Mozambique - Capacity Building : Public Sector and Legal Institutions Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 21415 IMPLEMENTATION COMPLETION REPORT (IDA-24370; TF-23425) ON A CREDIT IN THE AMOUNT OF SDRS 10.9 MILLION (US$15.5 MILLION EQUIVALENT) TO THE REPUBLIC OF MOZAMBIQUE FOR A CAPACITY BUILDING: PUBLIC SECTOR AND LEGAL INSTITUTIONS DEVELOPMENT PROJECT MARCH 15, 2001 Public Sector Reform and Capacity Building Unit Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Metical (MZM) - pi. Meticais US$1 = MZM 2,809 (10/29/1992)) = MZM 15,610 (06/30/2000) MZM I = US$0.000356(10/29/1992) = US$0.000064 (6/30/2000) FISCAL YEAR January 1 to December 31 ABBREVIATIONS AND ACRONYMS ADR Alternative Dispute Resolution BR Boletim da Reptublica, Bulletin of the Republic CAS Country Assistance Strategy CEJ Centro de Estudos Juridicos, Legal Studies Center CNFP Conselho Nacional da Funcao Puiblica, National Council for the Public Service DAF Departamento daAdminsitraado e Financas, Department of Administration and Finance DCA Development Credit Agreement DO Development Objectives DPAC Departamento Provincial de Apoio e Controle, pronvincial office of MAE FILMUP Zambia Financial and Legal Management Upgrading IDA International Development Agency IMC Inter-Ministerial Committee IP Implementation Progress IPAJ Instituto Para Assistencia Juridica, Institute for Legal Aid MAE Ministerio da Administraqo Estatal, Ministry of State Administration MOJ Ministry of Justice MSF Management Strengthening Fund MULEIDE Mozambican Women's Law Association NDA Nzkcleo de Desenvolvimento Administrativo, Administrative Development Unit NGO Non Governmental Organization PMU Project Management Unit PU Planning Unit SAR Staff Appraisal Report SCR Sistema de Carreiras e Remuneraqao, Career and Remuneration System SIDA Swedish International Development Agency SIF Salary Incentive Fund UEM Universidade Edoardo Mondlane, Edoardo Mondlane University USAID United States International Development Agency WG Working Group Vice President: Callisto Madavo Country Manager/Director: Darius Mans Sector Manager/Director: Brian Levy Task Team Leader/Task Manager: Harry Garnett/Jocelyn Mason  FOR OFFICL USE ONLY MOZAMBIQUE CAPACITY BUILDING: PUBLIC SECTOR AND LEGAL INSTITUTIONS DEVELOPMENT PROJECT IMPLEMENTATION COMPLETION REPORT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 10 5. Major Factors Affecting Implementation and Outcome 17 6. Sustainability 19 7. Bank and Borrower Performance 19 8. Lessons Learned 21 9. Partner Comments 23 10. Additional Information 23 ANNEXES: Annex 1. Key Performance Indicators/Log Frame Matrix 24 Annex 2. Project Costs and Financing 25 Annex 3. Economic Costs and Benefits 27 Annex 4. Bank Inputs 28 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 29 Annex 6. Ratings of Bank and Borrower Performance 30 Annex 7. List of Supporting Documents 31 Annex 8. List of Persons Interviewed 32 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization.   - 1 - Project ID: P001810 Project Name: Capacity Building: Public Sector and Legal Institutions Development Project Team Leader: Harry C. Garnett TL Unit: AFTI2 ICR Type: Core ICR Report Date: February 12, 2001 1. Project Data Name: Capacity Building: Public LIC/TI Number: IDA-24370; Sector and Legal Institutions TF-23425 Development Project Country/Department: MOZAMBIQUE Region: Africa Regional Office Sector/subsector: BI - Institutional Development KEY DATES Original Revised/Actual PCD: 11/14/90 Effective: 06/01/93 02/28/94 Appraisal: 04/01/92 AdTR.: 11/30/95 11/05/96 Approval: 11/19/92 Closing: 06/30/99 06/30/2000 | orwerlImplemflentiingAgency: | GOMITBA l | ~~~~Other Partners: I STAFF Current AtAppraisal_ Vice President: Callisto Madavo Edward V. K. Jaycox Country Manager: Darius Mans David Cook Sector Manager: Brian Levy Steen Jorgensen Team Leader at ICR: Harry Gamett Jacomina De Regt ICR Primary Author: Jocelyn Mason 2. Principal Performance Ratings HS - Highly Satisfactory, S = Satisfactory, U = Unsatisfactory, HL = Highly Likely, L = Likely, UN = Unlikely, HUN = Highly Unlikely, HU = Highly Unsatisfactory, H =High, SU Substantial, M = Modest, N = Negligible Outcome: S Sustainability: L Institutional Development Impact: M Bank Performance: S Borrower Performance: U l __________;_____.____ ____ .______ QAG (if available) Quality at Entry: U Project at Risk at Any Time: Yes DO and IP ratings unsatisfactory for one year prior to Mid-Term Review  - 2 - 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: 1. Work on developing the Legal and Public Sector Capacity Building Project began in 1990, a time when civil war was still raging in Mozambique. The end of the Cold War and the fall of the apartheid regime paved the way for an end to hostilities in Mozambique, and the General Peace Agreement was signed in October 1992. Over the next two years, the Government reformed macro- economic policies and introduced more liberal constitutional provisions, demobilized 80,000 soldiers, resettled several million refugees and displaced families back to the countryside, and began the process of re-establishing its administrative presence throughout the national territory. The first multiparty elections were held in 1994, and over 80 percent of the eligible population voted. Peace also brought an acceleration of the process of transforming the Mozambican economy from a centrally administered to a market system. The Legal and Public Sector Capacity Building Project, along with its twin, the Human Resources Capacity Building Project, were therefore being developed at a critical time in Mozambican history, in which large changes were taking place rapidly. 2. Bank engagement in Mozambique had begun in the mid 1980s, with adjustment operations in 1985 and 1987 in support to the country's economic reform program. The adjustment process was accompanied by a series of investment projects, mostly emergency-type projects aimed at mitigating the effects of adjustment and of the civil war. Hindering Mozambique's development, and thus more narrowly the implementation of the Bank's and all other donors' effort to support the Government's program, was an extremely serious shortage of capacity, in both the public and the private sectors. The public sector in particular lacked skilled individuals of any level, and especially well-educated and experienced people capable of providing leadership in planning and management. In addition, the public sector was increasingly unable to compete for the scarce talent available. This was due to its relatively much poorer pay (that was being eroded by devaluation of the local currency), employment conditions, and work environment in comparison to donor-supported projects and to the nascent private sector. The Legal and Public Sector Capacity Building Project was part of an effort, based on the findings of a substantial study undertaken by the Bank and the Swedish International Development Agency (SIDA) among others, to maintain and expand local capacity to plan and manage Mozambique's development. As originally designed, a single project was meant to respond both to the need to improve the public sector's effective demand for (incentives and conditions of employment and professional training, both pre- and in-service) and the supply of trained and qualified personnel. However, the project was deemed too large and complex, and was divided into two. Between them, these projects were expected to result in thousands of additional Mozambicans trained in professional domains, and that concurrent actions in the areas of public pay and employment, the reform of public institutions and the management of technical assistance would encourage these professionals to join and remain in the civil service. 3. It was (and remains) easy to underestimate the capacity constraints facing Mozambique in the early 1990s. For example, there were only 90 lawyers in a country of over 16 million in 1990; there were almost no judges with formal legal training - in Nampula province, the country's most populous, there was one judge for the whole province; in the public service two-thirds of senior civil servants had only secondary level education, one third had only primary education; and in the provinces the situation was much worse, only 74 provincial directors out of 206 had schooling beyond the primary level. Compounding these problems, the Government had no reliable mechanism to control its civil service: in 1989, no ministry had an accurate tally of who was on its pay-rolls; some ministries did not have a human resources department; rules were unclear and lines of authority were not formally established; and there were over 6,000 different job categories, with no ministry sharing the same qualification requirements. A process for achieving effective establishment control was only completed in 1999, when 89 percent of the 105,000 civil servants were accounted for, 'ghosts' and  - 3 - other anomalies removed from the payroll, and all civil servants were included on a unified information system. Indeed, the situation in Mozambique was so parlous that even the Government consistently overestimated what it was capable of achieving. The implications of this for the project were that the Bank and other donors were compelled to start from a very low base. Limited objectives and a very clear and contained set of outputs were the most that could be achieved over the short-term, or that the Government could absorb and to which it could be committed. 4. The Legal and Public Sector Capacity Building project's main objective was to build and maintain capacity in key public institutions and skill areas by expanding the supply of well-trained senior planners, policy analysts, managers, and technicians, and by enhancing pay and other incentives and conditions of employment for senior civil servants. Specific objectives included: (a) strengthening legal institutions and professional capabilities; and (b) developing public administration and management skills and enhancing civil service personnel systems. 5. Experience over the decade since the preparation of the project began has shown that the strategic focus of the two components of the project was appropriate. By focusing on improving civil service administration in one component, the project was supporting a critical first stage of a process of public sector reform, in which the Bank is now playing a central role. The Bank should also be commended for taking what was, for the time, a rather advanced view of the importance of legal reform to private sector development. The Bank had not typically addressed this issue, and the Government and donors in Mozambique had not fully recognized its importance. Now reform of the legal sector is one of the primary concerns of the Government's reform program and the Bank's CAS for Mozambique. Inclusion of the legal component played an important role in stimulating interest in the sector, beginning to address the issues, and laying the ground for future Bank involvement. 6. Nevertheless, the objectives of this project suffer from a number of weaknesses. First, the objectives are very broad, and give an inadequate idea of what would be achieved within the limited time and resources available during the project. As a result, the project did not help to prioritize the actions it was supposed to support. Moreover, in the absence of concrete indicators, it is difficult to assess whether the project has achieved its objectives. This first weakness was compounded by a second, that is the project overestimated the capacity of the project's implementing agencies . In the words of one Bank task manager, there was a "lack of capacity to build capacity." Third, there was little link in practice between the objective of strengthening legal institutions and that of developing the public administration. In effect the project amounted to two entirely separate sub-projects, with no connection between them other than the Bank task manager. Finally, the design of this project was influenced by the emergency-type thinking that was informing most other projects at the time, but at the same time it embraced longer-term ambitions. The fact that project objectives were shaped loosely, as might be appropriate in an emergency-type operation, derives from this type of thinking. Similarly, the way the salary incentives fund was developed (see paragraph 56 below) bespoke a short- term solution in lieu of a more systematic reform. Emergency-type thinking combined poorly, however, with the inherent nature of both long-term policy reform and the nuts-and-bolts reforms which formed the core of the project. 3.2 Revised Objective: 7. The project was revised through several amendments, which had the combined effect of significantly reducing its scope. With these amendments, the project now aimed at more modest practical steps aimed at laying the foundation for more embracing reform.  -4 - 3.3 Original Components: 8. The original project was split into two in response to concerns, expressed during the Yellow Cover review meeting and echoed during negotiations, about the supervision requirements of the proposed project, given the size and complexity. The other project was the Human Resources Capacity Building Project. Briefly after negotiations, the Legal and Public Sector Capacity Building project was to be a technical assistance project, which did not require a Staff Appraisal Report (SAR), but before going to the Board management decided to present an SAR for this project as well. While these last minute technical changes need not have, in themselves, adversely affected the outcome of the project, they suggest ambiguity on the part of the Bank about the nature and scope of the project's objectives, and the best instrument to achieve them. 9. An Inter-Ministerial Committee on Capacity Building was meant to coordinate the activities of the two projects, as well as a broad range of other capacity building activities in Mozambique. However, the links in practice between the two projects were weak, and, were it not for the limited size of each component, could have as easily been split into four separate projects. During implementation, the connection between the twin projects weakened further. In the first years of implementation they shared a task manager, but from mid-term onwards the connection had faded completely. For the borrower there was never a practical connection between the two projects. 10. The Legal and Public Sector Capacity Building project, as noted, consisted of two completely separate components, one for each of the two main objectives. There was no practical link between these components. The Strengthening Legal Institutions and Training component had a base cost of US$7.2 million, while the Public Administration and Development Management component had a base cost of US$6.5 million. 11. The Strengthening Legal Institutions and Training component was implemented under the auspices of a Legal Working Group comprising representatives of each of the five major legal institutions. These were the Ministry of Justice (MOJ), the Supreme Court, the Administrative Tribunal, the Attorney General's office, and the Faculty of Law at Eduardo Mondlane University. The MOJ was the main counterpart for the Bank, and was meant to be the 'first among equals' as the Government's line agency for legal affairs. The component's objectives were "to create a group of competent lawyers capable of designing and implementing a sophisticated legal and regulatory framework underpinning a market-oriented system; and build a corps of lawyers, paralegals, judges and court officers to ensure the proper administration ofjustice." It consisted of activities including: a) Design and implementation of a long-term strategy and action plan to strengthen the legal profession and the judiciary. This included support to three workshops to a) review the capacity, scope of work, institutional arrangements, and staffing of the main legal institutions, and b) recommend eventual adjustments to the functional, organizational, staffing, administrative, and professional aspects of these institutions; to fund diagnostic studies of the Law School with a view to improving the training programs provided; to support a planned National Conference on the legal profession in 1996; and by strengthening the legal institutions through training and exposure to other legal systems and traditions (US$800,000 or 11%). b) Improvement in the quality and efficiency of legal services, as well as training for the lawyers and judiciary. This included professional and in-service training for legal practitioners, law teachers, judges, magistrates and students, as well as the design and implementation of a pre-service and in-service training program as part of a follow-up of a  - 5 - pilot launched by the MOJ; and support for legal reform as it relates to private sector development (US$3.6 million or 50%). c) Facilitating the dissemination of the law and the administration of justice through the design of modem automated legal information systems; upgrading resources and operations of law libraries; assisting the legal reform process through providing consulting services, conducting seminars, and preparing legal materials. This included improving the legal libraries, notably the acquisition of books; and support to a modem legal information system (US$932,000 or 13%). d) Assisting in the development of legal institutions, such as the Administrative Tribunal, the Center for Judicial Studies, the Bar Association and the women's law association, MULEIDE (US$640,000 or 8%). e) Strengthening the institutions responsible for Project Implementation, and funding for the component management unit (US$1.3 million or 18%). 12. The Public Administration and Development Management component was to be implemented by the Ministry of State Administration (Ministerio da Administracao Estatal, MAE). It was designed in close coordination with and to supplement SIDA's public administration program. It consisted of the following sub-components: a) The establishment of a Planning Unit within MAE, staffed by a combination of local staff, locally-recruited advisors and foreign consultants. The unit was to be created as a condition of effectiveness, and its functions were to i) act as secretariat for the Capacity Building Inter-Ministerial Committee (IMC), including developing a plan for implementing the National Capacity Building Strategy; ii) develop strategies and implementation plans for civil service reform; iii) develop an institutional strategy and plan for MAE itself (US$580,000 or 9%). b) Support for the ongoing program of reforming the civil service personnel management system. The first phase of this program had been funded by SIDA; the project was to add IDA support to SIDA's for a second phase, which would include integration of all civil servants into he information system and the introduction of a new standardized career and remuneration system (US$1.8 million or 28%). c) Support to MAE's institutional development, including assistance to the Department of Administration and Finance (DAF); the provision of scholarships; and a contribution to a local cost fund to support MAE's leadership and coordination role in relation to civil service reform and overall capacity building (US$1.1 million or 17%). d) Financing for a Management Strengthening Fund, an internal grant making mechanism to support innovative reforms and capacity building proposals made by public sector agencies, to support management improvements within the public service and the development of emerging NGO professional organizations (US$3 million or 46%). 13. The component also was responsible for the administration of a salary incentives fund (SIF) to be funded by other donors. While the SIF was not to be funded by credit funds, but rather from bilateral grants, it was conceived as an integral part of the government's capacity building program and its goals, modalities and management mechanisms were evaluated as part of the Bank's appraisal of the program.  -6 - 3.4 Revised Components: 14. There were three major amendments in all, and a number of minor ones. The first of these took place as the project became effective. It recognized that the independent legal institutions would not recognize the authority of a Legal Working Group over the planning and allocation of component resources, even if constituted by a representative from each of these institutions. This amendment reduced the role of the Working Group to one of simple coordination without any decision-making powers, in effect splitting the component into five mini-components. With this amendment the development of a strategy for reform for the legal sector as a whole was also implicitly abandoned. This failure to relinquish leadership of the sector to one authority is a reflection of the weakness of the institutions themselves, and of the Ministry of Justice in particular. This latter ministry had neither the technical capacity to guide reform in the sector, nor the political strength to assert leadership. At the same time, no other legal institution was in a position to substitute for it. 15. The second amendment took place a year after effectiveness. It abolished the Planning Unit (PU) within MAE, and the IMC. Some of what would have been the PU's responsibilities were transferred to a newly-created Nuicleo de Desenvolvimento Administrativo (NDA). The difference between the NDA and the PU was that the latter was conceived to have a more executive role-as the project management unit for the capacity building project, as well as in supervising the implementation of reform and capacity building activities both within and beyond MAE and reporting on them to the Bank and to the Government via the IMC. Ministry officials contended that the PU duplicated activities of bodies that already existed within the Ministry, and proposed a more limited role for the NDA, as a sort of in-house 'think-tank' for producing position papers and seminars to inform the Ministry's ongoing activities. As part of the same amendment, the policy and oversight role of the IMC was reallocated to the existing Conselho Nacional da Funqio Puiblica (CNFP), a sub-committee of the Council of Ministers headed by the Minister of State Administration. The CNFP had a less ambitious role than that of the IMC, limited to civil service management reform. Although the DCA continued to give NDA the role of secretariat for the CNFP, this role was no longer necessary or possible given the new character of the NDA. The effect of these changes was to sharply reduce the policy and institutional reform agenda of the project, so that it simply supported the civil service reforms already under way and managed by the National Directorate for Public Administration (Direqcdo Nationalpara Fun!do Ptiblica, DNFP). 16. The third amendment took place at mid-term. This amendment abolished the Management Strengthening Fund (MSF), and cancelled US$3 million from the project. The MSF had remained dormant since implementation, and MAE was clearly neither willing nor capable of administering it effectively. It is a measure of MAE's implementation weakness that, while indisputably short of resources and in need of support for reform activities, it was unable to avail itself of $3 million of relatively 'easy' money. 17. Also, several minor amendments re-allocated resources across categories within the components, mostly with the effect of increasing the resources spent in the Legal component on training, and on increasing the resources in MAE spent on purchasing equipment as part of the computerization of the new career and remuneration system. Finally, the project was extended by one year to allow for the complete introduction of the new civil service career and remuneration system and complete work on its computerization. These amendments reinforced the impact of the major amendments.  -7 - 3.5 Quality at Entry: Project Design 18. Project design suffered from a number of weaknesses. The idea of 'capacity building' as it was conceived during project preparation was new to the World Bank's Africa Region at the time, and this was the first of three early projects in the Region to attempt to put it into practice. Preparation of the other two projects, the Tanzania Financial and Legal Management Project (TZ-2817) and the Zambia Financial and Legal Management Upgrading (FILMUP) Project (ZM-8258) was launched shortly after this project. 'Capacity building' has since been superceded as an organizing concept for project design, because of its excessively all-embracing nature, because it lacks specificity, and because it inevitably takes an overly determinist, 'top-down,' approach. The design of this project reflects these shortcomings, mainly in the lack of clarity in its objectives, and by seeking to establish the IMC and the WG to coordinate and drive capacity building in their respective sectors. Moreover, it appears the Bank was not fully aware of the supervision demands of this type of project; the need for high levels of supervision in Mozambique in general, and for institution building and technical assistance projects in particular, is a lesson that has become clearer with more recent experience (see for example the Implementation Completion Report for the Local Government and Engineering Project.) 19. Second, the project included significant policy-reform elements, to be headed by the PU. These related particularly to the integration of the district, provincial and central levels within the management career structure and to other aspects of the incentive structures for civil servants. Many of these reforms were already being addressed in an incremental way by the human resources management reform activities supported by SIDA and to be further supported by this project. The more sweeping reforms implied in the SAR to be led by the PU seem not to have been at the time properly understood or accepted by MAE. The SAR also speaks of reform of the public sector more generally, which the PU was meant to help coordinate. As was typical of a number of projects at the time this one was designed, while the project aimed to address broad policy issues, it did so with technical reforms as a starting point. Based on the limited success of this approach, recent projects, throughout the Region, have tended to tackle the policy issues directly, according to a clearly defined policy agenda developed as part of the preparation of the project. 20. The effect of the amendments was to address these failings, refocusing the project away from leading broad sectoral reforms onto more narrowly defined institutional strengthening and training activities. The Strengthening of Legal Institutions and Training component now focussed primarily on training legal personnel and facilitating the dissemination of laws, i.e. the items described in paragraph 11 items b) to d) above. Similarly, the Public Administration and Development Management component now focussed mainly on the pay and remuneration system reform, institutional strengthening of critical departments within MAE, and training. In short, the Bank shifted its approach to one that was more 'incrementalist,' more appropriate to Mozambique's evident implementation capacity constraints, and to the complexities, unpredictabilities, and need for long termn engagement inherent in this type of reform. Recent thinking, both in Mozambique and in the Bank more generally, supports this approach. However, although many of the problems in project design were addressed through the amendments, some remaining difficulties, such as the delays in effectiveness and implementation caused by the poor management structures, and the awkward connection between the components of the project, could not be corrected. These design flaws remained at the root of many of the project's future problems. Implementation Arrangements 21. The implementation arrangements as originally conceived did not, in practice, meet the needs of the project. Firstly, the Bank did not adequately account for the weakness of the MOJ. The MOJ  - 8 - was seen by the project as the natural, indeed unavoidable, leader for the Government's reform program of the legal sector. However, the MOJ fell short of being able to perform this function adequately, and was itself in need of substantial institutional strengthening. The MOJ appears to have been often absent during the preparation and the early implementation stage of the project, and it demonstrated a low level of commitment to the project from the start. More focus on reform and strengthening of the MOJ would seem to have been a prerequisite for that ministry to play the role defined for it. 22. After the first amendment, the WG became merely the coordinating body for five separate sub-components. The legal institutions retained the PMU to serve them each separately in facilitating their dealings with the Bank. Specifically, the unit handled day-to-day relations with the Bank, managed the procurement process, and tracked expenditures and provided all reporting requirements. It did not monitor the quality or effectiveness of activities and did not initiate activities, all of which was the responsibility of the individual institutions. This unit worked well in the limited role it had. In fact, despite its limited official role, the PMU was able to unofficially stimulate action and improve the effectiveness and quality of activities, and the effort and dedication of the project manager and his staff should be recognized as contributing to the successful outputs of this component. 23. However, the WG as mere coordinating body created a number of problems. First, the WG was unable to assert the leadership required to reallocate resources across institutions to accommodate different rates of implementation, and the Bank was not in a position to impose such reallocation. This led to resources being used less effectively. In addition, while strict accounting across institutions was maintained for the purposes of financial management, there was a wide variation in how well the different institutions kept records of their activities, monitored their achievements, and itemized their purchases - the records of none can be said to be adequate. Second as a result of this atomization of activities without any organizing body, the component was deprived of any means of supporting sector-wide activities or of laying the ground for future sector-wide approaches. A notable exception to this is the computerized legal data-base, which was largely managed under the auspices of the Project Management Unit (PMU) (see paragraph 36 below for more details of this activity.) Finally, institutions not represented on the WG were given short shrift. MULEIDE and the new Bar Association were allocated relatively small amounts of money at the beginning of the project, and no new resources thereafter. This again speaks primarily to the weakness in the institutional arrangements, and secondarily to the failure of the Bank during supervision to prevent these institutions being ignored. 24. Implementation arrangements for the Public Administration and Development Management component were also problematic during the first half of the project. Technical responsibility for management was to lie with the Planning Unit in MAE. When this unit was abolished the Ministry delegated management of the component to the National Director for Public Administration (DNFP). However, he was not able to devote the time, nor was it an appropriate role for him, to manage the project which cut across several functional directorates within MAE. In addition, due to the scarcity of qualified staff within the Directorate, no-one was delegated these responsibilities. Day-to-day administrative responsibility was to rest with the DAF, with the project supporting an additional accountant and training for the supplies officer in procurement. At the mid-term review, the Bank and the Ministry agreed to hire a component manager to assume responsibilities for procurement, financial management, coordination and reporting, after which implementation of project activities and the utilization of project resources improved considerably. 25. The lack of any practical link between the two components of the project also led to difficulties in implementation. These included rigidities in reallocating resources as necessary between the two components, and duplication of the local staff needed to manage the project. In  - 9 - addition, since the performance of the two components was different, it inhibited the Bank from adjusting the project as a whole to reflect the performance of the individual components. 26. The IMC was to be established to coordinate the Capacity Building Program in Mozambique, of which this project and its twin were part. It was to be presided over by the Minister of State Administration. Its membership was to include the Ministers of Planning and of Finance (these two ministries were joined after 1994), as well as the Ministers of Education, Justice, and Labor. The Rector of UEM and the Attorney General could also be asked to join (in keeping with the educational focus of the other CB project). The creation of the IMC was problematic in several ways. First, the mandate of the IMC was unclear, appearing at once too narrow and too broad. On the one hand it was devoted exclusively to the coordination of the two Bank projects. On the other, it's mandate was based on the concept of 'capacity building' which is very broad but is not programmatically specific enough, and covers too wide an array of sectors to provide an adequate framework for activities or to warrant coordination at that level. Second, an inter-ministerial committee already existed to preside over reform of the civil service (the Conselho Nacional da AdministraVdo Putblica, CNAP), and the IMC appeared to duplicate its work in some of the areas of its substantive focus. Borrower commitment 27. Borrower commitment to the project is difficult to evaluate in this case. Despite the laudable efforts made by the Bank during preparation of the project to ensure that the project was developed in close partnership with the borrower, a number of factors suggest that the borrower had not, in fact, fully 'bought into' the project. Principal among these, of course, are the lengthy delay between Board approval and effectiveness; the amendments required before project activities took place that effectively reduced the scope of the project; and the very slow disbursement until mid-term, particularly in the Public Sector component. Later experience suggests that, at least in the case of the Public Sector component, while the political level of government, at least initially, understood and was committed to the project this was not the case at the technical level, where there persisted a poor understanding of what the project sought to achieve and how to integrate it into existing activities. In the case of the Legal component, the commitment to the project of the five legal institutions seems to have been compromised by the discomfort they felt over the role of the WG before the first amendment. Moreover, subsequent actions on the part of the MOJ cast doubt on the degree to which it understood its role in, and was committed to the broader objectives of the project. Throughout the project, the MOJ's commitment-both to the project and to a variety of specific reform and capacity- building initiatives remained ambiguous. 28. In summary, quality at entry of the project was unsatisfactory. However, in mitigation of the design's shortcomings, it must be acknowledged that the Bank undertook this project at a time of considerable uncertainty in Mozambique. It reinforced the Bank's nascent relationship with Mozambique early in the process of public sector reform; it established the Bank as one of the institutions that recognized and supported early on the importance of overcoming the capacity challenges that Mozambique faced, particularly in the public sector; and it recognized, unusually both for the Bank and for other donors, the importance of the legal sector to private-sector led and equitable development. Despite its design flaws and implementation difficulties, the project successfully broke ground in two critical areas for donor support, and established important relationships with the borrower on which significant future operations have and will continue to depend.  - 10- 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: 29. When a project's objectives in the SAR are poorly defined, it is difficult to evaluate its success against achieving them. In this case, the lack of concrete objectives combined with the weakness of the implementation arrangements make it difficult to evaluate its success. However, the amendments effectively defined the project's objectives more narrowly and specifically, and against these objectives the project did achieve some notable outputs. As aimed for, it laid the foundation for undertaking a broad reform of the public and legal sectors in Mozambique. On balance, therefore, it is rated satisfactory. 30. Specifically, the project strengthened legal institutions and professional capabilities in the judiciary as well as among advocates, and developed public administration structures and enhanced civil service personnel systems and their management. Notably, 1) legal institutions previously existing in name only are now functioning, recourse to trained judges is now available to citizens in the provincial capitals, and a process of long-term reform of the legal sector has been launched; and 2) the civil service career management system has been substantially reformed and a long-tern public sector reform program is under way. The Bank was among a small number of donors working in these two sectors, and the Borrower agrees that, in the absence of the Bank's credit, much of this progress could not have been achieved. Importantly, the project paved the way for the Bank's future work in these sectors by developing the relationship with MAE and with the legal institutions. 4.2 Outputs by components: 31. The SAR listed a number of outputs for each component. They are listed below, with a description of the project's success in achieving each of them. Strengthening Legal Institutions and Training 32. The Legal component of the project was, on the whole, successful in achieving a majority of its planned outputs. It can take credit for having supported much of the substantial improvements, particularly in staffing, that have taken place in the sector over the last decade. 33. Design and implementation of a long-term strategy and an action plan to strengthen the legal profession and the judiciary in Mozambique. This output was implicitly dropped following the first amendment. 34. Improvement of the quality and efficiency of legal services, as well as trainingfor the lawyers and judiciary. This, and the following two activities and their outputs were substantially achieved. The majority of resources, almost US$5.5 million, or 64 percent of resources were spent on training. This included in service training for lawyers, judges, court officials (such as bailiffs and court stenographers), and other staff involved with the administration of justice (such as prison administrators and prison guards); short study tours abroad, and long-tern training. It also included establishing bilateral links between Mozambican institutions and those of other countries. The impact has been positive on the whole. The following data, by institution, make the point. The Supreme Court, which is responsible for deploying judges in all provincial and district courts, could deploy just eleven judges with legal training in the country outside Maputo, and the dispensation of justice was left largely to community tribunals whose officials had no legal training. Under the project, at least two trained judges are now  - 11 - present in each of the provinces, and a judge has been installed in each district court. In total, some 70 new judges were trained and appointed, and 50 existing judges were provided with appropriate re-training. The courses ranged from three to six months each. The project also provided training to court officials. The Attorney General's Office was established in 1989 and, in 1992 when the project was approved, was constituted by two trained individuals. By the end of the project, the Office was constituted by seven senior attorneys, a qualified attorney was installed in each of the provincial capitals, and training had been provided to attorneys in the districts (although training of attorneys in the districts was not completed.) The project provided training for all of these attorneys, and supported the purchase of office equipment for all provincial offices. After the project, this institution remains the weakest, however, and despite improvements in training, is largely ineffective. The year after the close of this component, the President of the Republic dismissed the Attorney General and all six Assistant Attorneys General on allegations of corruption. Nevertheless, the institution and the physical offices remain, as do the supporting legal staff. * The Administrative Tribunal only existed on paper in 1992. It is responsible for the legality of official expenditures, and therefore provides the audits for all official accounts and approves official expenditures (notably, it approves all new appointments and promotions in the civil service) and acts as ombudsman for the public administration. The Tribunal now comprises 70 trained jurists, as well as appropriate support staff and equipped offices. The Tribunal has a critical role to play in the civil service reform process. While it is still short of adequate staff, and will require substantial reform of its institutional framework, it nevertheless has for at least four years been providing the services required of it by statute. * In the Ministry of Justice the Minister's Office, the National Directorate of Registers and Notaries, the National Directorate of Prisons, and IPAJ, the Institute for Legal Aid, all benefited from the project. In the Minister's Office, senior officials benefited from international seminars; in the National Directorate of Registers and Notaries 186 notaries and registry officials received specialized training and improvements in the registry systems for motor vehicles were advanced; the National Directorate of Prisons received substantial support for training for its staff, all the way from senior level prison guards to prison administrators and departmental officials (a total of 270 individuals). It is important to note that, not only did the majority of officials with legal responsibilities lack adequate basic education levels, there had been no training whatsoever for specialized staff (such as notaries and prison administrators) in Mozambique since independence. In addition, included among the seminars and courses offered were courses in alternative dispute resolution (ADR), aimed at complementing an USAID funded effort to develop legislation for ADR in Mozambique, for commercial law to start off with. This will help expedite resolution of the simpler and more straight-forward conflicts in the commercial arena, and lay the foundation for similar alternatives in other areas of the law. * The Faculty of Law at Edoardo Mondlane University, by far the most important law school in Mozambique (there are now others at the Catholic University campuses in Nampula and Beira), had been dormant for a considerable period since independence, and closed altogether for four years. Opened with a new, modernized, curriculum in 1991, it suffered from a severe shortage of qualified teachers, as well as an almost complete lack of text-books and library books. Under the project, nine tenured faculty were provided with full-time post-graduate training, while an additional 10 were provided with substantial short-term courses, mostly abroad. Due to Mozambique's position as a  - 12- Portuguese speaking country with a Civil Law legal system surrounded by Anglophone countries with a Common Law legal tradition, the law faculty needed to be able to provide training in both traditions in both languages. The Law Faculty, while still suffering from a number of weaknesses (notably due to a lack of adequate compensation for its faculty) now does dispose of an adequately trained faculty, admits classes of between 100 and 150 students, and has increased the number graduating from 8 or 9 in the early 1990s to around 90 per year. The project also significantly increased the supply of textbooks and library books. 35. In addition to the training and seminars provided to the five main institutions, the project also provided three training courses to MULEIDE. The project was also meant to create a Center for Judicial Studies (CEJ), to provide in-service training to legal professional on a continuous basis. The CEJ was created only in name, however. Once again, the main reason for this was the atomization of the project among the five main institutions. 36. Facilitating the dissemination of the law and the administration ofjustice through the design of modern automated legal information systems; upgrading resources and operations of law libraries; assisting the legal reform process through providing consulting services, conducting seminars, and preparing legal materials. The project was substantially successful in achieving this set of outputs. First, the project supported the creation of a legal database of all of Mozambique's laws published in the Boletim da Repzublica (BR), the official legal publication in which laws become effective the moment they are published. Prior to the database, few if any complete sets of the BR were known to exist in Mozambique; a few partial sets were held by various institutions including the Supreme Court and the MOJ. Since this is the only place where the country's laws are published, it was, in effect, possible for only a few people to access the entire corpus of the law. In addition, since the BR had been published on very poor quality paper, particularly during the war, even those copies that did exist were decaying rapidly. With support from the project, the BR was transferred onto a CD disk, first in analog form, and then in readable text form. The data was made available on-line to all legal institutions in the country, including libraries, and is published on the web on a fee based system. This system has the impact of significantly improving access to Mozambique's published laws. 37. Second, the Faculty of Law in particular has benefited from substantial increases in the number of books and textbooks available to it. Where the law faculty had been mostly devoid of textbooks and library books at project inception, students are now provided with at least one copy of each basic text per four students, and the library possesses an adequate collection purchased with credit funds. In addition, the Supreme Court also used credit funds to improve its own library collection. Finally, the project supported the development of a management system for inter-library loans among the various law libraries. 38. Third, the project supported a variety of efforts, particularly on the part of the Law Faculty, to increase understanding and improve access to the law. For, example, in an effort to increase awareness of certain important aspects of the law among local lawyers and legal practitioners in the provinces (including the basics of the legal system, environmental law, and law of international contracts), the Law Faculty sponsored twelve seminars in six provincial capitals; and, based on experience gained through a study trip abroad, the Faculty has established a system whereby law students provide 'legal clinics' gratis to individuals and small businesses. 39. The project design also included resources for a law reform program. The SAR did not specify which laws it would support reforming, or how law reform supported by the project should take place. As a result, the project did not adequately achieve this objective. The project did support preparatory work in reforming the commercial code, the family law, and the penal code. However,  - 13 - work on these activities only began late in the project's life, and these laws are currently being reformed, supported by the Bank, under the new PODE project and by other donors. 40. Assisting the development of the legal institutions, such as the Administrative Tribunal, the Center for Judicial Studies, the Bar Association and MULEIDE. As indicated above, the project provided considerable support for the five main legal institutions, and the Administrative Tribunal and the Attorney General's office in particular, as well as to the Bar Association. This support consisted largely in computer and other office equipment. The CEJ was not created as planned, and while MNULEIDE benefited from seminars and training programs, it was not provided with equipment or technical support. 41. Strengthening the institutions responsible for Project implementation. As documented above, the WG was divested of its responsibility for project implementation and became simply a project coordinating unit. In consequence, beyond resources devoted to staff and to equip the coordination office, no additional resources were allocated to the WG. Resources to support institutional development of the five main institutions, that were responsible for implementing their individual sub- components of the project, have been discussed above. 42. The sustainability of these outputs is mixed, on the one hand, the institutional improvements, from the establishment in practice of the Attorney-General's office and the Administrative Tribunal to the improvements in the Faculty of Law, will be sustainable. On the other, as with most training projects, the sustainability of the improvements in the quality of senior legal personnel is more questionable because of the continuing difficulties the public sector faces in retaining qualified personnel, and the lack of resources for continued training. Public Administration and Development Management 43. The Public Administration component of the project was, on the whole, also successful in contributing to the modernization of Mozambique's public sector. 44. The establishment of a Planning Unit within M4E, staffed by a combination of local staff locally trained advisors andforeign consultants. The PU was abolished shortly after it was created because, as conceived in the SAR, it challenged the mandates of existing directorates. Moreover, the Bank had overestimated MAE's capacity to lead a broader reform of the Mozambican public administration, and the concept of the PU potentially overreached MAE's technical and political capacities to effect change in the broader public sector. On the one hand, MAE was too weak a ministry to represent effectively the Government's views, and, on the other, it had not the authority to lead a cross-sectoral Government activity. 45. The Nucleo de Desenvovimento Administrativo (NDA), created in the stead of the PU with a more limited role, was purely a 'think-tank' to support the Ministry's ongoing activities. Yet, even in this limited role, it failed to complete any but a very small number of the tasks for which it was responsible. The Ministry was never fully convinced of the usefulness of the NDA, and the staff it appointed were weak. In the end, the project's support for the NDA was minimal and most of the resources originally earmarked for the PU were re-allocated. The lesson here is that, even though it is the responsibility of a ministry like MAE to lead reform of the public administration, when such ministries lack political legitimacy and technical capacity the reform process must be led from elsewhere, arguably at the level of the Cabinet, and must be Government-wide. 46. Notable successes, however, that come under the rubric of this activity were one public sector reform workshop and two anti-corruption seminars. The former, which comprised representatives from a wide range of branches of government (legislative as well as administrative,) as well as  - 14- representatives from civil society, was cofinanced by the Governments of the Netherlands and Sweden. It produced a framework which has since been used extensively to help articulate the Government's policy (at Consultative Group Meetings, for example) as well as guide future reforms. This framework informs all public sector reform activities now, and is arguably the single most important intellectual achievement of the component. The anti-corruption seminars addressed the question of corruption in public fora for the first time, and began what is now an ongoing process of finding solutions. 47. Support for the ongoing program of personnel management, now entering the second phase of the program covering the extension of the personnel system to all other sectoral ministries at both the central and provincial levels. The majority of component resources were devoted to this activity and it produced the component's major achievements. The Bank, in cooperation with SIDA who had been involved in the reform of the civil service management system since the 1980s, supported the identification and integration of roughly 105,000 state employees into a unified database; the creation of a new career and remuneration system and the reintegration of all existing civil servants into the new system; the development and promulgation of the necessary legislation to implement the new system; an information campaign to familiarize civil servants with the changes that would affect them; and the design of a computerized network and management system. Importantly, the reform also included the decentralization of most routine personnel management to provincial governments. The main effects of the new management system are fourfold. First, where before the Government had almost no knowledge of who was on its pay rolls, there is now a single, comprehensive, database of all civil servants and their positions with a system in place to insure that it is kept current, ('ghost workers' and other irregularities have been removed). Second, the many individuals inadequately qualified for their positions have been identified and targeted for retraining or other action. Third, the civil service classification structure has been simplified, and career streams introduced, allowing for promotions that reward in-service training, experience and good performance. Fourth, the new remuneration structure is greatly simplified, based on the minimum wage, and includes a significant decompression of wages (the highest non-management wage is about 16 times that of the minimum wage, for example, while the highest line management wage is 21 times the minimum wage). These advances represent an important first stage of a comprehensive reform of the civil service. The length of time taken over it (over 12 years in all) and the substantial amounts of foreign T.A. explains the Government's success in producing this significant output despite the severe weaknesses of the ministry responsible. The Bank's role was to support the decentralization of the management system, the purchase of the equipment and materials needed, the computerization of the system, and the information campaign that accompanied its introduction. 48. As a result of the improvements in personnel management the project achieved, the stage has been set for the next phase of civil service and institutional reforms which promise to promote a more performance oriented public sector. A next stage of the reform would address the fact that wages are still not competitive enough to attract and retain adequate numbers of qualified personnel, especially at the technical levels; the civil service still contains large numbers of under-qualified personnel; the decentralization process, both of the civil service management system and of the public sector more generally has only just begun; and the tools and responsibilities for personnel management in the ministries is still oriented towards establishment control rather than active personnel management. 49. Support to MAE's institutional development, including assistance to MAE's Department of Administration and Finance (DAF); the provision of scholarships; and a contribution to a local fund to support MAE's leadership and coordination role in relation to civil service reform and overall capacity building. DAF benefited from considerable support and was largely transformed from a simple accounting department with hand-written ledgers into a unit equipped with computer databases and staff trained to use them, able to undertake basic financial management and management of the  - 15 - ministry's assets. It was also provided with the experience of managing the financial aspects of a Bank project. 50. Another output from this activity included in-service training of MAE staff, as well as four scholarships for first degrees in France and in Brazil. The training programs were mostly directed at staff in the provincial offices of MAE (the DPACS) and the human resources staff of other ministries (notably education) for the purposes of deconcentrating management of lower level staff to the provinces, and introducing the new personnel management systems. This was practically the only training many of these individuals had ever received, and therefore its impact was disproportionately large, in terms of improvements in performance and of bringing the provincial offices of MAE and other ministries into the reform process. In all some 600 individuals received training. In contrast, the scholarship program had minimal impact, mainly due to very poor implementation. 51. The other unit in MAE that benefited from substantial support in the form of training and equipment was the Administrative Inspectorate, identified as a new strategic priority during the Mid- Term Review. It was supported by the Bank as a means of ensuring that administrative rules and regulations were applied fairly throughout government, and that MAE officials in the field were adequately informed of what the rules were. This was particularly important given that the administrative processes were being reformed and simplified. The project funded training for the MAE inspectors, a course of training for 45 inspectors in other ministries who are under the guidance of MAE for administrative procedural matters, and a series of large seminars in the provinces designed to familiarize provincial staff of administrative procedures. The impact of these efforts, while hard to quantify, were significant: they were almost the only effort on the part of the government to inform provincial staff of administrative procedures and to standardize these procedures. 52. As part of the implementation of the personnel management system, the Ministry with support from the project also created a local cost fund for provincial directorates to support decentralized personnel management. This rotating fund was administered by each provincial directorate to support the minor administrative costs of personnel management. The amounts involved were very small - a balance of $250 per directorate, which was replenished only once or twice during the two years of its existence. However, it was the only fund for which the provincial directorates were given exclusive discretion, and for which they had to practice appropriate accounting and reporting methods. The Ministry reports that this experiment, though small in monetary terms, had a disproportionately important impact in empowering the provincial directorates and building local accounting capacity. 53. On the whole the training programs seem to have had the desired results, and management of lower level staff is now deconcentrated and the new systems introduced. There was clearly scope for more training, but despite continued efforts on the part of the Bank to develop a strategic approach to training, the Ministry never produced one. Similarly, there was no effort to use the resources for bachelors and masters degrees strategically, or to make the selection process methodical. The four candidates were chosen ad hoc by the National Director, and monitoring of performance was extremely lax. One returned to the Ministry and continues to work in the Personnel Department. The remaining three have not yet returned and there is no guarantee they will remain within the public sector. While one could question whether Government borrowing should be used for higher education abroad, and whether the Bank should be associated with such an activity, an opportunity was lost to target these resources to those most likely to make the greatest impact on the public service in general and MAE in particular. 54. Financing of a Management Strengthening Fund to support management improvements within the public service and the development of emerging NGO professional organizations. The management strengthening fund was a missed opportunity. It was never established, and this activity, along with the resources associated to it, were cancelled at mid-term. The Bank overestimated MAE's  - 16 - ability to lead public sector reform in Mozambique, and underestimated the capacity constraints the ministry itself faced. MAE was simply not in a position, (and the Government itself too poorly articulated,) for it to administer a fund for the Government as a whole (let alone including non- governmental entities.) 55. In addition, to the activities listed above, credit funds were to be used to finance consultant services to assist in the monitoring of a salary incentive fund (SIF) and in assessing its impact on civil service staffing and performance. The aim of the SIF was to counteract the effect on the salaries of the most valuable civil servants of the rapid devaluation resulting from the structural adjustment loans. Funds for the salary enhancements were to be provided by the Governments of Norway and Switzerland. Swiss resources were used prior to project effectiveness, while Norwegian funds (circa $3 million), that were conditional on project effectiveness and administered by the Bank, were used thereafter. However, the Government unilaterally expanded the original target of 2,600 civil servants to include approximately 4,400 additional beneficiaries, mostly in the provinces and with four rather than six years' experience. A commonly held explanation for this is that, through internal pressure both bureaucratic and political the Government was not able to resist 'category creep' wherein more civil servants with less critical roles and less formal qualifications were brought under the scheme. Another, posited by some commentators, is that the selection of those with higher degrees and long experience meant de facto skewing the fund to civil servants in Maputo. This was politically not viable at a time when the first multiparty elections were to take place and the ruling party needed votes in the provinces. Bank staff working on preparation of the project report that some MAE officials had, in fact, opposed the scheme, on the basis that it interfered with the development of the new salary framework, and that it was politically too sensitive. The result, in any case, was that neither donor was willing to replenish the fund until it had been restructured and made sustainable. Whatever the explanation for the ballooning numbers of affected civil servants and the consequent failure of the SIF, it is clear that it was premature, and did not take into consideration either the administrative or political realities of Mozambique. 56. This component produced important outputs. Its failure, however, to achieve all the planned outputs and to take a more strategic approach to reform, is due to two factors, in addition to the remaining problems inherent in the design. On the one hand, it is clear that managers at the technical level were only 'bought into' the project in as much as it was useful to support existing activities with which they were engaged. The project was never used to open new avenues for action, nor was there much thought given to using project resources strategically. On the other, it speaks to the weakness of MAE more generally. Technical managers were fully engaged in 'fire-fighting' and had neither time, nor the staff, nor the ability to look beyond immediate needs. In this sense, "there was no capacity to build capacity." This confirms the project may have been premature in wanting to undertake major policy reforms and a major training program. In the event it was the project's 'nuts and bolts' activities that laid the ground for more embracing reform. 4.3 Net Present Value/Economic rate of return: Not applicable. 4.4 Financial rate of return: Not applicable.  - 17- 4.5 Institutional development impact: 57. The institutional development impact of this project is rated modest. This project was exclusively focused on institutional development, and its impacts are detailed above. Given how weak both the legal institutions, MAE and the public sector generally were when the project was developed, the Government made considerable progress in reforming them. Both the Bank and the borrower agree that reforms supported by the project are a first stage in a lengthy process. 5. Major Factors Affecting Implementation and Outcome S. 1 Factors outside the control of government or implementing agency: 58. The Ministry's ability to tackle a large number of challenging issues at the same time is impaired by the severe scarcity of qualified staff within MAE. Throughout the project, Mozambique was going through a major political transformnation, much of which demanded that MAE play an important role. From 1992, when the project was approved, to 1999, when the project closed, Mozambique signed a peace agreement that ended the civil war, held two national elections (in 1994 and 1999) and local elections (in 1998). Each of the elections, as well as the legislation that made them possible and the two electoral censuses, were administered through MAE, and the legislation creating independent municipalities was developed by MAE (with support from the PROL project funded with a Bank credit.) In addition, the Ministry, like the Government, needed to extend rapidly its administration throughout the country, much of which had been inaccessible during the civil war. Finally, rapid transformations in other ministries, such as health and education, also needed to be absorbed and reflected in MAE's activities. The result of these external pressures is that the Ministry could rarely escape a 'crisis management' mode of operation in order to develop a longer, more strategic view. This, among other things, accounts for MAE's preference for pursuing more limited, practical objectives and postponing taking the more encompassing approach to public sector reform the project design had envisioned. 59. A second factor outside the Government's control was the competition for qualified staff from outside the public sector. This came both from the private sector and, especially, from donor agencies and donor funded projects. The years following the 1994 election saw dramatic increase of donor support for Mozambique, with the concomitant need for local staff to manage it. These years also saw a great increase in private sector activity, that also required talented personnel. Together these factors put large new demands on an already very small pool of qualified individuals. The civil service, whose low salaries were further undermined by the sharp devaluation of the Metical as part of the structural adjustment program, had great difficulty retaining trained staff or attracting new ones. In addition, high top-ups and other fees paid to sitting civil servants severely distorted the salaries within the civil service. The effect on the project was twofold. First, it faced the same difficulties as other projects recruiting qualified personnel to manage it. More importantly, the project's positive impacts, of the career and remuneration reforms especially, were reduced in the face of the onslaught of donor project funds and, to a lesser extent, private investment capital. 5.2 Factors generally subject to government control: 60. The success of the project in achieving a significant number of its objectives, particularly in the public administration component, is largely due to the Govemment's commitment to tackling the problem of weak capacity in the public sector, and especially to reforming the civil service management system. Conversely, the fact that the project was not able to go further in developing a long-term strategy for the legal sector, is a reflection of the weakness of the MOJ and the lack of commitment on the part of Government to a concerted reform effort.  - 18- 61. One of the main factors affecting implementation within the control of the Government was the delay in effectiveness. This was in part due to the fact that the Minister of MAE changed just before the project became effective. It resulted in the delays that follow inevitably from a change in management. However, more important was the fact that the Government was not willing to go through with the implementation arrangements which had been agreed to as conditions of effectiveness. This reflects the Government's inexperience with the Bank, its unwillingness to challenge the Bank's proposals, as well as its inability to come up with a viable alternative. It also reflects the Bank's over-estimation of the Government's capacity. 62. Also affecting the project was an unreliable provision of counterpart funds and slow Government approval of contracts. Misunderstandings over the Bank's procurement procedures also contributed to delays. The difficulty of securing the small amounts of counterpart funds needed for the legal training activities in particular, persuaded the Bank to lift the requirement for direct counterpart funds in September 1997, limiting the Government's contribution exclusively to providing dedicated staff and administrative support. 5.3 Factors generally subject to implementing agency control: 63. Both the Legal institutions component and the MAE component were slow to implement to start with, even after the lengthy delay in effectiveness and the subsequent amendments. In the case of MAE, this was due in large measure to the fact that, once the Planning Unit was transformed into the Nzicleo the management of the project fell to the National Director for Public Administration, who was unable to put into place an effective arrangement to manage the implementation details. Following the Mid-Term Review, MAE hired a project manager, and implementation improved considerably. A contributing factor to the weak management of the project prior to hiring the project manager was the unwillingness in MAE to share resources between projects within the same Ministry. Thus, the procurement expert attached to the PROL project, who had been trained in Bank procedures, was not used adequately to support this project component. This speaks to a tendency to balkanize activities between projects and directorates within MAE, inhibiting efficient collaboration and creating a duplication of effort. 64. While accounting and auditing generally met the minimum standards required by the Bank, monitoring and control overall remained very weak, both for MAE and for the Legal Institutions. On the legal component side, the project management unit functioned well, and installed an effective system for tracking expenditures. However, once the project was over, activities had ceased, and the final audit concluded, the data was erased from the hard disc on which they were kept and no detailed record remains. In MAE, records were generally poorly kept. In neither case was any effort devoted to establishing a system to monitor the outcomes of the project's activities, for the purposes of informing the design of other activities or the development of reinforcing links with other ongoing activities. 5.4 Costs andfinancing: 65. The project was implemented within the financial package as agreed in the DCA. 66. As shown in Annex 2, the total cost of the project was US$12.73 equivalent, or 73 percent of the projected project costs of $17.5 million. The shortfall was due mainly to the cancellation of the Management Strengthening Fund sub-component. 67. IDA credit disbursements accounted for just over US$12.3 million, 79 percent of available funds and 97 percent of total project costs. However, as noted in the amendment canceling the 10% counterpart fund requirements for the training components, the total project costs estimated in the SAR  - 19 - do not include all costs. Notably, funding from Swedish SIDA for the new civil service career and remuneration system was not included in the project costs. Other donor funding includes circa US$3 million from the Government of Norway, and 3 million Swiss Francs from the Swiss development Corporation. 68. The life of the project was originally for six years from February 1993 to June 1999. The project was extended for one year. 6. Sustainability 6.1 Rationalefor sustainability rating: 69. The sustainability of this project is considered, on the whole, likely. The overall improvements in the institutions the project has supported are sustainable. In the case of the legal component, the increased number of trained judges and lawyers will likely persist, as the new salary regime has greatly improved remuneration, but are threatened by continued excess demand for qualified staff elsewhere in the economy. The increases in the number of law students to which improvements in the law faculty and in the law library have contributed, will also persist. More importantly, the momentum continues that the project began towards law reform and reform of the legal sector. 70. In the case of Public Administration, the new human resources system is now fully entrenched, both in law and administratively. In addition, the exigencies of the system itself will push further reform, in completing the computerization process for example, and shifting the system beyond simple establishment control and more towards human resources management. Some of those provided training to administer the system have left for more attractive positions outside the public sector, but many have remained. New cadres of technical staff can be trained, moreover, as the training programs were designed and largely delivered by in-house staff. Most importantly, the Government is clearly embarked on a process of public sector reform for which this project helped provide the foundation. This process will likely continue, as public sector reform is now a declared priority for the Government and its development partners. 6.2 Transition arrangement to regular operations: 71. As this is a capacity building project, the issue of transition to regular operations does not arise. 7. Bank and Borrower Performance Bank 7.1 Lending: 72. The Bank's efforts during preparation to address Mozambique's capacity constraints in a flexible and innovative way, and, in particular, to recognize the importance of reform of the legal sector to Mozambique's post war reconstruction and long-term development, all deserve considerable praise. Moreover, that this was done in a difficult political and institutional environment should also be acknowledged. The Bank team was especially foresighted in being among the first to recognize as central themes many of the issues that are now universally recognized as critical concerns today. However, the design of the project, which resulted in an unsatisfactory quality at entry rating overall,  - 20 - was flawed. In general there appears to have been a lack of adequate assessment of the policy and institutional environment for implementation, as preparation tended to focus on technical analyses of problems and assessment of potential solutions with insufficient regard to their practicability given the weak position of key implementing institutions. For these reasons, the Bank's performance during preparation is rated unsatisfactory. 7.2 Supervision: 73. There were two distinct periods during project implementation. During the first, which preceded the mid-term review, the project advanced slowly, and was rated 'unsatisfactory' for one year on implementation progress (IP) and achievement of development objectives (DO). This was largely due to the failure of the MAE component to exhibit any real progress, but also reflected the slow start of activities under the legal component. This rating is the result of failings on the part of both the Bank and the Borrower. On the Bank side, the weaknesses were due mostly to poor communication with the borrower, and a poor recognition of the very high supervision requirements that the project needed. Nevertheless, the Bank should be commended for being willing to amend the project and reduce the scope of the objectives to meet the Borrower's requests and implementation realities. During the second half of the project, performance improved significantly, and the Bank can take some credit for this. Far-sighted and constructive intervention during the mid-term, eliminating the Management Fund sub-component and raising the profile of the project with the Borrower and the Bank, helped. This, combined with better communication and much improved performance on the part of the Borrower, led to the Bank upgrading the project to 'satisfactory.' The improvement in the quality of the relationship between Bank and the implementing agencies was the result of an increased willingness on the part of Bank supervision missions to allow the project to adapt to fit the implementing agencies' evolving needs. In addition, given the very weak management in MAE in particular, the Bank was closely involved in the day-to-day management of the project, visiting the project an average of five times a year during the second half of the project. While a majority of these visits were not formal supervision missions, it allowed the Bank to supplement the implementing agencies' very weak management capacity. However, the Bank was also remiss in allowing very weak management practices to persist, especially on the part of MAE, and for not addressing more vigorously the evident problems in the legal WG arrangement and the MOJ's lack of commitment. On balance, however, Bank supervision performance is rated satisfactory. 7.3 Overall Bank Performance 74. Despite the weaknesses of the project at entry, because of the Bank's efforts to address difficult issues in an innovative fashion, and the satisfactory performance during supervision, the Bank's overall performance is rated satisfactory. Borrower 7.4 Preparation 75. Borrower performance during preparation was mixed. On the one hand, MAE participated fully in discussions concerning Project component identification and displayed flexibility in Project design. The Borrower should also be given credit, particularly MAE, for being willing to consider innovative ways to address Mozambique's capacity constraints. On the other, the MOJ participated with less reliability between approval and effectiveness, and did not attempt to act swiftly in resolving the difficulties surrounding the institutional arrangements; and MAE was short-sighted in supporting untenable implementation arrangements through project preparation. On balance, Borrower performance during identification, preparation and appraisal was unsatisfactory.  - 21 - 7.5 Government implementation performance: 76. Overall, the Government's implementation performance was unsatisfactory. The necessary legislation for both the reforms in MAE and those in the legal sector were made. The Government devoted the necessary time and resources to make the significant budgetary changes that the Career and Remuneration system required, committed itself fully to the framework and strategy that MAE adopted, and promoted it appropriately to its constituents and to the donor community. However, in the case of the legal component, the Government allowed the balkanization of the legal sector to persist, did not take the opportunity to foster more embracing reform of the legal sector, and did not act swiftly to ensure that needed laws were reformed. Moreover, it did not provide necessary counterpart funds in a timely fashion to either component, and delays related to contract approval were also problematic. In the case of MAE, it failed to establish the links between MAE and the other ministries, especially the MPF. 7.6 Implementing Agency. 77. Performance on the part of the implementing agencies of both components was mixed. MAE's performance, particularly in earlier part of the project was unsatisfactory. Performance improved considerably during the second part of the project, and implementation progress improved. However, management practices remained weak, even with the new project manager. 78. Performance of the implementing agencies for the legal component is difficult to assess. The management unit was very effective, and, within the limits of its remit, was able to provide considerable stimulus for and coordination of the diverse activities that the legal sector was using credit funds to support. However, the WG format, in which each institution was the de facto implementing agency for its part of the component, did not work well. The perforrnance of the MOJ, meant to beprimus inter pares among the legal institutions, was unsatisfactory. Administratively, the performance of both the legal implementing agencies and MAE was weak. However, a majority of outputs aimed at in the amended project were achieved. Thus, on balance the implementing agency performance was satisfactory 7.7 Overall Borrower performance: 79. Overall Borrower performance was unsatisfactory. 8. Lessons Learned * Capacity constraints in Mozambique are severe even by African standards. The Bank should therefore guard against overestimating borrower capacity to implement. Specifically, project objectives should be clear, concrete and achievable in the short terrn. Implementation arrangements should be simple. Objectives and outputs should be practical. * Where coordination and leadership in the Government and the implementing agencies is weak, managers at the technical level are unlikely to 'buy into' a project beyond its use in supporting activities with which they are already engaged. Where a project seeks to open new avenues for action, a greater effort is required to build support among technical managers and to strengthen the leadership and coordination in the appropriate agency.  - 22 - * Capacity building, especially when there is so little capacity to start with, is inherently a long term activity. The duration of a normal Bank project can serve to launch and undertake a stage of the process. Capacity building projects should be designed within a framework of long-term Bank commitment (such as is now possible with an APL.) Work on the human resources management system, for example, had begun in 1989, and its success in large part was the result of an unusual level of long-term commitment and patience on the part of both SIDA and the Bank. * Where implementation capacity is especially scarce, projects should ensure that local staff have the necessary skills, and that a dedicated personnel for the project are identified and mobilized. Bank projects should consider developing the necessary implementation skills during project preparation or through an initial period devoted to training those responsible for the management of the project. * When projects are developed in a situation of post-war reconstruction, a clear distinction should be made between those activities that are principally emergency, and those that are meant to lead to long-term reform led by substantial policy change. The latter require longer planning, a higher level of specificity in their objectives, and a broader political and resource commitment on the part of the Government. * In capacity constrained environments where the government is poorly articulated, no single ministry will be able to lead government-wide, multi-sectoral reforms. These will need to be led at cabinet level or the equivalent. Moreover, inter-ministerial committees in a capacity constrained environment created to generate and coordinate policy can only be made effective if there is a widely held recognition of their usefulness at the political level. For this reason, they are likely to be successful only in a limited number of cases. To be successful, they should have a clearly defined mandate, and should be created with the explicit agreement and clear 'buy-in' of the affected ministries. * Coordinated reform of the legal sector can be difficult in environments with weak public institutions and scarce human capacity and resources. Ministries of Justice, who should normally lead the Government's program in the legal sector, are often the weakest among the legal institutions. Moreover, a renewed enthusiasm for the rule of law after a period of Government interference can strengthen individual legal institutions' desire to guard their independence, and consequently weaken the leadership role of a MOJ. Reform of the legal sector, therefore, presents a peculiar challenge. A reform program should, therefore, focus first on establishing an appropriate leadership role for and building capacity within the MOJ or its equivalent for the purposes of the reform. Special attention should be given to designing and following through with a mechanism for coordinating the institutions within the sector behind a unified program. * Salary Incentives Funds, while increasingly important as possible solutions for entrenched public sector uncompetitiveness, are inherently politically sensitive and administratively difficult to implement. They should be applied only in the context of comprehensive civil service management reform program.  - 23 - 9. Partner Comments (a) Borrower/implementing agency: MAE and the Legal Working Group both wrote completion reports, which were discussed during seminars prior to the closing of the respective components. These reports are available in hard copy in the project files. Their findings concur with those in this report. (b) Cofinanciers: (c) Other partners (NGOs/private sector): 10. Additional Information Not Applicable  - 24 - Annex 1. Key Performance Indicators/Log Frame Matrix This project was designed before a Log Frame Matrix became a mandatory part of project preparation. The SAR required that each component lead agency develop, as part of the launch exercise, clear indicators for monitoring progress, closely linked to the quantitative targets for each component. However, the implementing agencies did not comply with this requirement. At mid term, three broad indicators were included in the PSR. These were: 1. Training received by 200 judges, 300 court clerks, 600 administrators; course and legal instruction received by 70 teachers; refresher training received by 120 practicing lawyers. Result: Targets met. 2. An effective career development, training and remuneration system in place and able to atttract quality civil servants. Result: Target met. 3. MAE enabled to take a leadership role in public sector reform. Result: Target met.  - 25 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appraisal Project Cost By Component US$million US$million Strengthening Legal Institutions and Training 7.20 9.05 126 Public Administration and Development 6.50 3.68 57 Management Total Baseline Cost 13.70 12.73 Physical Contingencies 1.80 Price Contingencies 2.00 Total Project Costs 17.50 12.73 Total Financing Required 17.50 12.73 Note: The additional amount spent by the Legal Component over the appraisal estimate is due largely to the appreciation of the SDR to the dollar during the 1990s, and the fact that the unallocated amount was allocated to each component proportionate to their allocation of project resources at mid-term. Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) Procurement Method 1 Expenditure Category ICB NCB Other 2 N.B.F. Total Cost 1. Works 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 2. Goods 0.50 0.00 0.40 0.00 0.90 (0.50) (0.00) (0.40) (0.00) (0.90) 3. Services 0.00 0.00 4.80 0.00 4.80 (0.00) (0.00) (4.30) (0.00) (4.30) 4. Miscellaneous 0.00 0.00 10.50 0.00 10.50 (0.00) (0.00) (10.30) (0.00) (10.30) 5. Miscellaneous 0.00 0.00 1.30 0.00 1.30 (0.00) (0.00) (0.00) (0.00) (0.00) Total 0.50 0.00 17.00 0.00 17.50 (0.50) (0.00) (15.00) (0.00) (15.50)  - 26 - Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) Procurement Method 1 Expenditure Category ICB NCB Other2 N.B.F. Total Cost 1. Works 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 2. Goods 0.72 1.38 1.30 0.00 3.40 (0.72) (1.38) (1.30) (0.00) (3.40) 3. Services 0.05 0.30 1.49 0.00 1.84 (0.05) (0.30) (1.49) (0.00) (1.84) 4. Miscellaneous 0.38 0.00 6.69 0.00 7.07 (0.38) (0.00) (6.23) (0.00) (6.61) 5. Miscellaneous 0.00 0.00 0.42 0.00 0.42 (0.00) (0.00) (0.42) (0.00) (0.42) Total 1.15 1.68 9.90 0.00 12.73 (1.15) (1.68) (9.44) (0.00) (12.27) Notes: 1. The first miscellaneous category is Training; and the second is Operating Expenses for the legal working group. 2. Differences between tables here and data in the Loan database are due to exchange rate fluctuations between the SDR and the Dollar 1/ Figures in parenthesis are the amounts to be financed by the IDA Credit. All costs include contingencies. 2/ Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Project Financing by Project (in US$ million equivalent) Percentage of Appraisal Project Appraisal Estimate Actual/Latest Estimate IDA Govt. CoF. IDA Govt. CoF. IDA Govt. CoF. Legal and Public 15.51 2.00 12.27 0.46 79.1 23.0 Sector Capacity IBuilding Project  - 27 - Annex 3: Economic Costs and Benefits Not applicable  - 28 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle Performance Rating No. of Persons and Specialty Implementation Development Month/Year Count (e.g. 2 Economists, I FMS, etc.) Progress Objective Identification/Preparation Appraisal/Negotiation April 1992 12 ML,2 L,4 ED,4 CB, 1 other Supervision October 1993 2 2 CB S S December 1993 1 CB S S May 1994 1 DC S S June 1994 1 CB S S December 1994 2 CB, ID S S December 1995 2 CB, ID S S February 1996 2 CB, ID S S June 1996 2 CB, ID U U November 1996 5 CD, CB, 2 ID, L U U February 1997 2 2 ID S S October 1997 2 2 ID S S October 1998 2 2 ID S S May 1999 1 ID S S November 1999 1 ID S S ICR April 2000 ID S S Notes: 1. CB= Capacity Building Specialist, CD = Country Director, DC Division Chief, ED Education Specialist, ID = Institutional Development Specialist, L= Legal Specialist. 2. Official missions between 1997 and 2000 were supplenented by frequent visits during which the TL was on other missions. In total, visits during this period averaged 5 per year. (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ (,000) Identification/Preparation 56.2 139.4 Appraisal/Negotiation 22.3 54.3 Supervision 113.2 309.9 ICR 9 25 Total 200.7 528.6 Notes: 1. As this project was prepared jointly with the Human Resources Capacity Building Project, the figures for Identification/Preparation and Appraisal/Negotiation reflect the resources spent on both projects. 2. Recorded staff time on supervision is an underestimate. As noted in the text, during the second half of the project, the task manager was in Mozambique an average of 5 times per year, and on each mission spent time with this project even when not officially on mission for it.  - 29 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Sector Policies M Institutional Development M Social Poverty P'eduction N Gender N Public sector management M  - 30 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending U Supervision S Overall S Bank Lending performance on balance was unsatisfactory, but should be commended for taking innovative approaches to address Mozambique's capacity constraints, for recognizing the importance of legal reform, and for having developed this project in a difficult political and institutional environment. 6.2 Borrower performance Rating Preparation U Government implementation performance U Implementation agency performance S Overall U Borrower performance during preparation was weak, mainly because of its contribution to the poor institutional arrangements. Borrower implementation agency management practices were and remain very weak. The satisfactory rating for implementation agency performance is due to the success, despite these difficulties, in achieving a majority of the desired outputs.  - 31 - Annex 7. List of Supporting Documents Avalia,co Final do Projecto Capacity Building, Componente Administra,ao Publica, June 2000, Ministerio da Administracao Estatal, Direc,ao Nacional da Fun,cao Publica -- the Ministry of State Administration's official completion report. Exercicio Capacity Building, Componente Legal - Avalia,co Final, draft, September 1999, EMCO -- official completion report written by consulting firm contracted by the Legal Component Project Management Unit but not yet officially sanctioned. Legal Capacity Building (Mid-Term Review), November 1996, Susanna Brito -- consultant review for the World Bank's mid-term review mission. Staff Appraisal Report, Capacity Building: Public Sector and legal Institutions Development Project, October 29, 1992, World Bank Staff Appraisal Report, Mozambique Capacity Building Project, March 11, 1992, draft, World Bank Mission Supervision Reports, April 1992 to May 2000, World Bank  - 32 - Additional Annex 8. List of Persons Interviewed Jacomina de Regt -- World Bank Phyllis Pomerantz -- World Bank Dr. Aires de Amaral -- Ministry of Justice Mario Rui -- Director, Project Management Unit, Legal Component H.E. Aiuba Cuereneia -- Vice Minister, MAE Dr. Higino Longomane -- National Director, DNFP, MAE Dr. Mario Da Paz -- Project Coordinator, MAE Louis Helling -- Consultant Lars Tengroth -- Embassy of Sweden The Legal Component Working Group (seminar) MAE representatives (seminar)