogeneity of applications, ranging from sophisticated Oracle database at the Ministries of Education and Oil to paper ledgers in remote spending units. It is fragmented among the 1400 spending units, which report only line items to the Ministry of Finance and with uneven consistency to their sectoral ministries and MOCSAI. As mentioned above, this situation weights negatively on the establishment of a comprehensive and accurate HR management information system. Progress has been made on the modernization of the civil service legal and regulatory framework, with the adoption of a new law (No 43/2005) on employment, wages and salaries system, as well as its numerous bylaws on: (i) transfers to the three employment scales, (ii) rules for allowances, (iii) more transparent recruitment and employment procedures, (iv) the set up of a Civil Service Fund, to process redundant employees, and (v) the establishment of civil service directories. This numerous legislation is detailed in annex 2. However, as mentioned above, its enforcement is only partial. 2. The second component, Civil Service Fund, aiming at rationalizing the size and capacity of the civil service is rated unsatisfactory, considering the component's objective "a workforce streamlined and becoming more efficient", and the evaluation criteria foreseen in the PAD: "the reduction of the number of civil servants on the payroll and the retention of high quality workers". The reform and the project did not achieve the objective, as evidenced by the important increase (24%) in the number of civil service employees on the wage bill, raising form 330 000 in 1998 to 410 000 in the budget 2010. Likewise, the payroll increased steadily during the last decade, with a 26% increase in 2007, after the introduction of the new wage bill. However, the bill did not enhance the civil service effectiveness or attractiveness for high quality workers as it did not improve the prospects of wage progression (wage compression of 1 to 4), resolve the problem of patronage, lack of transparency and uneven application of the civil service law (no ghost worker has been officially removed) or introduce performance orientation/ compensation. Most structural weaknesses of the civil service remain, with few but notable exceptions, such as MOCSAI's transparent and competitive selection of its management. It has yet to be replicated elsewhere and lead to a performance based framework. The National Wage and Salaries Strategy, adopted in 2005, was partially implemented, resulting in an important increase in the wage bill, which was already not sustainable, without correcting its structural weaknesses nor improving the efficiency of the civil service. The strategy provided for (i) an important wage increase (between 20-40% across the board), (ii) a revised employment, and wage scale, only partially addressing the compression problem, and (iii) a revision of the numerous allowances. However, it did not fully take into account the major recommendations of the National Salary Strategy Report prepared under the project, such as the move from an education to a job centered pay structure, a greater pay progression to enhance motivation, undertaking a unit based headcount and EDB update before the transfer to the new scale, and delinking, initially, the increased wages from pension benefits to avoid important deficits. Furthermore, the report suggested introducing the Full Time Equivalent as HR unit, in order to take into account the limited working hours in the Yemeni civil service (5/ day on average) and increase flexibility in HR management and restructuring. This proved too complex for the existing administrative structures and monitoring systems. Due to the overall wage increase, the wage compression improved only slightly, from 3-1 to 4-1 and thus did not really address the lack of incentives from low wage progression. The allowances, 19 initially 65, have been significantly consolidated in a common work nature allowance and the general allowances have been integrated into the basic salary. Rules and procedures for allowances have been clarified, making their attribution more transparent. But, this reform has not led to a reduction in the amount of allowances, even if their relative share to the basic salary has decreased due the wage increase. Regional hardship allowances remain insufficient to attract competent civil servants in remote areas. Technical assistance was provided for the development of standard job descriptions, but they have not yet been implemented. Thus, salaries remain solely based on administrative grades and education levels, which is not conducive to performance evaluation and compensation. Even if initial savings could be made by referring employees to the pension and the civil service funds, these savings remained modest due to the delayed restructuring agenda and limited results in removing double dippers and ghost workers from the payroll. They only partly offset the substantial wage bill increases in 2007 and 2008, estimated respectively at +26% and +15%ii. Some results have been achieved with the enforcement of retirement rules. From 2006 to 2009, 28 097 employees, who had served past retirement age, were retired. This improvement is the result of MOCSAI's census and identification work, as well as of the strengthening and enforcement of existing retirement rules. A Cabinet decree (211/2007) and a circular from the Prime Minister were passed to specify actions to be taken to enforce retirement in the public service. Furthermore, since 2000 a total of 44 703 civil servants have retired, on average 4500 persons per year. However, this number represents merely 0.5% of the total civil service and is well below the normal retirement ratio of 3% noted in other civil services with similar population age. This could be partly due to weak information in existing HR databases (an estimated 7% of EDB population has no birth/starting date) and probably to an uneven application of the rules. The removal from payroll of ghost workers and double dippers has achieved modest results. Despite initial public announcements of 60 000 double dippers and ghost workers in the civil service, at the beginning of the reform, only 3792 double dippers were removed from payroll as of end 2009. From this total number 3767 employees came forward on a voluntary basis to relinquish one of their jobs and only 25 double dippers were identified based on the HR databases. No ghost workers have been identified and removed from payroll in official records. These results were validated by a joint working group co-chaired by the Minister of Finance and Civil Service to evaluate the benefits of the reform. A similar exercise is underway to complement the 2009 and 2010 figures. The contribution of the Civil Service Fund (CFS) to the streamlining reform was effective but modest, notably constrained by limited referrals and financial resources. The CSF was established with the enactment of Law 1/1999 and Law 314/2000 and was gradually set up, with branches in four Governorates. However, due to legal intricacies with the existing legal and regulatory civil service framework and delays in its organizational set up, the Fund became operational only by the end of 2004. It processed 14,127 redundant staff, notably from defunct or privatized state enterprises, during Phase I from April 2006 to March 2009. 2,250 employees received their statutory pension, 1 100 were redeployed to other public services and 10,777 taken care of by the CSF. Among these, 93.7% opted for early retirement and were transferred to the pension authority, and 6.3% opted for severance packages (10,300 USD av.). The total cost to the Fund was US$ 17.8m. However, as no budget allocation was foreseen to offset the resulting pension contribution deficit, exacerbated by the wage increase, the Pension Authority refused new referrals. New redundant workers are expected to be processed in 2011, based on budget availability. 3. Under the project's third component on Government agency reengineering many studies have been completed, but there is little evidence of implementation. Thus, in light of the stated objective "to streamline operating procedures in selected agencies" and performance indicator "workflow delivery improved against service standards, in restructured agencies", the results are considered 20 unsatisfactory, as there is very little evidence of outcome in terms of restructured agencies or improved workflows/ service delivery. However, it must be noted that the project' s activities focused on the reengineering studies and that their implementation was supposed to be supported during a follow up project, which never materialized. Despite the project's two extensions, no support has been provided for the implementation of approved studies. A study was conducted in 2006, to review the Government's mandate in order to clarify the responsibilities of ministries and main agencies, identify overlaps, and propose measures to rationalize and streamline ministries and agencies. However, after completion of the study at the end of 2006, there was no evidence that the Government approved or started implementing this inherently political and sensitive study. A study on the introduction of standard job classification in the civil service was conducted in 2006. A Cabinet order n°146/2009 regarding the new job classification has just been adopted and needs to be implemented. Further, a related study was conducted to develop performance indicators, trying to link them directly to the mission, the vision and the strategic goals of nine ministries and agencies. However, the final report raised important quality issues and lacked specific and operational recommendations. It was thus not implemented. Over the course of the project, approximately US$ 4.17 million have been spent on a number of "re- engineering studies" of individual government agencies and departments. The methodology followed to implement these studies aimed to increase local knowledge and appropriation. Most studies were conducted in close collaboration with focal groups of officials set up in the respective agencies. Nevertheless, the studies did not always benefit from the necessary involvement and commitment from the agency leadership and thus did not translate in internal reform action plans. There are however some exceptions, like COCA and the Custom authority, were the studies appear to have played a useful catalytic role in crystallizing reform initiatives that were developed by agency staff. A restructuring plan is being elaborated in both agencies. However, apart from MOCSAI, who has started a formal reorganization with the adoption of Republican resolution No 235/2007, and the President resolution n°5/2010 on internal audit, there is limited evidence of the project funded studies' impact on the restructuring and streamlining of Government ministries and agencies. Anecdotal information suggests that the Customs authority started restructuring, supported bilaterally by Jordan. However, contrary to the initial objective, none of these agencies have referred redundant workers to the Civil Service Fund, which seems to confirm the limited restructuring and streamlining of their workforce, at least at this stage. This component included an innovation fund aimed at providing (i) short term, innovative and visible benefits to the citizens to sustain the reform momentum, (ii) to test the idea of creating a common fund to channel donor financing for agency restructuring, and (iii) to leave room for changing circumstances and opportunistically capture reform initiatives by other agencies. Based on the limited information available these objectives do not seem to have been achieved for the following reasons. First, it took seven years until six projects were selected and approved and between one to two years to implement them, which does not qualify as "quick fixes". Second, while most selected projects were geared towards the citizens, their degree of innovation varied. For instance, the set up of a help desk at MOCSAI or an information center at the Airport, can be considered innovative compared to existing administrative practices, the installation of three ATMs for the postal office or medical equipment for the Republican Hospital's emergency services might be less innovative. Third, in the absence of any information on these projects' outcomes, it is not possible to assess the effectiveness of the intended benefits to citizens. Fourth, as the innovation fund has been entirely managed by an external consulting company with very limited interaction with the reform teams, it does not seem to have informed the reform process or support its acceptance/ momentum, as intended. For instance, it has not contributed to the set up of a common fund to channel donor financing for agency restructuring, which does not exist. 21 4. The capacity building component can be rated moderately unsatisfactory, as its objective "to enhance the capacity to sustain reforms and manage the public administration", was only partially achieved based on the project's indicator : "an effective program plan for follow-on reforms developed. Civil service training program implemented with agreed number of civil service servants trained in management of public administration and specialist work skills for operating core systems". Even though an important number of training activities were delivered, there is no evidence of comprehensive qualitative assessment or impact evaluation. The quantitative information available indicates that a total of 27 training programs have been delivered to approximately 1,500 civil servants, including 5 management training programs for approximately 600 senior Government managers. The Dutch Trust Fund financed also the development and set up of a master for business administration (MBA) and a master for public administration (MPA) at Sana'a University. The MBA and MPA, were part of the sample of the Trust Fund's final external evaluation, finalized on January 28, 2008. While providing interesting insights and lessons learned, the evaluation's contribution to this assessment is limited as it covers only two projects. The evaluation noted notably that the training project's relevance depended strongly on the degree of local ownership, which was overall limited, but better for the two courses set up at Sana'a University. Further, it found the MBA effective both in terms of institutional development and human capacity building, while the MPA less effective in institutional development. It also highlighted the generally negative impact of brain drain on the effectiveness of such training programs. 3.3 Efficiency The above-mentioned efficiency criteria are not applicable for an institutional capacity building project like this one. Nevertheless, the project's efficiency is limited considering (i) the limited output and outcome specified above and in annex 2, (ii) the two restructuring and extensions, stretching its implementation from 5 to 10 years and (iii) the project's high preparation and management costs, exceeding 17% of the budget. The addition of the AFMIS component required the set up of a second and separate PIU, thus increasing management and supervision costs. WB management and supervision cost (BB budget) for this project totaled US$2.41 m or 8% of the total project budget. 3.4 Justification of Overall Outcome Rating Rating: Moderately unsatisfactory, based on the assessments under sections 3.1, 3.2 and 3.3 and particularly considering the AFMIS outcome. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development: N/A (b) Institutional Change/Strengthening The AFMIS resulted in important institutional changes, reinforced by the technical assistance on the systems' conceptual design. The iterative development of the AFMIS application and its conceptual design was an opportunity to review, clarify, and specify the responsibilities and functions across the budget cycle both at the Ministry of Finance as well as in the four major pilot Ministries (Education, Health care and Population, and General Operations and Works). It highlighted the functions that need to be further developed, such as treasury and cash management, debt and project management, etc....). Even if the work on restructuring has not yielded many benefits at this stage, some of the reengineering studies were accepted and applied by the corresponding ministries. For instance MOCSAI revised and clarified its organization with the adoption of Republican resolution No 235/2007 and undertook a competitive external selection of its top management. 22 Training efforts supported by the project and the Dutch Trust Fund, both locally and internationally, have contributed to build expertise and long-term capacity at the country level, as evidenced by the MBA and MPA introduced at Sana'a University. (c) Other Unintended Outcomes and Impacts (positive or negative) Positive: As most service contracts under the project were awarded to local companies (AFMIS software development trainings, ....), there were positive externalities in terms of capacity building and transfer of know how to local consulting firms. Negative: The project was instrumental in the development of the Civil service modernization reform and the National Wage and Salaries Strategy adopted in 2005. However, as this strategy has been only partially implemented, it further increased the un-sustainable wage bill without addressing its structural weaknesses. The project's high visibility and strong ambitions in terms of civil service streamlining and removal of double dippers and ghost workers, has substantially exposed the project and the reform champion to fierce internal resistance and opposition, as evidenced by the removal of the Minister of Civil Service, at the beginning of the project. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops: N/A 4. Assessment of Risk to Development Outcome Rating: significant In line with the IEG methodology, this risks assessment combines their impact and their likeliness of occurence. The overall sustainability risk to the project development outcome can be considered significant and not high, because the impact of the below specified risks is commensurate with the modest project outcome even if their likeliness of occurrence is high. Especially the wider civil service reforms and the biometrics system are at risk. The BIS needs to be fully institutionalized and adequately maintained for the sustainability of data collection, storage and updating, as well as for future card printing. Furthermore, the AFMIS, which is the project's main outcome will receive continued support under the forthcoming public finance modernization project is less exposed to risks. In the civil service component, most risks have already materialized and translated in very limited outcomes. The few remaining outcomes, such as the reengineering of MOCSAI or Customs or the new employment structure and payment scale introduced by law N°43/2005 are thus less exposed to risks. At the time of this evaluation the main risk to the project development outcome are the following: The political and security risks are high, considering the fragile truce with the northern tribes, the increased discontent and secessionist aspirations of the southern communities and the increased terrorist threat from Al Qaida of the Arabic Peninsula (AQAP). This risk is further increased by the prospect of upcoming presidential elections in such a tense situation. The governance risk is high, as highlighted in the country's Governance indicators and CPIA and Transparency international ratings. The prevalent corruption and patronage, notably in public finance management and civil service appointments are the biggest risk to the project's development outcome. The risk of reduced government ownership and commitment is considered significant but not high, mainly because the AFMIS seems to have reached a critical mass of support. It is presently 23 rolled out to core ministries representing more than 60% of the budget. As most risks affecting the civil service component have already materialized, outstanding sustainability risks are limited. The risk of lack of institutional support is considered moderate to significant for the same reasons as above. It might, however, be more important for the accounting and control dimensions of AFMIS given the high number of civil servants involved/ affected, notably in the Governorates and their potentially diverging interests. The economic risk is considered moderate. The likelihood that the current fiscal crisis has strong adverse economic consequences is high. The impact of worsening fiscal and economic conditions on the project's main output, AFMIS, is unclear, as a difficult fiscal situation might also strengthen the political will and public acceptance of structural public finance management reforms. The social risk is significant, considering the limited employment and revenue generation opportunities, at least in the short and medium term, until the social safety net and private sector reforms underway bear some fruits. In the meantime the pressure will remain strong on the current welfare and income distribution functions of the civil service and of public expenditures. The financial risk is significant, because of the partial and fragile outcome of the civil service component. The continued financing, from scarce budget resources, of the biometric information system's completion and maintenance represents a significant risk. The overall technical risk is significant. It is high for the finalisation and maintenance of the BIS and EDB, considering the very limited IT capacity in MOCSAI (staff reduced from 10 to 4) and the dismantling of PIU and contractor teams. It is moderate for AFMIS, given the strong ownership and continuation of support. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry rating: Moderately unsatisfactory The project preparation was thorough (3 years), timely (along with CAS and WB/ IMF CS budget supports) and in line with the Bank's focus on governance and civil service reforms at the time. It was consistent with the Government's strategy, supported by an earlier IDF grant. As noted above, the project was too ambitious and complex for a low income, low capacity, fragile state like Yemen, plagued by weak governance and severe security issues. It did not sufficiently take into account the existing formal and informal institutional and organizational set up prevailing at the time. The risks and lessons learned were reviewed in the PAD but not sufficiently reflected in the project design, as they would have warranted for a more prudent, gradual and sequenced approach. The project design also suffered from weaknesses in the timing and phasing of certain reforms/ activities, such as the introduction of new human resource policies, while developing HR management information systems and processes and conducting reengineering studies. Likewise, the introduction of complex IT solutions, somewhat disconnected from the country's realities, and without prior improvement of the basic HR processes and management proved to be questionable. The integration of the AFMIS component, which was a project of its own, substantially added to the overall complexity and strained limited management, coordination, and monitoring capacity. The results framework was too succinct and project specific, in the absence of a Government reform implementation and monitoring framework. The project's two outcome indicators, "the production of consolidated financial statements for budget execution in a timely manner, and employment based on positions and a transparent application of civil service law", only partially reflect the PDO, as they do not 24 capture the objective to create operational HR (EDB and BIS) and financial management systems nor the objective of sustainable utilization of human and financial resources (streamlining and restructuring). The Bank's internal quality assessment was mixed. In its Quality of Supervision Assessment (QSA7) the project's quality at entry was rated unsatisfactory. Whereas, the 2008 Quality Assessment of Lending Portfolio (QALP), rated the project's quality at entry as moderately satisfactory. (b) Quality of Supervision rating: Unsatisfactory Overall, the Bank has invested heavily in project supervision (total of US$2.4 m), with mixed results: Despite important efforts to get the project back on track, the project revision failed to fully integrate the lessons learned from the problematic project implementation and to streamline the project accordingly. An external review in July 2003, highlighted the project's over ambition, great complexity and lack of integration with the political and social realities. It recommended "to consider substantial modifications of the over-ambitious problem components, AFMIS, human-resource management, and training". The Bank considered canceling the AFMIS component, but the sophisticated and challenging EDB and BIS were not questioned. The training component was later transferred to a Dutch trust fund financing training capacity building in Yemen. Likewise, the project results framework and M&E, whose weaknesses were acknowledged in the 2004 management review meetings, was not strengthened. Nor was the PDO revised. The frequency of supervision and reporting varied. Supervision missions and reporting (ISR) took place on a regular basis until the first project restructuring was completed, in the middle of 2005 and the project showed initial improvements. ISR reporting became much less regular until end of 2009, with lags of up to 1.5 years between two ISR, although there were supervision missions in between (ie September 2005). There is little evidence of management comments on the ISR. The missions' skills mix was overall appropriate for the projects' scope and complexity, considering existing expertise and resource constraints. Most missions included the following expertise: public sector, AFMIS, financial management and procurement. In July 2009, an external consultant was recruited for a thorough assessment of the core human resources information systems supported by the project. Testing/ validation of AFMIS was entrusted to KPMG, who undertook 4 field visits between 2005 and 2009. These two missions provided very specific information on the core systems' progress or lack thereof. The candor and quality of ISRs and aide memoires varied greatly between the different teams and did not always reflect the accurate status of project implementation and results. Information collected during supervision missions was not always appropriately verified and assertions about project performance and outputs were sometimes too optimistic and unsubstantiated. Date ISR No. DO IP Archived 1 6/17/2000 Satisfactory Satisfactory 2 2/28/2001 Satisfactory Satisfactory 3 8/02/2001 Satisfactory Satisfactory 4 8/03/2001 Satisfactory Satisfactory 5 04/10/2002 Satisfactory Satisfactory 6 7/17/2002 Satisfactory Satisfactory 7 12/11/2002 Satisfactory Satisfactory 8 12/19/2002 Satisfactory Unsatisfactory 25 9 6/16/2003 Satisfactory Unsatisfactory 10 12/12/2003 Satisfactory Unsatisfactory 11 6/30/2004 Satisfactory Unsatisfactory 12 12/27/2004 Satisfactory Unsatisfactory 13 05/04/2005 Satisfactory Moderately Satisfactory 14 06/26/2006 Satisfactory Moderately Satisfactory 15 06/22/2007 Satisfactory Satisfactory 16 12/03/2007 Satisfactory Satisfactory 17 09/23/2008 Satisfactory Satisfactory 18 02/02/2010 Moderately Satisfactory Moderately Satisfactory 19 07/04/2010 Moderately Unsatisfactory Unsatisfactory The complexity and broad scope of the project, the difficult working environment characterized by strong information asymmetry and security issues, and the absence of an effective monitoring and evaluation system in place, made the supervision of this project particularly challenging. The very high turnover of project TTLs (7 during the project life) and teams weighted negatively on the consistency, effectiveness, and quality of project supervision. The project revision and important reallocations from the civil service and training component to the core IT systems has not resulted in a revision of PDO and performance indicators. Excessive attention seems to have been devoted to the development of core systems, whose budget allocation rose from 50% to 75%. It was to the detriment of the more difficult but necessary institution building and revision of processes. Too little attention has also been devoted to the employee database despite the fact that it is the core HR system and indispensable for an operational BIS. The Bank internal Quality of Supervision Assessment (QSA7) rated the Bank supervision overall moderately satisfactory, with the candor and quality of ISR rated moderately unsatisfactory. The QALP 2008 rated the quality of bank supervision as moderately unsatisfactory. The ICR shares this assessment, based on the missed opportunity to considerably streamline the project after the mid-term review in 2003, the lack of continuity in project management and weaknesses in reporting, while taking into account the challenges and the efforts made to turn the project around in 2004. However, the Borrower's negative assessment of the Bank's performance and attitude in the attached self-assessment, warrants and overall unsatisfactory rating (c) Justification of Rating for Overall Bank Performance rating: Unsatisfactory The Bank's overall performance is rated unsatisfactory, based on the above-mentioned weaknesses both in terms of quality at entry and supervision, while considering the important governance and security challenges, which strongly affected this highly sensitive reform and project. The Bank showed important efforts and commitment, at the highest level, during the two project revisions in 2004/2005 and 2007, with some success, as evidenced by the turnaround of the AFMIS component, the project's main outcome. It is however questionable, why an additional financing of US$ 14 million was processed and approved in April 2008, given the overall poor track record of project implementation so far, the difficult governance and security environment, and the limited implementation time left (2 years). The project's past disbursement rate did not support the expectation that the substantial additional funds could be disbursed within the last two years (the additional financing assumed that this 8 year-old project could multiply 11 26 times the disbursements made in the last nine month). The additional financing never reached effectiveness creating unnecessary frustration on both sides. 5.2 Borrower Performance (a) Government Performance rating: Moderately unsatisfactory The Government showed strong commitment and involvement at project preparation and design stage, particularly on the civil service component. It enacted a comprehensive and ambitious Strategic Framework for Civil Service Modernization, on which the project was based. This process was very participatory, culminating with a National Conference, opened by the President in 1998. However, due to various internal and external reasons mentioned above, the commitment was not maintained on the most difficult and sensitive aspects of the civil service reform, particularly on the streamlining and restructuring component, which were only partially implemented. The implementation of the National Wage Strategy, prior to the restructuring of government agencies and the streamlining of the civil service, led to an important increase in the wage bill, contrary to the reform's objective. During the first four years of project implementation, the AFMIS component made very little progress and was poorly managed. This situation resulted in an official request from the Bank to change the project manager and take critical mitigating measures to save this component, which the Government did, with some success, as evidenced by the positive development of the AFMIS component. Despite Government efforts and an elaborated project oversight and implementation framework, coordination and reform phasing issues adversely affected the project. For instance, the AFMIS component proceeded with hardware acquisition before the identification of critical PFM reform actions and prior to the conceptual design of the system. The human resources information systems and databases supported under the project still suffers from coordination, cooperation, and integration issues both at central government level and with local spending units. The set up in March 2009 of a joint working group co-chaired by the Minister of Finance and the Minister of Civil Service and Insurance, was a welcome step to enhance cooperation and data exchange, notably on civil service streamlining and corresponding adjustment to the payroll. Such a coordination mechanism should have been in place since the beginning of the project. This enhanced cooperation was particularly necessary to address the lack of integration between the Government's HR and financial information system. Strong capacity constraints in complex project management and in very specialized IT skills, negatively affected the Government's performance. The planned capacity building and training did not sufficiently address this issue. The reform led, with the support of the project, to a substantial revision of the Government's civil service legal and regulatory framework. More than 30 laws, regulations and Council resolutions were passed. The challenge now lies in their effective, consistent, and transparent implementation. (b) Implementing Agency or Agencies Performance rating: Moderately Unsatisfactory Until the change of the program manager of the AFMIS component in 2005, the performance of the PIU was poor and led to significant implementation and disbursement delays. Recruitment of PIU staff and procurement did not always comply with Bank procedures. The situation improved with the new management team and technical assistance to the PIU. The overall Civil Service PIU's performance was better but uneven across components and not sustained over time. Performance in terms of management, supervision, and reporting was better for the core 27 systems and streamlining (Civil Service Fund) than for the reengineering, innovation fund and capacity building components, as evidenced by the limited information provided by the project on the status and outcome of these activities. Lack of pro-active problem resolution caused undue delays, as evidenced by the near paralysis of biometrics data processing for almost two years, when new ID cards had to be purchased for quality reasons. The project implementation suffered from considerable procurement delays at the beginning. The initial procurement plan was delayed by two years, and some contracts, like for the BIS were signed only one year before the initial end of the project. An independent procurement review noted generalized poor procurement filling and contract management in both PIUs. Asset control was almost non-existent. In the AFMIS PIU there was no procurement staff appointed for a long time. Financial management was an issue particularly at the beginning of the project and suffered also from staff turnover, but improved gradually and was considered satisfactory at project end. The monitoring and evaluation system foreseen in the project document was not fully implemented, as evidenced by the absence of an automated M&E system and independent performance audits. The absence of a genuine M&E system at project level, combined with an incomplete results framework and no detailed reform implementation plan, affected negatively project supervision and the possibility to take remedial actions in due course. (c) Justification of Rating for Overall Borrower Performance rating: Moderately Unsatisfactory Despite a strong initial commitment from the Government to the comprehensive and participatory reform agenda, its implementation was uneven and adversely affected by strong capacity constraints and coordination issues. PIU's performance was uneven across components and suffered from the lack of an operational monitoring and evaluation system, thus influencing the overall rating. 6. Lessons Learned The following lessons can be learned from the ICR's findings: The project reflects the numerous and often painful lessons learnt from now 30 years of Bank support to civil service reforms: projects of this kind need to be much more focused and in line with the country's context and administrative capacity. The phasing of reform needs to ensure that the basic human and financial resources management functions and processes are in place or addressed before envisaging complex IT solutions. The approach needs to be more incremental and opportunistic, thus requiring more flexibility. Given the highly political nature and sensitivity of government reengineering and civil service reforms, their feasibility needs to be based on a thorough analysis of the country's political economy, governance, and corruption situation. Whenever, public employment plays an essential welfare function, like in Yemen and many other countries, credible alternatives, such as social transfers or safety nets need to be envisaged to alleviate the social impact of streamlining and thus opposition to the reform. As evidenced by this project, appraisal needs to focus on the effective demand as much as on the need for reforms and avoid the supply of conventional technical solutions. For instance AFMIS 28 suffered from the lack of demand for public finance management reforms. Technical solutions alone cannot appropriately address public sector management deficiencies as they fail to capture the human dimension and the related bounded rationality affecting the decision making process (ref. D. North). The set up of an HR information system was a necessary but insufficient condition to improve HR management, which is strongly influenced by socio-political considerations and informal rules and procedures. Therefore, much more attention must be paid to the "soft" nature of public administration reforms as well as to the informal institutional rules and organizational arrangements. At operational level, this project highlights the importance of a specific and realistic results framework, based on a reform strategy with a detailed and time bound implementation plan and informed by an appropriate monitoring and evaluation system at project level. The M&E arrangement need to be clear from the start. It further, shows the importance of a detailed and candid risk analysis and a realistic assessment of the mitigating measures. The project's ambitions and scope need to reflect this risk assessment. It also shows the need for continuity in project management and an ongoing dialogue with the project counterparts and main stakeholders. This requires a strong and clear involvement of the field office. Finally, two lessons learnt that could be useful for the upcoming public finance modernization project are: (i) the importance to address the payroll fragmentation and lack of integration with AFMIS and the HR systems and databases, and (ii) the need to systematically include HR and Payroll controls in internal control and external audit of spending units (not limited to the financial flow as it is currently the case). 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies The Borrower's comments on the Bank's performance have been taken into account and warrant an unsatisfactory rating. The data on wage bill evolution has been modified and replaced by data from the approved IMF art. IV reports. The impact of the Wage strategy on the wage bill increase has been qualified. The role of increased allowances has been added. The baseline has been changed from 2006 to 2005 as requested. The new personnel referred to the Civil Service Fund in 2010 and expected to be processed in 2011, beyond the project's time frame, are mentioned but cannot be taken into account in the assessment. The origin of the referred workers has been qualified. Regarding the reengineering studies, the report mentions that the project activities focused on the first phase, i.e. conducting the studies, and that implementation was foreseen in the second phase, after 2006. It also mentioned that these studies were delayed and that their implementation has not been supported, despite the project's two extensions. The statement about the declining number of IT personnel at MOCSAI has been corrected. The 17% spent on project preparation and management includes the project preparation costs. 29 The ICR considered the statistics on retirees, double dippers and ghost workers removed from the payroll mentioned in the comments and decided to keep the official data produced by the Government in the 2009 joint report from the Ministers of Civil Service and Insurance and Finance, confirmed by various Bank assessments and external reviews. All supporting documents are listed in annex 7. Most comments provide useful additional information on the country context and reform implementation. They are summarized in annex taking into account the information already provided in the self-assessment. (b) Cofinanciers None (c) Other partners and stakeholders None 30 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD (USD millions) Appraisal millions) CORE SYSTEMS 14.1 23.4 IMPROVEMENT CIVIL SERVICE FUND 4.9 0.28 STREAMLINE AND 1.5 4.95 RATIONALIZE BUILDING CAPACITY 3.9 0.62 PROGRAM MANAGEMENT 2.5 2 PPF REFINANCING 2 0.92 Total Baseline Cost 28.9 32.17 Price and cost Contingencies 1.1 0.00 Total Project Costs 30 32.17 Total Financing Required 0.00 0.00 (b) Financing Appraisal Actual/Latest Type of Percentage of Source of Funds Estimate Estimate Cofinancing Appraisal (USD millions) (USD millions) Borrower 3 0.00 .00 International Development Association 30.00 0.00 .00 (IDA) 31 Annex 2. Outputs by Component After more than 10 years of implementation and two restructuring, the project achieved the following modest outputs: 1. Component ­ Improve Core Systems 1.1. Human resources management framework Advice and technical assistance for the National Wage and Salaries Strategy. The project financed the National Salary Strategy Report prepared by ARADO and Ernst&Young. The Strategy was adopted in 2005, providing for (i) a substantial wage increase across the board (between 20-40%), (ii) a revised employment and wage scale, only partially addressing the compression problem, and (iii) a revision of the numerous allowances. However, it fell short of taking into account the report's major recommendations on the necessary wage decompression, the consolidation of the wage scale, the integration of basic allowances into the wage, and job description based pay. As mentioned under component 2, the wage and salary increase was general and not linked to agencies reengineering and streamlining, thus resulting in a strong wage bill increase. Advice and technical assistance for the revision of the corresponding legal and regulatory framework for civil service. It includes notably the adoption of a new law No 43/2005 concerning employment, wages and salaries system, its' executive regulation issued by Republican resolution No 99/2007, and the following bylaws: Resolutions of the Council of ministers Nos. (165-237-238) in 2005 specifying the rules of transfer to the general scale of employments, wages and salaries for the higher authority, civil servants the military; Resolutions of the Council of ministers Nos. (239, 273,287) between 2005 and 2007, providing for the executive procedures for the first phase (increase) of the wages and salaries national strategy; Resolutions of the Council of ministers Nos. (136,137, 144) in 2006, specifying the bases and rules for awarding the different allowances ; Cabinet Decree No 397/2007 providing for the executive procedures for the second phase (increase) of the wages and salaries national strategy Input for the revision of civil service recruitment and employment rules and procedures: Cabinet decree No. 278/2004, No 143/2006 and No 302/2007, as well as Prime Minister Circular No 1/2004 pertaining to the procedures governing new recruitments and the adherence to the new automated recruitment system. Resolution of the Council of Ministers No. 413/2006 concerning the appointment system in public employment; Resolution of the Council of Ministers No. 161/2007 related to the approval of the Civil Service Fund Program for the period 2008-2010. Assistance for the reorganization of the Ministry of Civil Service and Insurances (MOCSAI), through the Republican resolution No 235/2007, based on the reengineering study. 32 Technical assistance on the civil services directory, resulting in the adoption of initial procedures for the establishment, adoption and dissemination of government services directories, with the adoption of Cabinet Decrees No 51,184, 256/2006; 123/2007 and 126/2008. 1.2. The Employee database (EDB) A civil service census was completed and an initial Employee Database (EDB) was established, based on the existing personnel data in 1998 with an IDF grant, during project preparation. This Database was developed by MOCSAI, based on Oracle and initially excluded the defense and internal security personnel (around 50% of employees). EDB is not fully integrated. Despite the law 43/2005, providing for the establishment of a central EDB within MOCSAI and regularly updated by all Ministries/ Spending units, the Database is still not fully connected / integrated with the 3 Databases for the Military, the security sector and the insurance sector (retirement). EDB is incomplete and not up to date. The data populating the original database is a snapshot of the civil service in 1998, with the addition of new employees. Partial information was provided for employees' age, length of service, unit of employment, pension contributions, as well as on retirements, head count/ grade, .... The data accuracy is not guaranteed given the missing data (cf retirement dates), unrecorded personnel changes (departure, transfer, retirement) and changes to personnel details (life events/ dependants,...). The database also suffers from insufficient verified documentation substantiating benefit claims and the absence of systematic data correlation. EDB data presents some major consistency issues. For instance, it records only 693 pensioners, 21 transferred and 16 dismissed employees, for the whole civil service, but more than 11 647 on overseas study, representing 2% of the total employees. HR data ownership issue and reporting deficiencies from spending units. The HR data is largely created and "owned" by the 1400 spending units, managing their respective payroll. Many of these units report primarily to the 333 Districts, organized in 23 Governorates, with only a dotted reporting line to the ministries. Consistent, timely and accurate data collection and verification is thus a particular challenge. In an attempt to improve this, 22 Regional Information Centers (RICs), reporting directly to MOCSAI were established in order to update directly the central EDB, based on the data they collect from the spending units. To date, equipment, training and capacity building of these centers is still ongoing. 18 are in principle "connected" to the central EDB, but IT infrastructure and connectivity remains an issue. 3 RICS still have to be established at the Ministries of Education, Defense and Interior, representing ¾ of civil servants. The acceptance and cooperation of service units with the centers vary, so does the consistency, format and accuracy of data provided and flowing into the central EDB. Thus, there was no evidence at this stage of a significant improvement of EDB. The EDB is not connected to the fragmented Payroll system, managed on an aggregate level by the Ministry of Fnance, based on line item inputs from the 1400 spending units. MOCSAI has developed a MS Access based Payroll database, which is presently being tested in 10 sites. However, this work has not been completed by project end nor have the main constraints affecting the EDB been resolved. The following issues regarding the HR MIS and EDB seem outstanding: (i) MOCSAI's mandate and its acceptance, (ii) a clear, standardized and documented process for primary HR and payroll data creation, collection, reporting, updating and control, (iii) clear, timely and effective functional reporting and control lines between the spending units and line Ministries, (iv) the full 33 connectivity/integration between EDB, Payroll, HR databases from the Ministries of Education, Defense and Interior, as well as with Pension and CSF databases, and (v) IT infrastructure and connectivity constraints both at the central and decentralized levels. There is no evidence that ministries and spending units use the EDB based data and reports for HR management nor HR policies or decision making, probably due to the above-mentioned limitations. 1.3. Biometric information System (BIS) The project has supported the establishment of a biometric information system, capturing employee number (ID), signature, fingerprints and pictures for each individual employee within the civil service, along with the Military, Judiciary and Police (added by presidential decree in 2004). As off project closure, work on the BIS is well advanced but the system is not yet operational. Following activities remain outstanding for the completion of BIS: Biometrics data remains to be captured for more than 173 950 civil servants; Half of the data captured is still on CDs or temporary storage solutions, and needs to be uploaded in the BIS; Data in BIS has to be verified and reconciled with the data hold in the EDB, and correlated with payroll data, before the ID cards can be printed. Currently, there is an important discrepancy between different HR datasets. The EDB has 540 000 entries, the BIS has projected 489 000 entries, MOCSAI's 2010 budget is based on 424 000 employees, MOF's 2010 budget 421 000 employees and the payroll includes 400 000 employees. These datasets need thus to be verified and reconciled. As mentioned above, the EDB data is partly obsolete (1998 snapshot) and of limited accuracy. It needs a complete verification and update, before BIS data can be validated. An institutionalized process and adequate capacity needs to be maintained for the sustainability of data collection, storage, and management, as well as for future cards printing, given the project closure and MOCSAI takes over these functions from the contractor. An appropriate maintenance contract needs to be concluded for the BIS and funded from the GoY budget. In conclusion, while valuable work has been done, it did not result in an operational system at this date and sustainability is a matter of concern. 1.4. Transparent hiring and job descriptions The project provided advice and technical assistance for the modernization of civil service recruitment and employment process, which resulted in the following output: Cabinet decree No. 278/2004, No 143/2006 and No 302/2007, as well as Prime Minister Circular No 1/2004 pertaining to the procedures governing new recruitments and the adherence to the new automated recruitment system. Resolution of the Council of Ministers No. 413/2006 concerning the appointment system in public employment; Resolution of the Council of Ministers No. 161/2007 related to the approval of the Civil Service Fund Program for the period 2008-2010. 34 A pilot project for merit-based recruitment and remuneration was launched by MOCSAI in October 2008. By March 2009, 3 heads of sector (deputy minister) and 16 general managers were recruited through a competitive procedure. However, the two-tear remuneration system envisaged has not yet been put in place. A study on standard job classification was prepared by Team International. However, the new job classification and description have not yet been implemented. Republican Decree of 2007 (in the context of the CSF action plan) regulates new hiring and links it to identified needs and job descriptions. The latter still need to be developed and implemented. 1.5. Payroll modernization The project put little emphasis on the payroll system consolidation and modernization. At project end, the payroll system is still fragmented and heterogeneous: MOCSAI has an Oracle database and has just developed a MS Access based Payroll database, which is presently being tested in 10 sites. The Ministries of Education and Oil run a more sophisticated and up to date payroll system, based on an Oracle database. Fragmented data production and ownership by the 1400 different spending units. Spending units use a variety of different payroll applications, developed or adapted internally and report just line item carry forward data. Some more remote spending units still use manual payroll ledgers. No standardized system is in place for payroll data input, validation, reporting, modification or control. Payroll controls seems limited to the verification by MOF that the monthly drawing of each spending unit does not exceed 1/12th of annual allocation and that total annual ceilings are respected. Payroll database is not linked/ integrated with other HR and financial databases/ information systems, such as EDB, BIS, AFMIS. 1.6. Accounting and financial management information system (AFMIS) Despite a late and difficult start, the AFMIS sub-component has made significant progress since 2004. By project closure, a computerized system with the core modules has been developed, tested, and rolled-out to the four envisioned pilot ministries (Finance, Education, Health Care and Population, and General Operations and Works). Its roll-out to five additional institutions started: Customs Authority, Tax Authority, Ministry of Water and Environment, Ministry of Oil and Minerals and the Ministry of Electricity. The following outputs were achieved: Development of the following modules: (i) budget preparation, (ii) budget execution, (iii) account payable, (iv) fixed assets, (v) revenue, (vii) final accounts. The core modules (budget preparation, execution (safe commitment and procurement controls, and final accounts) have been finalized and are currently pilot tested in the Ministry of Finance and the above- mentioned three line ministries. 35 The budget preparation module is used by the Ministry of Finance since 2007, for the 2008 budget, although based on the uploading of budget proposals from spending units in the form excel spreadsheets, as the module has not yet been rolled out. The budget execution and the final accounts modules are pilot tested since 2008 in all four pilot ministries, representing a substantial part of the budget (60%). The other modules (accounts payable, fixed assets and revenue) are currently being finalized, based on the external verification done by KPMG and will then be tested in the four pilot ministries before being rolled out. The data exchange module with other systems, such as HR, Customs, Taxes, Payroll, Debt, LGMIS for externally funded projects, needs to be developed. The cash management module has been postponed, in the absence of a treasury function at the Ministry of Finance, despite recommendations since the beginning of the project that this function, presently carried out by the Central Bank be transferred to the Ministry of Finance, in accordance with international practices. The commitment control module has been postponed to the second phase and will be covered by the follow up project. AFMIS has been tested four times (December 2005, April and December 2007 and December 2009) by KPMG and most findings/ issues have been acknowledged and addressed. Work is notably ongoing to develop functionalities for the expenditure module, the revenue module, the accounts payable module, the fixed assets module and the budget closing/ opening function. Bank statement reconciliation and reporting is also under development. IT infrastructure capacity constraints remain, as the current hardware/ servers do not support the number of envisaged users. It does not guarantee business continuity or integrity in case of server failure (ref. KPMG tests). In sum, the three core modules have been successfully developed and are currently being implemented in pilot ministries. Three out of the nine modules are been further finalized before being pilot tested. Three modules have not been developed yet. The data exchange interfaces with other financial management systems (still needs to be developed. 2. Component ­ Civil Service Fund 2.1. Enforcement of retirement rules and removal of ghost workers and double dippers from payroll, The CSMP's reengineering studies and HR databases (EDB/BIS) did not result in the identification of many redundant employees nor the ghost workers and double dippers on payroll. Most of the redundancies, referred to the CSF, came from defunct state enterprises and only 25 double dippers and no ghost worker were identified through the biometric database. 3767 employees voluntary departures were also recorded. These figures came from the conclusions of a joint working committee between the MOF and MOCSAI on 03/09/2009. A Cabinet decree (211/2007) and a circular from the prime minister were adopted to strengthen the existing legislation and specify actions to be taken to enforce retirement in the public service. 36 A MOCSAI circular was passed authorizing individual agencies to hire new staff conditional upon their phasing out redundant employees. Agencies are only allowed replacing 40-60 percent (depending on qualifications) of employees that are declared redundant. However, there is no evidence of redundant employees reported by agencies, other than the ones referred to the Civil Service Fund. From 2006 to 2009, 28 097 employees, who had served past retirement age, were retired. However, this number is a small percentage of the total civil service. Furthermore, since 2000 a total of 44 703 civil servants have retired, on average 4500/ year, representing 0.5% of the total civil service. This is well below the normal retirement ratio of 3% noted in other civil services with similar population age. This confirms, as mentioned above, the serious shortcomings in HR data (7% of EDB population has no birth/starting date) and suggest that retirement rules are not consistently enforced. 2.2. Civil Service Fund (CSF) The legal framework to set up the CSF has been developed during project preparation. The Fund was established with the enactment of Law 1/1999 and Law 314/2000. However, legal intricacies with the existing Legal and regulatory framework governing the civil service and budget constraints prevented the Fund from being operational before end of 2004. The CSF set up and equipment of branches in four Governorates was supported by the project. Its staff was trained under the capacity building component ­ 4 and the Dutch Trust Fund. CSF policies, rules and procedures were prepared under the project. The CSF processed 14 127 redundant staff, mainly from defunct or privatized state enterprises, during Phase I from April 2006 to March 2009. 2 250 employees received their statutory pension, 1 100 were redeployed to other public services and 10 777 taken care of by the CSF. Among these, 93.7% opted for early retirement and were transferred to the pension authority, and 6.3% opted for severance packages (10 300 USD av.). The total cost to the Fund was 17.8 m USD. A complaints system was put in place and reviewed 800 complaints of which 300 were approved. A second phase was launched end of 2009 to process 3626 employees, identified mainly from State enterprises (Yemenia airlines) and from political parties. However, it stalled in the absence of financial compensation by the MoF of the extra cost incurred by early retirement, the most common option chosen by employees referred to the CSF. The public service units prepared preliminary lists of the redundant employees for 12,000 employees to be reviewed by the CSF in 2011 3. Re-engineer Government agencies 3.1. Government's mandate study A study on the Government's mandate was completed in 2006 by Yeoman Ward (US$268,000). It compiled the current roles and responsibilities of ministries and main agencies, identified overlaps and ambiguities in roles and responsibilities, develop recommendations to rationalize and streamline ministries and main agencies, identified areas for Government disengagement and enhanced private sector role and outlined new roles and responsibilities of ministries and main agencies. 37 The set up of a Presidential Commission on the Government's mandate was foreseen to review and approve the study's recommendations with the support from the project's technical secretariat. The presidential decree setting up this commission was not by the end of the project. 3.2. Introducing job classification in the Civil service A study was conducted on standard job classification by Team International (US$296,000). A Cabinet order n°146/2009 regarding the new job classification has just been adopted and needs to be implemented. This activity aiming at introducing the building blocks for a merit based civil service and for which substantial training was foreseen has thus to achieve the expected results. 3.3. Performance indicators study The study, realized by Scientific Council for Applied Sciences (US$200,000) was completed in May 2007 for selected ministries and agencies by engaging focal teams in the process of defining the objectives of monitoring and evaluating performance, using performance measurement criteria within and across institutions. Performance indicators were developed, trying to link directly to the mission, vision and strategic goals of the following nine ministries and agencies: (1) General Secretariat for the Cabinet; (2) Ministry of Finance; (3) Ministry of Higher Education and Scientific Research; (4) Ministry of Local Administration; (5) Ministry of Oil and Minerals; (6) Ministry of Public Works and Highways; (7) Ministry of Telecommunications and Information Technology; (8) Public Authority for Electricity; and (9) Yemen Standardization, Metrology & Quality Control Organization. However, the final report raised important quality issues, as the performance indicators were not objective, focused, measurable nor practical. They were mainly internally oriented and not focused on the performance of the ministry or agency in terms of service delivery, quality or value added for users. Furthermore, the study was lacking specific recommendations on how to develop these performance indicators. 3.4. Re-engineering studies Five reengineering studies have been completed between 2006 and 2007 for the following agencies: (i) COCA (US$1,261,250 by PWC), (ii) Taxation Authority (US$758,000 by PWC), (iii) Internal Audit Agency (US$555,000 by PWC), (iv) Customs Authority (US$1,098,173 by WSP), (v) MOCSAI (US$493,900 by Yeoman Ward). The study on COCA was concluded only in September 2009, after contractual and financial issues were resolved. The studies followed a participatory SWAT methodology involving key internal and external stakeholders of each ministry and agency through workshops and surveys and made recommendations on: (i) human resources management policies, (ii) organizational charts and job classifications leading to right-sizing, (iii) streamlining of selected back-office and front-office processes, (iv) service quality improvements through simplified processes and increased transparency, (v) required legal and regulatory changes, (vi) implementation strategies, output and impact indicators, (vii) the adoption of international auditing standards for COCA and Internal Audit. Based on these studies the following actions were taken : Service guides were developed, including description of the procedures required under applicable laws and regulations, and the fees payable for individual procedures. Guides are posted in visible places of central and local offices, and published on the agency website. Cabinet Decree No. 5 of 2009 approved the guiding model and terms of reference for the reengineering and restructuring process in Government agencies. 38 Cabinet Decree No. 125 of 2008 provided the widespread utilization of the outputs of the CSMP reengineering activities for streamlining administrative reforms in ministries and other agencies. Cabinet Decree No. 127 of 2008 approved the Draft Republican Decree on the rearrangement of the roles and responsibilities of Internal Auditing in Government agencies. Republican resolution No 235/2007 on the reorganization of MOCSAI. President resolution n°5/2010 on internal audit. However, the restructuring process of these agencies has remained limited to date, notably due to internal resistance, change in leadership (Customs and Tax Authorities), disagreements with the consulting companies (COCA, TAX authority studies), and pending changes to the legal (MOCSAI, COCA) and regulatory framework to implement key recommendations of the reengineering studies. 3.5. Innovation fund The Fund, with an initial budget of 2.34 m USD was managed by the Government, who contracted the Arab Administrative Development Organisation (ARADO) to review the proposals submitted to the Fund. It experienced important implementation delays, as it started only in June 2004 and signed the contracts for the 6 selected projects by end of 2007. The Fund aimed at providing (i) short term, innovative and visible benefits to the citizens to sustain the reform momentum, (ii) to test the idea of creating a common fund to channel donor financing for agency restructuring, and (iii) to leave room for changing circumstances and opportunistically capture reform initiatives by other agencies. Based on the limited information available these objectives do not seem to have been achieved for the following reasons: It took seven years until six projects were selected and approved and between one to two years to implement them, which does not qualify as "quick fixes". While most selected projects were geared towards the citizens, their degree of innovation varied. For instance, the set up of a help desk at MOCSAI or an information center at the Airport, can be considered innovative compared to existing administrative practices, the installation of three ATMs for the postal office or medical equipment for the Republican Hospital's emergency services might be less innovative In the absence of any information on these projects' outcomes, it is not possible to assess the effectiveness of the intended benefits to citizens. As the innovation fund has been entirely managed by an external consulting company with very limited interaction with the reform teams, it does not seem to have informed the reform process or support its acceptance/ momentum, as intended. For instance, it has not contributed to the set up of a common fund to channel donor financing for agency restructuring, which does not exist. 4. Capacity building: Activities under this component have been transferred to the Dutch Trust Fund (TF-24755) after the project's mid-term review in October 2003. The Trust Fund was closed on December 2007. 27 training programs have been delivered to approximately 1,500 civil servants, including 5 management training programs for approximately 600 senior Government managers, of which the following: Activity Cost Target/ output Objective/ Content 39 Restructuring USD Training 29 heads and Build capacities of focal teams in administrative strategy and 49,000 members of focal teams reform supervision on Principles of supervision on consultants who will consultants be conducting restructuring in their agencies Review the Lebanese experience in this field Information USD Training 108 IT and RIC Build the capacities of computer operators in IT systems and 205,912 staff on the following: center and RICs in governorates computer Computer operation Improve the skills of engineering working in IT Advance computer center and RICs programs Senior and USD Implemented workshops for Introduce participants to modern theories and middle 262,530 154 ministers, deputies and methodologies of administrative reform managers in chairmen of agencies state agencies Training of USD Training 30 trainers from Improve skills of trainers in government training trainers 140,207 government training centers and institutes institutes and centers Provide them with necessary skills to use modern methodologies in training Transferrable USD Training of 570 employees Provide participants with the following skills and skills 270,210 working in offices of senior capacities: managers in targeted Report writing agencies by restructuring Organization and documentation of meetings process and other agencies Principles of computers Making presentations using modern technologies Budget preparation using excel sheets English language Managers of USD Training of 90 managers of Provide participants with modern concepts in personnel 372,670 personnel affairs in central administrative and organizational reforms and affairs in civil and local authorities their relation to HR system development service (1) Gain skills to apply HR systems Focal teams USD Training 30 members of focal Introduce participants to the modernization and 120,340 teams in 9 targeted agencies development of administrative systems targeted by by restructuring restructuring activities in their agencies Change management and resistance Project USD Training 20 people of those Provide participants with necessary skills to help planning using 4,000 nominated to implement them plan and implement innovation programs computers innovation programs from 6 they supervise agencies Computer USD Training 48 engineers and Build capacities of managers and developers at IT systems 451,000 managers working in IT center and RICS on Oracle 9, Windows OS, Unix, managers and centers and RICs in Developer 2000 developers governorate Computer skills USD Training 150 staff from Computer proficiency and project 149,640 targeted agencies by Providing those working in planning departments planning restructuring and other in restructuring agencies by necessary skills to use 40 agencies MS project (50 employees) Employees USD Train 50 staff responsible for Provide trainees with new concepts on government responsible for 193,050 public help desks in services implementation government agencies that Reviewing the Jordanian experience in this field of public completed government service guides service guides General USD Training 50 Personnel Inform managers on the accomplished HR systems managers of 199,090 Manager in government and how to deal with them personnel agencies affairs in civil service (2) Specialized USD Qualify 75 government Provide government agencies with leaders certificate in 78,000 employees in agencies understanding modern concepts of administrative public targeted by restructuring and development to help the CSMP upon initiating administration other agencies in the restructuring activities concepts of public administration and administrative reforms Executive MPA USD Qualify 37 people from Provide government agencies with qualified leaders to 162,000 government senior managers create necessary momentum for restructuring activities to obtain MPA Executive MBA USD Qualify 15 leaders from Provide economic government units with highly 97,000 government agencies of qualified leaders to improve their performance economic nature Source: CSMP Project Implementation Unit Annex 3. Economic and Financial Analysis N/A ­ Governance institutional support project 41 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Supervision/ICR Fabian Seiderer Sr Public Sector Mgmt. Spec. MNSPS ICR Robert Beschel Lead Public Sector Mgmt. Spec. MNSPS TTL Olaf Smulders E T Consultant MNSPS Giulio De Tommaso Sr Public Sector Mgmt. Spec. AFTPR TTL Catherine M. Laurent Sr Public Sector Mgmt. Spec. MNSPS TTL David F. Varela Sr Public Sector Spec. LCSPS Ardita Abdiu Sr Public Sector Spec TTL Moad M. Alrubaidi Financial Management Specialis MNAFM Arun Arya Sr Public Sector Mgmt. Spec. MNSPS Linda Van Gelder Sector manager TTL Robert E. Hindle Country manager MNA TTL Akram Abd El-Aziz Hussein El- Sr Financial Management Specia MNAFM Shorbagi Lydia Habhab E T Consultant MNSPS Josephine Masanque Sr Financial Management Specia OPCFM Abduljabbar Hasan Al Qathab Senior Procurement Specialist MNAPR Mikael Sehul Mengesha Senior Procurement Specialist MNAPR Reginald W. Miller Consultant QLP Mikhail V. Pryadilnikov Consultant ECSPE Harjit Singh Consultant AFTPR Hisham Ahmed Waly Sr Financial Management Specia MNAFM Alexandra C. Sperling Senior Program Assistant MNSED David Biggs Consultant Johanne Angers Senior Operations Officer Jamal Abdulla Abdulaziz Senior Procurement Specialist Martin G. Rama Lead Economist Afif Al Saqqaf Consultant 42 (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY98 226.93 FY99 201.73 FY00 26 144.64 FY01 2 5.82 FY02 -0.09 FY03 1.62 FY04 0.00 FY05 7.85 FY06 0.00 FY07 0.00 FY08 0.61 Total: 28 589.11 Supervision/ICR FY98 0.00 FY99 0.00 FY00 1 2.51 FY01 23 77.72 FY02 34 186.63 FY03 23 164.64 FY04 47 198.87 FY05 38 264.51 FY06 32 146.24 FY07 25 190.95 FY08 4 93.89 Total: 227 1325.96 43 Annex 5. Summary of Borrower's ICR and/or Comments on Draft ICR Republic of Yemen Ministry of Civil Service and Insurances Civil Service Modernization Project Report on The Outputs, Outcomes and Effects of the Civil Service Modernization Project Submitted to the Higher Council of Civil Service The level of implementation of the project activities per components: Component 1's level of implementation: Apply systems and policies to strengthen the basic systems: a. Building units for archive in 1154 main and subsidiary units and preparing the functional file (a file of documents) for every employee. A fixed and a mobile file have been prepared for every employee. The number of those who had files used to be 20 thousand employee only. This task has been carried out based on the unified documentation and archiving system. The percentage of employees whose functional files have been supplied with the basic legal documents is 100%. b. Implementing the first statistical survey of the administrative body units and the public and mixed sectors (the Functional Survey, Census and Public registry August 1998-January 2000). Towards the achievement of all strategic and procedural objectives of this survey, the project secured and provided all the requirements for the successful implementation (financial, human and training). For the purposes of this report, we will only refer to the key factors that enabled the project to succeed in survey implementation with high level of efficiency and effectiveness. c. Supplying the civil service officer in the Capital Secretariat and the governorates with the systems and programs through which the employees' data and information can be provided at the local decentralized level. The capacity of the local authority has been further enhanced by information centers that contain a database and sub-information related to the employees of public service units at the level of the local authority and is networked with the main information center and the central database of the Ministry's HQs. 44 d. Setting up the central database of the State employees which contains all the employees and affiliates to the civil, military, security and judiciary bodies along with the retired employees. The main database lists the employees of each sector alone with sub databases that represent the sub unit at both the central and local levels in civil service, military body, security forces, judiciary and retired employees. e. Rationalizing the employment policies and consolidating the principle of equity and equal opportunities through the following tasks; Adopt a number of policies and procedures that aim at getting rid of the reasons leading to deficiencies and corruption in public office, on top of which are; Placing strict controls over the recruitment process. Regulate the transfer movement among units. Address the conditions of employees who get paid without running a job. Opening the allowed time period for the unpaid leaves. Connecting between staying in office and performing the job. Unifying the entry point to the public post so that MOCSI becomes the sole and main gate to new recruitments. Developing a new policy and an e-recruitment system to be circulated over all the administrative units and public service units. Developing a unified scheme for salaries and wages and applying it in all units of public service. Preparing and applying the appointment system which contains the basis, criteria and specifications for occupying each public post. f. Strict implementation of the retirement law through the following: Revising the Insurances and Pensions Law No. 25 of 1991 in accordance with Law No. 1 of 2000 which lifted the minimum retirement pension to match the minimum wages. The law also stipulates that allowances shall be covered by the insurance deductions and the retirement pension shall be linked to the last salary. Simplifying the retirement procedures. Setting up a database and an e-system for retirement. g. Developing an electronic system and implementing a comprehensive plan to replace the non- Yemeni employees with Yemenis. This process has resulted in the replacement of over five thousand employees. h. Networking the sub information centers in the Capital Secretariat and governorates with the Main Information Center to constantly and timely update the functional developments in later stages. 45 i. Issuing the payrolls of local authority units through the information centers. j. Implementing the fingerprint and photo system: The biological fingerprint and photo system constitutes an entry point towards building a valid and updated database as well as to plan and manage the human resources. Such system will revoke all cases of repetition or shadow employees. The progress made until mid February has been as follows: 97% of the Civil Sector; 98% of the Judicial Authority; 50% of the Military Sector and 65% of the Security Sector. Progress made in the second component- rationalizing the number of employees- Civil Service Fund: Within the CSMP, the Civil Service Fund proposes a number of solutions to mitigate the economic and social impact of rationalizing the number of public employees who will be leaving the public office due to the restructuring processes and due to the reduction in the number of employees to fix the deficiencies in the payrolls under the administrative reforms. The following outcomes have been realized as a result of the Fund's work: The redundancy made under the first chapter (employees' wages and compensations) of 4,684 employee in a total annual cost of 1,848,915 YR (only one billion eight hundred forty eight million nine hundred and fifteen thousand YR) including 978 employees with an amount of 294,763 (two hundred ninety four million seven hundred and sixty three thousand YR). Redundancy made under the third chapter (subsidies, grants and social benefits) of 9,101 employees with an annual cost of 3,542,429 (three billion five hundred and forty two million and four hundred twenty nine thousand YR) including the autonomous units of 291 employees with an annual cost of 88,672 (eighty eight million six hundred seventy two thousand YR). These units used to pay the wages on their own. The progress made in the third component: rationalizing and restructuring the governmental apparatus: The strategic objective of restructuring is all about regulating and rationalizing the State apparatuses through achieving ancillary objectives of; reforming and developing the civil service setup, reorganizing and structuring the administrative processes and service provision through a high-scale restructuring of guiding governmental bodies as well as a detailed organizing process of the various civil service sectors. The restructuring process goes through two phases; the first is when the project carries out a mission of reorganizing and structuring in order to minimize the aspects of deficiency and organizational flaws. Upon finalizing the restructuring of the guiding governmental bodies, the 46 second phase starts with enough expertise and knowledge to carry out a massive restructuring process of all governmental bodies and institutions. Through the project's activities, 11 studies have been conducted to form a benchmark for any change to take place at the legislative and organizational levels and shift in work approaches at the state's level. The progress made in the fourth component: building the institutional capacities and training: Within this component, a number of training events have been held to achieve the specific goals defined in the project document particularly provision of the training needs to keep up with the project's other components in restructuring, basic systems, project management and preparation adaptation of the targeted bodies and institutions for the second phase. During the project, 29 training programs have been implemented in various relevant fields and specializations such as strategic planning, change management, human resource management, computers...etc. with the participation of 1,685 administrative manager and public servants. The National Strategy of Wages and Salaries: GOY adopts a comprehensive program for economic, financial and administrative reform in conditional partnership with donors to meet the obligations of the Yemeni government to rectify deficiencies, dry up the sources of corruption and conduct a radical review of the laws, policies and regulations in connection with the good governance principles. Among the components of the Government's general program are reforming the jobs, salaries and wages system and improving the livelihoods of employees and retired people by drafting a national Strategy for wages and Salaries. The financial effects of applying the first and second phases of the National Strategy for Wages and salaries on the State's general Budget: The National Strategy for Wages and Salaries was aimed at improving the conditions of public servants and granting them a decent life through improving the incomes gained through public office .And to a large extent, protecting the civil servants from the negative influence. However, the high inflation rate that has accompanied the implementation of the first and second phases of the Strategy has swallowed the increases obtained in the public employee's salary. Moreover, the National Strategy for Wages and Salaries has addressed a number of deficiencies and made some reform to the payrolls at the level of public service unites including the following; 1. Correct and clean the payrolls from any illegal positions revealed such as; Downgrading all the employees who had reached one of two terms (retirement age or end of service) dead or permanently disabled after transferring them to the public setup and retiring them. Inventorying the entire abundant workforce and transferring them to the Civil Service Fund. Downgrading the double, discontinuous and shadow employees. 2. Replacing the non Yemeni employees with Yemenis. 47 3. Lessening the financial redundancies that had been illegally disbursed from the budgets and allocations as a result of implementing reforms or removing deficiencies that reflected on the State's general budget and preparing the budgets based on the actual implementation of the Strategy. The reforms made and related to the implementation of the law and the strategy have reflected positively on the State's general budget and the budget of the economic, autonomous and annexed units and the private funds through the following; 1. Making financial redundancies that have been cut down from the State's general budget of 55,034,373 thousand as a result of downgrading 99,894 employees as follows; 70,021 employees were retired in an annual cost of 26,788,000 YR. 14,210 employees were transferred to the Civil Service Fund which in turn has handled their conditions either by early retirement or the financial compensation in an annual cost of 5,108,175 YR including an amount of 3,453,757 YR that has been reduced from Chapter three for 8,810 employees. Downgrading 15,663 double and discontinuous employees and shadow and fake jobs in an annual cost of 19,387,000 YR. Reducing an amount of 3,751,198 YR as a result of replacing the non Yemeni employees with Yemenis and other redundancies. 2. Making financial redundancies from the unclassified units' budgets after processing the retired double and shadow employees numbered 818 employees. 3. Making financial redundancies from the budgets of the economic, autonomous and annexed units along with the private funds as a result of reducing the cases of deficiencies and illegal positions for 5298 employees as follows: 4,015 employees were retired. 291 employees were transferred to the Civil Service Fund 992 double, discontinuous and shadow employees were cut down. The key outcomes of the Project's first phase: It must be noted that the project, within the limitations of its basic budget, has implemented additional tasks that weren't mentioned in the main plan. These tasks were rather prioritized in the government program and the President's electoral program leading to the Project being entitled to fulfill these tasks without any extra costs on its basic budget. The tasks undertaken by the project in the first phase have had direct and indirect positive impacts on development of the public management system, administrative reform and combating corruption and consequently the State's general budget. We can say that the Project, thanks to the support of the President, has managed to achieve substantial results that made the government and its partners of the international organizations and donor countries proud. 48 On top of these achievements are; diagnosis of the current situation and administrative malfunctions, corruption manifestations and the wrong culture in dealing with public posts. A consensus has been reached of the importance to fix this situation by highlighting the risks of it. Today, the good governance programs and public management systems have become a priority for the government, the political parties and across the entire society. This was a necessary access point to implement the administrative reform program by changing the culture and attitudes towards public posts and getting rid of the wrong practices in order to fulfill a combination of strategic and procedural tasks. a. Results achieved by the project that contribute to the reform and development of the public management system: 1. Unifying the documentation systems to become one system under which archive units have been built in 1154 main and ancillary units along with the preparation of a fixed and mobile file for each employee after having only 20 thousand files. 2. Enabling the government to own the necessary digital tools as an objective and scientific basis to implement the reform programs. 3. Providing the employees data and information at the local and central level and strengthening the local authority's capacity through the information centers that involve an ancillary database for the administrative unit's employees that is networked with the main information center and the central database Ministry's HQ. 4. Building a central database and main databases for the judiciary, military sector and security and later on for retirement and social welfare. 5. Installing and applying the system of bio-fingerprint and photo in the State's administrative units, the public and mixed sectors, judiciary, military and security sectors. 6. Building a comprehensive database for job applicants from the graduates of secondary schools that enabled the study of supply and demand and to define the specializations that need to be reviewed in the high education institutions and those to be targeted leading to the introduction of new specializations and restricting the enrollment in others. 7. Approaching the educational and functional characteristics of employees and their geographic distribution to enable channeling employment towards the specific specializations, public service units and governorates. 8. Applying the principle of equity and equal opportunities in obtaining a public post and getting rid of all the reasons leading to malfunctions and corruption along with largely linking staying in the office to performing the job. This is a radical change as retention, 49 promotion or salary used to be taken for granted as every one used to consider the public post a right in itself and not a duty. 9. Rationalizing the number of employees by ending the rapid increase in public posts due to an average of 35 thousand new employees per year as from 1990 to 1997. Employment through substitution has also been revoked with strict application of retirement for 62,415 employees in addition to replacing the non-Yemeni employees with Yemenis for in 5,793 of cases. 27,856 double and shadow employees were downgraded with 14,501 of the redundant workforce through the Civil service Fund. Other employees have been redeployed from the central authority units to the local units. In general, the rationalization procedures have led to downgrading a total of 100,185 employees which resulted in having redundancies that reduced the general budget by 55,736,455,000 YR and covering about 40% of the increases that have been paid in the first and second phases of the National Strategy for Salaries and Wages. 10. Tightening the monitoring over the inputs and outputs of public posts and the changes at the level of the administrative unit and the employee while rationalizing the expenditure over wages and salaries which has led to cleaning the payrolls from the double and shadow employees. And this will help in preventing this from happening in the future and control the additions and omissions to the payrolls as well as the premiums and deductions from the employees' dues. 11. Improving the relationship between the job and the labor market while creating an objective relationship between the salary and the position along with building a unified setup for wages and salaries based on lifting the minimum wages to match the poverty line and mitigating pressure between the minimum and maximum limits. The salary is also linked to the actual exercise of the job and the type of it with all the cadres and wage systems are enrolled in the same system within unified groups, categories, degrees and job titles after revoking anything else. Any new cadres or allocation changes have been suspended by integrating them within the main salary upon transfer to the general setup while keeping the accommodation, appearance and rural areas allocations. 12. Ending the application of the functional degrees system and shifting to the position system. 13. Wiring the employees' salaries to their workplace and ending all the illegal deputation, seconding, part-time and unpaid leaves and the educationally stumbled people in and outside the Republic. 14. Studying the State's organizational and functional setup and defining the aspects of repetition and overlapping in competencies among the public service units. This is 50 accompanied with identifying the structural and organizational deficiencies and proposing solutions by specifying the roles and functions the government is going to keep (allocating competencies between the central and local authorities and proposing the number of ministries and institutions and the roles and functions for each public service unit) and identifying the roles to be given up by the state for the private sector and CSOs. A higher committee is also formed to study the results thereof. 15. Finding public service unit that can be treated as a guiding model for restructuring the rest of public service units. 16. Setting a system and a complete manual for internal audits based on the latest criteria and the best practices. 17. Setting a system for the institutional performance evaluation and preparing the performance indicators for 9 guiding units as an introduction to doing the same in all the State's institutions. 18. Simplifying the service provision procedures to citizens and investors and easing information acquisition in timely manner through a process of inventory and documentation of the services and then including them in published governmental manuals containing all the data and conditions for service acquisition in 63 governmental bodies and 1800 service. These will be issued and published on the internet, through the media outlets and in printed manuals and guiding boards to be installed in the entrances of these units. Moreover, the organizational frameworks in the governmental bodies are set with bylaws and mechanisms along with an ongoing process of revising and developing the service provision procedures for those that are described as complicated and overlapping. 19. Setting the foundation for an ideal experience in administrative excellence, quality and creativity awards in development of government services and urging the spirit of competitiveness. The experience was about preparing and implementing six creative projects to improve public services. Such will provide a model in the future for other units to follow and compete. 20. Building the institutional capacities and equipping employees with modern sciences while introducing them to the successful experiences and practices along with the latest theories in administrative systems and organizational development. In addition, the visions of the senior and middle administrative leaderships are unified regarding the reform and development tracks to be taken and the requirements for such reform in their units through the workshops and training implemented by the Project. 51 21. Qualifying 197 of the members of the Parliament, deputy ministers, directors of institutions and general directors in the public service units in the Executive Masters Program and other certificates of specialty. b. Benefits of the Project that contribute to the State's General Budget: Upon planning the project, it was anticipated that the successful implementation of the Project's activities will lead to reducing some of the budget expenditures in subsequent phases. However, the success of the project in the first phase and its expansion to absorb extra tasks within its approved budget has actually diminished some repeated expenses in the general budget especially with regards to rationalization of wages and salaries as a result of controlling and reducing the employment process by over 50% in average from 1990-1998. This is in addition to activating the retirement system, downgrading the functional duplication cases and discontinuation and replacing the foreign labor force with a Yemeni one while implementing the Strategy of Wages and salaries. This has enabled the improvement of the financial situation of the State employees and security and armed forced personnel through the increases of the first and second phases of the strategy with direct and indirect redundancies made as follows; a. Direct financial redundancies: 1. Financial redundancies as a result of reforms made at the first phase of the National Wages and Salaries Strategy: in implementation of law No. 43 of 2005 regarding the jobs, wages and salaries regulation which took effect as of the second half of 2005, and in light of the Strategy's stipulations of substantial increases in salaries with these increases being linked to the implementation of a package of administrative and institutional reforms, a joint mechanism has been set between MOCSI and the Ministry of Finance that guarantees that such reforms are being reflected in a timely manner in the State's general budget and upon the development of proposals for the public service units budgets for the years 2006, 2007 and 2008. The total redundancies made and included in joint minutes of both ministries has been 55,736,453,000 YR. 2. Downgrading 27,856 double, discontinuous and shadow employees with an annual cost of 19,387,000,000 YR 3. Redundancies made as a result of replacing the non-Yemeni employees with Yemenis. 5,793 foreign teachers were replaced with an amount of 5,365,058 YR per year. 4. Financial redundancies made as a result of the reduction program: The program has made significant financial redundancies by considering the difference between the cost of the employees' retention in the public office and the remedial costs 52 (buying the service years + financial compensation) which are on the Fund. The total financial redundancies made from the reduction process are 27,073,704 YR with an average of 2,676 YR. this will be explained further later on. b. Indirect financial redundancies as a result of reforming the jobs' system: Making indirect financial redundancies of 32,5 billion YR as a result of suspending the employment by substitution and reducing the number of annual recruitment from 35 thousand to 12 thousand in addition to controlling the process of income tax collection and insurance subscription along with using modern systems in human resources. c. Increasing the salaries and wages tax revenues and alike charged on the workers in the State's administrative units by over 23,000,000,000 YR. Effects of Civil Service Modernization on the Governmental Management Issues and Administrative Development The success of the administrative reform programs is usually measured by the effects they bring about to the regulations, laws, attitudes and work methods more than the results or outputs. In terms of our situation and with taking into account the tasks and roles of the governmental bodies that are concerned with reform and administrative development, the CSMP has caused deeply rooted effects on the governmental administration unlike any other project. We can consider the laws, Republican decrees and Cabinet resolutions that have been issued in response to the outputs and requirements of the CSMP, assuming that these have been enforced through the administrative practice, the Project's effect. These are as follows; 1. The effects made through the Republican laws and decrees: Law No. 1 of 2004 regarding the Civil Service Fund. Law No 43 of 2005 regarding the Jobs, Salaries and Wages regulation. The Executive bylaw of Law No. 43 of 2005 regarding the Jobs, Salaries and Wages regulation issued by the Republican decree 99 of 2006. The organizational bylaw of MOCSI issued by the Republican decree 235 of 2007. The Presidential decree No. 1 of 2004 regarding the functional Card system. The President's resolution No. 5 of 2010 regarding the internal audit in the State's institutions. 2. The Cabinet resolutions issued in light of the outputs and requirements of CSMP as a change framework: a. In the field of implementing the National Strategy for Wages and Salaries: 53 The Cabinet resolutions No. 238, 239, 273 and 400 of 2005 regarding the executive procedures of the Law No. 43 of 2005 and the National Strategy for Wages and Salaries and the implementation of the first phase of it. The Cabinet resolution No. 238 of 2005 regarding the transfer rules to the general setup of jobs, wages and salaries annexed to Law No. 43 of 2005. The Cabinet resolution No.239 of 2005 regarding the executive measures of the National Strategy for Wages and Salaries and the Jobs, Salaries and Wages regulation. The Cabinet resolution No. 237 of 2005 regarding the rules of transferring the senior management staff to the general setup of Jobs, Salaries and Wages. The Cabinet resolution No. 237 of 2005 regarding implementation of the National Strategy for Wages and Salaries and the Jobs, Salaries and Wages regulation. The Cabinet resolution No. 165 of 2005 regarding the rules of transfer for the army, police and security forces affiliates. The Cabinet resolution No. 278 of 2007 regarding completion of the first phase tasks of the National Strategy of Wages and salaries. The Cabinet resolution No. 397 of 2007 regarding completion of the executive procedures for the second phase of the National Strategy of Wages and Salaries. b. Cabinet resolutions related to the public post reform: The Cabinet resolution No. 413 of 2006 regarding the appointment system for public office. The Cabinet resolution No. 136 of 2006 regarding the bases and rules of giving remote areas allowances. The cabinet resolution No. 137 of 2006 regarding the standardized system for "nature of the job" allowances. The Cabinet resolution No. 144 of 2006 regarding the bases and rules for eligibility of travel and transportation allowances. The Cabinet resolution No. 161 of 2007 regarding approval of the Civil Service Fund program for 2009-2010. The Cabinet resolution No. 286 of 2008 regarding the administrative and institutional reforms in the field of good governance. Law No. 31 of 2009 regarding the job rotation system. The Cabinet order No. 146 of 2009 regarding the job descriptions and classifications. The Cabinet resolution No. 18 of 2008 regarding the procedural manual to apply the functional card system. c. With regards to simplification of procedures: The Cabinet resolution No. 185 of 2006 regarding approval of the government services manual and methods of obtaining them. 54 The Cabinet resolution No. 123 of 2007 regarding implementation of the governmental services manual. The Cabinet resolution No. 126 of 2008 regarding shifting towards the implementation of the second phase of the governmental service manuals application. The Cabinet resolution No. 709 of 2009 regarding completion of the first and second phases of the governmental procedures simplification program. d. With regards to moving the burden of the administrative reform and development to the State bodies: The Cabinet resolution No. 125 of 2008 regarding making use of the outputs of the CSMP in generalizing and achieving administrative reform. The Cabinet resolution No. 5 of 2009 regarding the approval of a guiding model for TORs of restructuring the state apparatuses. The Cabinet resolution No. 127 of 2008 regarding the internal audit in the State bodies and institutions. e. Procedures related to reforming the public posts and their impacts on the State's general budget: Activating and improving the basic systems related to human resources like the admission, retirement and job description systems. Providing the elements and data related to the post and the public employee at the central and local levels on e-systems. Addressing the larger sum of deficiencies and illegal conditions of the payrolls by revoking the functional duplication. Providing considerable amounts of money to the State's general budget as a result of the major operations to develop and activate laws and regulations that govern the public positions. For example, an amount of 19,387,000,000 YR has been saved as a result of handling 23,269 cases of double and discontinuous employees. Making redundancies of 12,022,000,000 YR due to the implementation of the fingerprint and photo e-systems and the Wages and Salaries Strategy. Activating the laws upon budget preparation with the involvement of the MOCSI and MoF achieving redundancies in the State's general budget by 88,617,000,000 YR. Resistance of Change in the Project Life: Despite the large-scale preparations that preceded the Project and was demonstrated in a comprehensive national meeting over the modernization of civil service in the Economic Conference of 1998 as well as the implementation plan of the CSMS which had taken into account a number of potential risks, the Project faced several forms of resistance to change which negatively affected the early phases of the project. In order not to indulge more in the theoretical 55 issues of change resistance, we will review some forms of resistance either through some exercises of the administrative leaderships or from the dominance of the cultural inheritance over public posts or even through the applied regulations and procedures; The attempt to adapt the activities of CSMP to fit in rigid templates by focusing the project on partial issues with which the administrative leader feels comfortable. Maintaining the inherited and current practices in the public post system and fearing change even by the potential beneficiaries of it. Short integration between the State's apparatuses which leads to repeating the task and sometimes refusing the idea by one party and accepting it by another which hampers the implementation of certain civil service schemes. Unlike many international experiences, the task of following up, affecting and directing reform is handled by a high level governmental party that follows the Prime Minister of the President of the Republic which gives it the necessary power and influence to affect administrative reform. However, change in this project has been run at the level of ministries. For immanent and objective reasons, responsiveness with the comprehensive change has been reluctant sometimes and even repellent as long as change has become an urgent requirement from the government. The financial conditions and policies were not so helpful to make change in spite of the obvious support offered by the political leadership which has led to freezing some reform procedures or taking decisions that negatively affected the reform process. The unavailability of a comprehensive set to manage and coordinate the tendencies of the various state bodies in the fields of reform, institutional, organizational, administrative and financial development as there are sporadic attempts that don't make use of the local accumulations of successful reform and change experiences. Therefore, loans and support has been directed to State parties that work in isolation from other bodies despite the central and focal role of MOCSI and CSMP as an umbrella of the development and modernization processes. The multiplicity and repetition of the monitoring and evaluation process and sometimes lacking the clear M&E indicators and criteria lead to the overlapping of the monitoring process of the donor parties with the executive work while ultimately leads to loosing ownership and then the project could be directed towards issues other than the actual priority requirements of the local needs. The lack of a definite body that exercises monitoring and accountability over the leaderships of the governmental bodies that benefit from the loans and grants allocated for reform, development and restructuring so that the potentials for such important activity are not to be wasted. All these factors and phenomena have threatened the life of the project since its early stages and led sometimes to hampering the execution of its specific plans. The project couldn't overcome some local constraints until in a later stage. 56 The tasks and roles of the governmental bodies: It is taken for granted that the reform and change process in the governmental bodies is a comprehensive process that integrates and harmonize within the activities, roles and tasks of the various governmental bodies. On that basis, the tasks accomplished by the CSMP are linked to the defined tasks and objectives and in this field the project has achieved results that are worth appreciation. Therefore, the evaluation of the CSMP is supposed to be relying on the specific tasks and objectives and not to the modernization of the public administration system or even the elimination of all forms of corruption and deficiencies as these are governmental and community goals that require the implementation of a multi-phased reform process with all sectors involved (judicial reform, security reform, financial and economic reform...etc.). This emphasizes the importance of the State's bodies performing their tasks and roles within the State tendencies, strategies and programs. CSMP and MOCSI are part of the governmental set that undertakes specific objectives in the administrative reform track. The outputs of the CSMP have provided several change elements through the republican laws and decrees and the resolutions of the Cabinet along with the revocation of many hurdles facing the public posts. This has made a foundation for change that is supposed to be implemented by the bodies concerned with such laws and resolutions while making use of the successful experiences of other government bodies. In addition, this report will also include drafts for resolutions to benefit from the outputs of the CSMP and use them as basis for a new phase of practical implementation. Lessons learned: The track of the CSMP is an experience to benefit from due to the long lifespan and the various fields included in addition to being the last and most profound project in administrative reform and public posts since the administrative reform is the axis and force behind the comprehensive development. The lessons learned could be shown as follows: 1. Resistance to change: The administrative reform processes usually confront fierce resistance since it is related to the employee and the public post. This pushes the public employee to try to maintain the current situation and fight any change especially in light of the short administrative methods and lack of accountability. Therefore, administrative reform has failed in defeating the challenges more than its successes as it is being fought even by the potential beneficiaries from the administrative reform outcomes. 57 Therefore, it is important to consider the resistance to change as one of the biggest threats in administrative reform especially with the failure of the M&E processes and the principle of accountability. The Project experience has shown that preventing the resistance to change and unifying the visions of the administrative leaderships through training and workshops have left a positive impact but it wasn't a radical solution to this problem. 2. The sense of ownership: The frequent and sometimes the contradictory interventions with multiplicity of supervisors by the World Bank, the level of expertise and knowledge of some of them and the repeated unjustified visits as well as the commanding nature of some of them have caused the local staff to lose their sense of ownership and ability to influence. The international experts who want always to show their boss their ability to unveil the deficiencies and flaws in every visit and to provide advice and guidance to the Yemeni side, are under constant pressure and that sometimes takes away some objectivity. In Yemen, there is a local cadre that is capable of managing and implementing projects with no interventions of the WB experts who are not always working in favor of the projects. 3. The influence of the supporting parties: There are many ways to describe the relationship with the World Bank based on the amount of interest paid by the beneficiary. In our case, the interventions of the WB personnel suggest that they directly run the loan unlike the other forms of relationship when the WB provide the entire loan to the beneficiary in return of a higher interest but minimum interventions by the WB experts which makes the beneficiary country fully responsible for making the project works. The disbursement operation takes 11 approvals from the WB including 5-6 prior consents. We may imagine the period of time it takes to obtain ant service or special goods if we take into account the time space between one procedure and another. Therefore, we may make a comparison between the relationship forms with the WB. 4. The Project phases and overlapping: The CSMP was featured by the various implementation forms: 1. Some activities were prepared and initiated such as; the database, bio-fingerprint and photo, the National Strategy for Wages and Salaries, the Renovation and Creativity Program and the Civil Service Fund. 2. There are other activities for which studies have been prepared in order to implement them in the State bodies (11 studies). 58 There aren't enough funds to initiate the activities of the second group as they have to be put into practice. This requires that the allocations to implement the outputs of CSMP are made available. The future studies shall include the field implementation process in order to make use of any study in direct development and change. In addition, the evaluation of the quality or appropriateness of any study may only be seen through the successful practical implementation. 5. Integrity of the roles of the externally funded projects: It is noted that the relationship between the managements of the externally funded projects is almost absent. There are no regular meetings, workshops or exchange of experiences except for the ones arranged by the donors in rare occasions. This situation doesn't contribute to integration and benefitting from the accumulated expertise. It rather gives contradictory information about the projects leaving a space for the personal impressions on which the attitude of the international expert may be built. It is important to create a mechanism that enhances the integrity and coordination of roles between the leaderships of the externally funded projects, exchange of knowledge and experiences and sharing of support for the success of each project. 6. Benefitting from the projects' outputs: It is well-known and taken for granted that projects are there for a specific period of time- the project lifespan- after which the project achieves its goals with the launch of a new phase which is the implementation of the projects' outputs as a new task for other parties. For instance, the outputs of the CSMP are a number of laws, republican decrees and Cabinet resolutions which represent a new task for the State apparatuses which are to implement them. Thus, it is important to have a clear mechanism to move the projects' outputs to the day to day reality in the beneficiary bodies and under the supervision of the body responsible for following up and evaluating the administrative reform processes. The success of the implementation expresses the success of the reform and change objectives. Preparation and calculation of risks: The government and community preparation and the extent of agreement around the importance of change and reform have been unprecedented in all projects implemented in Yemen. The Project's plan which represented a tool for implementation of the CSMS has envisaged many potential risks either in implementation or objectives. However, the practical implementation faces several difficulties not only at the local level but also and largely by the funding agency. 59 Hence, many unaccounted for risks will always emerge produced by the local reality or made up by the funding agency and these shall be taken into consideration upon planning and specification of objectives. The cumulative summary of the Project's implementation level: Over the past period of time, the Project has managed with efficiency and effectiveness to implement the tasks listed under its first phase and even managed to carry out extra tasks as follows; a. Tasks of the first phase: the Project was able to implement all tasks with a 100% completion. The following table describes the tasks completed; A table showing the tasks of the Project in the first phase S Tasks completed by 100% Level of progress % Component 1: Applying the policies and systems to strengthen the basic 100 systems 1 Personnel Information Management 1-1 Review of the personnel files (build a unified information system and 100 apply it an all units of public service in the central and local authorities) 1-2 Design and implement the functional survey and census 100 1-3 Build and operate a main information center and ancillary centers in the capital secretariat and governorates 1-4 Import and install the fingerprint and photo system, develop the 100 operating schemes, train employees, prepare the functional card system, prepare the procedural manual and the executive plan to implement the functional card and all the requirements that enable the proper implementation of the functional cards. 1-5 Adopt a transparent recruitment scheme and review the updated 100 recruitment policies and procedures enabling recruitment based on efficiency and equal opportunities, set and implement a ten-year plan for recruitment through an integral e-system. 1-6 Build the retirement system and strictly implement it 100 2 Unify the methods and schemes of preparing the payrolls within one 100 system based on unified rules and procedures for additions, downgrading and deductions. 3 Develop a strategy for IT 80 4 Draw maps of the human resources management information system 70 (system engineering) 5 Develop the standardized system for job description and classification 6 Build and implement the functional settlements system 7 Build and apply the e- system for non-Yemeni labor force 8 Build a system for replacement of the non-Yemeni cadres with Yemenis. 9 Develop the National Strategy for Wages and Salaries Component 2: rationalizing the number of employees in the governmental 100 60 bodies 1 Apply the retirement procedures through the database 2 Approve the deletion of the double and shadow employees through the fingerprint and photo system and downgrade a number of double employees 3 Prepare the legal references, policies and regulations and set mechanisms that enabled the proper operation of the Civil Service Fund Component 3: rationalizing the volume of the governmental apparatus 100 1 Define the State's current and future role and function 2 Conduct restructuring studies for (7) units; MOCSI, Tax Authority, Customs Authority, the Central Organization of Control and Audit, Ministry of Social Affairs and Labor and Ministry of Public Health and Population) 3 Set up a program for renovation and creativity in public services development and prepare the legal references, regulations and executive procedures and finalize the first phase of the project 4 Define the current and future role and function of the State 5 Restructure the internal auditing 6 Develop a system for institutional performance evaluation 7 Develop and simplify the government service provision procedures Component 4: building and developing the institutional capacities 100 1 Train the middle and senior leaderships, technical training for users of the basic systems and train the focal teams in the selected administrative units 2 Develop the M&E systems and issue manual 3 Prepare a methodology for strategic planning and issue the manual 4 Set a functional analysis tool 5 Develop a methodology for restructuring and issue the manual 6 Develop the national training plan 7 Masters program in public management (Sana'a University) 8 Masters in business Administration (Sana'a University) 9 Specialized certificates in public management (Sana'a University) 10 International certificate program (computer + Microsoft Project) Tasks being implemented away from the tasks of the first phase: The tasks being implemented S Ongoing tasks Progress level 1 Implement the fingerprint and photo system in the judicial authority, 100%, 86% military and security sectors 87 respectively 2 Review the Civil service Law in light of the strategic tendencies 50 3 Build a central database and key databases for the judicial authority, 40 military sector, security, retirement and social welfare. 4 Refer the redundant labor force to the Fund based on the restructuring 50 process 61 5 Simplify the government service provision procedures to the public and Ongoing the private sector through examining and inventorying all the government services and collecting all the data related to the provision of each while studying the followed procedures and examining their legality. The provision procedures are unified and the illegal services eliminated. Prepare a manual for the government services, document and print all service manuals on guiding boards in the entrances of each government body. These will also be published through the internet and promoted through the media outlets for all the units of local authority. 6 Implement the first phase of the creative projects (won by the emergency 80 services in Al-Jumhouri Hospital, scholarship services in the higher education, simplifying the civil service procedures, tracking the transactions in communication, the Information system of Al-Mukalla Airport, the ATM in the Post Office). 7 Restructure the National Institute for Administrative Sciences 20 8 Apply the restructuring studies at the practical level in the State agencies 30 and institutions in coordination with such parties. Summary of Borrower's comments on draft ICR This section provides a summary of the comments received from the Project Unit at the Ministry of Civil Service and Insurance on the draft ICR documents. To the extent feasible we have tried to incorporate the project unit's comments (as addressed in section 7 of the ICR). The borrower points out that the Project has created direct and indirect impacts on the management, employees and the administrative leaderships in strengthening the reform, development and change orientations, through the individual reform initiatives in various organizations, ministries and through the Republican laws and resolutions as well as the resolutions of the Council of Ministers in a number of about (20) laws and resolution all supporting the reform and change process. All these are considered outputs of the Project in addition to the fact that the Project has realized savings in the budget of the State equivalent to the value of the loan. The project unit felt that certain conclusions and opinions were very rigorous and did not observe the circumstances, stages and impediments the Project faced neglecting taking into account the complicated and difficult situations existing before the start of the Project and the new situations ending up to mid 2008. The lack of activating monitoring and evaluation by the Bank and the multiplicity of supervisors of the Project during the life span of the Project renders evaluation through the final report short of conceiving the developments and tracing the multiple activities and outputs of the Project. COMPONENT ONE: CORE SYSTEMS National Wages and Salaries Strategy 62 ONE: The total cost of the increase of the salaries and wages bill during the years 2006 until the end of 2009 reflected in the estimates of the year 2010 amounted to YR 392 billion in a percentage of 134% compared to the year 2005. This cost is opposed with financial decreases due to administrative reforms and treatments of a number of imbalances, illegal situations and the review process of the salaries statements carried out during the implementation of the strategy and when the budget drafts were prepared. This has entailed net increase in the wages bill of an amount of YR 293 billion. TWO: The implementation of the national strategy of wages and salaries and the granted increases were linked to the implementation of a package of reforms and treatment of numerous imbalances and illegal situations. Without the reforms and remedies implemented before and during the implementation of the strategy and the law for a number of imbalances in the system of employments, salaries and wages and in the salaries statements those savings could not have been realized. Examples include: Elimination of all employees who have reached either term, the deceased and the permanently disabled after being transferred to the general scale and referring them to retirement in a number of 70,021 employees. Limit all redundant workforce and referring them to the Civil Service Fund in a number of 14,210 employees. Elimination of the double dippers, those who discontinued employment and the ghost employees who were unveiled with a number of 15,663 employees. Replacement of Non Yemeni employees by Yemeni nationals Deduction of the financial amounts which were disbursed illegally from the budgets and authorizations of agencies resulting from the implementation of reforms and removing imbalances reflected on the public budget of the State and preparation of the budgets of agencies based on the actual implementation of the strategy. Three: Double dippers and ghost employees: Reforms and remedies performed before and after implementation of the national strategy of wages and salaries and the correction of those imbalances and illegal situations (double dippers and ghost workers) resulted in exclusion and elimination of the majority of those cases from salaries lists and monthly authorizations. This appeared as actual financial savings (in number and cost) in the budgets of the agencies. These savings off-set increase in wages and salaries as a result of (i) the implementation of the wage strategy, additional allowances such as the grant of work nature allowance to a number of sectors (education, health, information and COCA), (ii) other increase of wages and salaries such as new employment, and continuous employment by replacement of double dippers, those discontinued and dismissed, ghost employees, retirees. In this respect it is important to keep in consideration various factors (prevailing economic condition, social dimension/entitlement to public employment, trade union pressure) impacting the application process and hampering the required reforms and limit the wages bill. 63 BIOMETRICS, INFORMATION SYSTEM AND DATABASE The consolidated salary system statement was prepared and circulated to all central and local public service units where all salaries statements are now issued in accordance with the consolidated system but difficulties exist in networking with the public service units. The biometric system is a continuous process and matching shall be performed when the highest percentage of biometric system is achieved in a percentage of approximately 99% and then matching shall be carried out between different civil, military and security sectors. In regards to obtaining the database of the sectors pursuant to the Republican Resolution no. (1) of the year 2004 related to the establishment of a central database and establishment of the necessary databases (demographic data) of which the basic columns for the issue of employment cards to all sectors we have prepared them and they are perfectly ready. In regards to unveiling the double dippers and ghost employees on the pay statements the Ministry of Civil Service represented by the information technology center was not lagging in unveiling and treating such cases and report to the Ministry of Finance to abolish them and consequently clear the salaries statements of such cases. The most notable impediments facing the Ministry in the process of networking the public service units as well as quality training of technicians to cope with the technological developments is that some systems and devices use systems not being taught in universities. COMPONENT TWO: RATIONALIZATION OF EMPLOYEES SIZE: ONE: With regard to the establishment of the Civil Service Fund, a law was effectively issued establishing the Fund under no. (1) of the year 1999 and its executive bylaw no. (14) of the year 2000. Upon preparation and drafting the technical, financial and organizational operation bylaws and systems during the subsequent period to the issuance of the Fund law assisted by a specialized technical team and getting assistance of certain international expertise and experience of the World Bank it became evident that some gaps and shortcomings exist in the Fund law. Based on this the law of the Fund and its executive bylaw were amended by issuance of law no. (1) of the year 2004 and the Republican Resolution no. (184) of the year 2004 related to the executive bylaw of the Fund law and thereby the Fund moved to actual operation commencing the year 2005 after completion of all legislations, systems and executive instructions and procedures necessary for operation. TWO: The Fund work strategy was not limited to the reduction of the redundant employees of the ailing economic units only but is seriously ongoing to reduce the redundant employees of different standing public service units mainly those of the public administration units which is currently under progress. In view of the level of implementation of redundant employees reduction strategy until the year 2010 we confirm the following: 1- Phase one implementation is finalized by completion of treatment and reduction of redundant employees in the ailing economic sector units as well as employees referred from the public and partisan organizations with the public administration units whereas a number of 16807 cases of employees were treated as follows: Referral of 2636 employees who reached retirement age 64 Reduction by early retirement and financial compensation of 13071 employees Early retirement of 12122 employees in a percentage of 92.7% in a financial cost of YR 2,643,403,331 equivalent to USD 12,294,899 Financial compensation to 949 employees in a percentage of 7.3% with a financial cost of YR 2,107,289,545 equivalent to USD 9,801,346 Redistribution to the public service units of 1100 employees of those redundant with experiences and qualifications 2- The second phase of the redundant employees' reduction program is initiated during the 2010 through the following: a- Counting the redundant employees existing in the standing public service units who are not filling any positions in light of the first and second phases results of the wages and salaries strategy by obligating the public service units to submit the names of those idle. The public service units prepared preliminary lists of the redundant employees for 12,000 employees. b- Preparation of documents for the criteria and procedures of referring those employees to the Fund and application of the referral shall be launched in the year 2011 following approval of the competent authorities. c- During the year 2009 - 2010 the number of redundant employees was reduced in a number of standing public sector units which completed the re-building and restructuring process namely: National Shipyard Company for 167 employees Yemen Airlines Company for 1000 employees Public Corporation for Fisheries Services and Marketing for 120 employees Public Corporation for Coastal Fishing for 370 employees THREE: The low acceptance by redundant employees of the severance pay compensation package (less than 8%) and preferring early retirement can be large explained by (i) the characteristics of the reduced employees (average age is 45 - 50 years and average service is 25 years, and 65% are unqualified and holders of preparatory school certificate or below), the economic circumstances represented by low economic standards and continuous rise of the prices and depreciation of the currency and high rate of dependency as the average number of households is 7 members. Consequently, most employees avoid the risk of receiving compensation amounts and seek private enterprises in order to gain a fixed income through retirement pension. By consequence, reducing more than 17,000 employees until the year 2010 has certainly led to decreased inflation of the government apparatuses. In view of the financial savings realized for the budget the reduction of the number of employees resulted in the decrease of the salaries bill by an amount of YR 5,112,077,380 equivalent to USD 23,777,104. The total realized savings resulting from the reduction of the cost of their stay in the public employment (for the remaining period in the employment until reaching retirement) is an amount of YR 44,266,864,462 an equivalent to USD 205,892,392. 65 COMPONENT THREE: RE-BUILDING AND RESTRUCTURING COMPONENT: In different documents of the Project the studies were forming a preliminary stage for the application of the re-building and restructuring in pilot agencies. One may refer to the contracts with consultancy firms based on which the studies approved by the World Bank were implemented to have the studies applied in a subsequent stage. Whenever the Project finds an opportunity for application it proceeds the same such as the procedures for simplification of the government services delivery progressively applied by the Project in the State apparatuses. In the outputs of this component, there are achievements that could not be easily measured as the activities of this component have created a state of awareness to the importance of the strategic planning in a number of government agencies. Those agencies initiated preparing their targets, messages and strategic plans and adoption of the rebuilding and restructuring process as a method to create the change. Those agencies received direct technical assistance from the Project and a support through a set of resolutions of the Council of Ministers. These results represent the start to transfer the leverage of the development and change process to the government agencies themselves by incorporating the development and reform process within the components of their activities contained by their annual plans. COMPONENT FOUR: CAPACITY BUILDING: Through the component of capacity building (29) training programs and workshops were implemented targeting the leaderships of the ministries and the government apparatuses and participants in different activities of the project. All activities implemented through this component overseen by numerous supervisors of the Bank were approved in advance without any objection to any proposed training program in the Project. Finally, the implementation of those programs has created a substantial leverage on transformation and change of opinions and practices of the higher leaderships towards understanding the fundamentals and bases of the strategic planning and adoption of the rebuilding and restructuring methodology as a tool for change. Annex 6. Comments of Cofinanciers and Other Partners/Stakeholders None 66 Annex 7. List of Supporting Documents PAD, Financing Agreement CSMP PAD, Financing Agreement, CSMP supplemental financing Restructuring MEMO 12/15/2005 Restructuring MEMO 12/20/2007 Management review meeting minutes and memo dated 10/26/2004; 12/03/2004; 07/28/2005 CSMP Implementation Supervision Reports (ISR) CSMP Aides Memoires IEG Country Assistance Evaluation, 2006 IEG Civil Service Reform Evaluation, 1999 QAG Lending Portfolio CSMP assessment, 2009 QG Quality of Supervision Assessment (QSA7), 2007 CSMP Mit-term review, 2003 CSMP performance audit, Geoffrey Shepherd, July 2003 KPMG AFMIS testing, 2009 HRS external assessment, Bill Monks, 2010 Dutch Trust Fund external evaluation, 2008 GoY CSMP assessments, including joint MoF-MOCSAI working group resolutions GoY CS related laws and regulations IRIS project documents and correspondence IMF art. IV mission reports 67 46ºE 48ºE 50ºE 52ºE 54ºE REPUBLI C O F This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information Y EM E N shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 20ºN SELECTED CITIES AND TOWNS 20ºN GOVERNORATE CAPITALS REP. OF SAUDI ARABIA NATIONAL CAPITAL YEMEN RIVERS MAIN ROADS GOVERNORATE BOUNDARIES INTERNATIONAL BOUNDARIES OMAN 18ºN 42ºE To Abha 44ºE Sanaw To Thamarit Habarut Thamud SAADAH Mah rat To Mirbat Saadah n Mt hu AL- ns. ak di M AL-JOWF Wa M A H R AT Zamakh HADRAMOUT Damqawt Maydi ¸ - Huth adi Al Jiz t W Al Hazm ser Al Ghaydah 16ºN HAJJAH AMRAN Wad n De Wad i Had Mar'ayt 16ºN tay i Jaw ram b'a f Al Ghuraf Nishtun ou Al Luhayyah Hajjah Sa t Amran t as Hawra MARIB mla Salif Al Mahwit Marib Ra Shabwah AL- SANAA t Qishn MAHWIT SANA'A a u o S H A B WA m Sayhut r a Re Al Hodeidah Al Gabain DHAMAR Harib Nuqub d HODEIDAH DAH a Shihr d (3199 m) Nisab Ataq H RAIMH Dhamar Rida Al Mukalla AL-BEIDA Se Wad Zabid i I B B AL- Al Huwaymi a 14ºN Al Beida 14ºN Ibb DHALE'E (3227 m) Bir Ali Al Dhale'e Taiz A B YA N W ad B i i Ahwar TA I Z ana Mocha Shaqra At Turbah ERITREA Lahej Zinjibar Gulf of Ade n At Turbah LAHEJ Qadub Aden Qalansiyah ADEN ETH. Socotra 'Abd al Kuri Samha ) DJIBOUTI 0 50 100 150 Kilometers Darsa TE IBRD 33513R A 12ºN (HA The Brothers N OR 12ºN DRAMO T GOVER U MAY 2009 0 50 100 Miles 42ºE 44ºE 46ºE 48ºE 50ºE SOMALIA 52ºE 54ºE