This chapter summarizes the lessons that have emerged from the cases studied, and offers recommendations on (1) general approaches to aid coordination that can be adopted for other sectors or cross-cutting issues, and (2) some specific actions, both for other sectors and cross-cutting issues which wish to implement better aid coordination, and for how to further improve aid coordination in the four cases studied. The recommendations on general approach are distilled from key factors that have contributed to the achievements of the cases studied. Some recommendations for specific actions are offered to address some key issues and challenges which were identified and analyzed in the four cases studied, if new partnerships and aid coordination are to advance further.

This study has identified key contributing factors to the achievements of the cases studied (see Table 4-1). These are concerned with (i) key approaches to aid coordination, (ii) institutional arrangements for aid coordination, and (iii) the process of aid coordination. These factors can provide a useful guide when other sectors or cross-cutting issues consider developing new arrangements for aid coordination.

(i) Key approaches to aid coordination

The overall finding from the current study is that one size does not fit all. Our review of mechanisms in Chapter 2 made clear that aid coordination arrangements are diverse among the cases studied. This reflects the fact that aid coordination arrangements have evolved over time to meet local needs and conditions in the respective sectors or cross-cutting issues. However, certain key approaches during this process have been crucial to their success. These are (1) a broadly inclusive, participatory process, and (2) learning by doing.

The four cases studied suggest the critical importance of getting the right process of developing aid coordination for a sector or a cross-cutting issue. The experience of Education and Health, in particular, highlights the importance of inclusive and participatory approaches to aid coordination management. The ministries made clear at the early stage that they had no intention to exclude any interested donors and NGOs and instead encouraged them to participate in the sector-wide process. In Education, this development was a response to the failure of the initial Statement of Intent (SOI) to obtain unanimous endorsement from participating donors. The lesson learned was that more attention should have been paid to obtaining consensus through intensive dialogues with broad participation of stakeholders prior to requesting endorsement of SOI. The ministry’s subsequent efforts to ensure a broadly inclusive, participatory process have created room for discussion and negotiation between the government and donors and among donors themselves. In retrospect, this prepared the ground for the development of a shared vision about sector-wide programs. Seila also took an inclusive approach to local governance and local development in which all stakeholders, including concerned central line ministries, provincial offices of line ministries, and all levels of local administration, were encouraged to participate. TCAP initiated a new program approach for planning and coordinating technical assistance for economic and public finance reform in which a number of donors participate under a common, comprehensive framework for the first time in Cambodia.

The effectiveness of a learning-by-doing approach was most explicitly demonstrated under the Seila program, but the other cases have also taken this approach to some extent. Seila started in 1996 by piloting small scale activities to develop concepts, instruments, and capacity for local governance and development. Its aid coordination mechanisms at the national and provincial levels have evolved through the process of learning by doing. This approach helped to identify what worked and what did not in Cambodia. In Education and Health, learning-by-doing has been practiced in the process of program planning, implementation and monitoring, and program plans and activities have been adapted flexibly to reflect lessons learned in the process.

The two approaches discussed above point underscore the point that there is no single aid coordination arrangement that can be applied to all sectors or all cross-cutting issues, even in the same country. In other words, imposing an arrangement without adequately addressing local needs and conditions unique to the sector or the cross-cutting issue and without broad participation of stakeholders involves high risk of failure, particularly if it is driven by donors and Cambodian ownership is weak or absent. However, an open, participatory process that involves the government, donors, and other stakeholders in developing a shared vision of sector priorities and agreed rules for cooperation can lay a solid foundation for successful coordination, and a flexible attitude towards constantly adapting plans based on what has worked can ensure continued effectiveness.

A key factor for successful aid coordination is the presence of sound formal mechanisms of coordination. Chapter 2 examined five types of institutional arrangements for coordination used in the cases studied: (a) overall coordination among the government, donors and NGOs, (b) coordination within the government, (c) coordination among donors, (d) coordination among NGOs, and (e) core task force or secretariat. In Education SWAp and Health SWiM, formal institutions were established or designated to handle all of the above functions. Of particular interest are donor coordination bodies -- ESWG in Education SWAp and Health Sector Partners Meetings in Health SWiM. Their presence reportedly helped greatly to build consensus among donors by creating opportunities to exchange views, share information, discuss and negotiate issues. This in turn helped reduce transaction costs of the government, since it no longer had to negotiate with each donor bilaterally.

The presence of sound institutional arrangements for aid coordination is clearly a factor for success, but this alone does not appear to be sufficient. Institutional arrangements need to be flexible to accommodate local needs. For example, Seila has created Seila Forum in 2002 to enhance an effective RGC-donor partnership development at the national level, responding to the increasing needs of aid coordination among Seila partners and broader stakeholders in local governance. There are also a variety of partnership arrangements with Seila, as Seila accommodates a wide range of sector or area-specific projects and programs, responding to the preferences and comparative advantages different donors who accept Seila’s reform framework.

It is also important to take into account the differences in implementation arrangements for service delivery among ministries and agencies when institutional arrangements for aid coordination are developed. For example, implementation arrangements for service delivery in Education and Health are considerably top down and centralized, which contrasts with decentralized service delivery systems in Seila. Responding to the above differences, institutional arrangements for aid coordination have been developed in a very different way between Education/Health and Seila. This is one of the reasons why institutional arrangements are necessarily diverse.

While formal institutional arrangements are important, they may need to be complemented by informal networks among the people involved. It was stressed by many interviewees that strong personal networks among donor representatives and advisors helped to disseminate information and coordinate activities. In Health SWiM, personal networks reportedly helped disseminate critical information that was not always shared through formal communication channels. The cases of TB Sub-Sector and TCAP also indicate that personal networks among advisors served as a complementary, effective coordinating mechanism, even though not formally institutionalized. However, informal networks cannot completely substitute for adequate formal coordination institutions.

(iii) Process of aid coordination

The government and development partners often have different views on the program to be developed in a sector or a cross-cutting issue. The views on the program may also vary within government ministries/agencies or among donor agencies. As we studied in Chapter 2, this was clearly the cases in Education and Health where both ministries needed to formulate a common vision of the program at the early stage of program preparation. A contributing factor to the achievement was commendable efforts of the government and development partners to undertake intensive, constructive dialogues among stakeholders, through which the differences of views and opinions were identified and solutions were sought to address them. As mentioned above, this was carried out through an open, transparent, and participatory process. The successful formulation of a common vision has paved the ground for subsequent development of the program in those sectors.

The experience of Education SWAp and Health SWiM suggests that a lot of effort should be made to disseminate the concept of sector-wide programs to broad stakeholders during the early stages of preparation. For example, as was reviewed in Chapter 2, MOEYS held a series of awareness raising and training workshops about sector-wide program concepts for stakeholders, including public officials, donor agency and NGO staff. MOH also organized study tours for public officials to some developing countries that have been undertaking sector-wide programs. Since many government officials (and some donor and NGO staff) were not familiar with the concept of sector-wide programs, this was crucial to laying the basis for subsequent discussion, participation, and ownership.

Information sharing is a critical factor for successful aid coordination management. Chapter 3 reported that information sharing has improved significantly with the introduction of SWAp in Education, and in Health to a lesser extent. It was also reported that institutionalization of information sharing among donors appears to have had an impact on the reduction of overlaps of assistance in the cases studied. Information sharing should be institutionalized as part of formal meetings with government and donors. Also the information system for aid management could be developed and managed at all ministries and agencies receiving foreign assistance, which is discussed in details in Section 2 of this chapter.

In the cases studied, most donors provide assistance through project-type support at present. Some components or projects under a program may be cost-shared or co-financed using trust fund arrangements. Direct budget support to the ministry is found in only one case (Education SWAp). Some donors feel that direct budget support should play a more important role, but many donors are concerned about transparency and accountability of fund management, since the government capacity to manage public finance remains very weak, as reviewed in Chapter 2. Reflecting this reality on the ground, the MOEYS has decided to accept diverse assistance modalities in Education SWAp, and the MOH decided to carry forward Health SWiM without direct budget support. In both cases, the government’s policy that ensures flexibility in assistance modalities enabled donors to seek and enhance complementarities by combining the strengths of different donors. That policy also helped develop partnership arrangements with the participation of all interested donors and thus maximize the resources available for development.

Joint sector or program reviews are a natural extension of aid coordination at the preparation and planning stage. Chapter 2 reported that joint sector or program reviews by the government, donors and NGOs are carried out in Education SWAp, Health SWiM, Seila core component, and TCAP. Also, donors supervise projects or sub-components of a program jointly on a regular basis and their reports are shared with other participating donors (TCAP).

In addition to the more general recommendations concerning approaches, institutional arrangements, and the process of aid coordination discussed above, this study has generated some specific recommendations, both for other sectors and cross-cutting issues which wish to implement better aid coordination, and for how to further improve aid coordination in the four cases studied. Although the government and its partners have made progress in advancing the new partnerships paradigm in the cases studied, there are some key issues that need to be addressed if the new partnerships are to advance further. Those key issues are discussed below and some recommendations for the government and development partners are proposed. Table 4-2 lists all recommended actions for the government and donors to address several key issues.

(i) Recommendations for other sectors and cross-cutting issues

In Education and Health, the government has made significant progress in the development of sector-wide policy and strategy. As was assessed in Chapter 3, policy and strategy development has helped the government and development partners forge a shared vision of sector development that has improved aid coordination, and the comprehensive scope of those programs has enhanced the government’s ownership and capacity to manage both specific programs and aid coordination. The development of SWAp and SWiM has also contributed to providing detailed sector-level information to the recently completed NPRS.

The achievements in Education and Health suggest that similar steps could be taken in other sectors in which sector policy and strategy are weak and coordination of donor assistance is urgently needed. The initial step for the government might be to identify some priority sectors consistent with the Socio-Economic Development Plan (SEDP) II and the NPRS. Immediate candidates could be, for example, agriculture and road transport, which are critical for development in Cambodia.²⁶ Donors and NGOs could assist the government in undertaking this initiative, for example, by providing technical assistance for analytical work, and support for workshops and seminars.

External assistance is much more effective when donors divide up the work according to what they can do best. Chapter 2 in this report revealed that there are a number of examples in which the government and some donors make efforts to combine their strengths of assistance schemes to enhance an overall impact of their assistance. For instance, Seila’s core components are funded by one set of donors who have the ability to provide funding through trust fund arrangements, and these core components then manage and supervise various local investment projects funded by other donors. In TCAP, some donors provide funds for operational costs, whereas the other donors finance and supervise technical assistance through co-financing arrangements. In Health, TB Sub-Sector Program has developed and applied a common TB treatment strategy (called DOTS) for Health Centers nationwide through complementary assistance schemes among donor-funded projects. In all the cases, the strengths of aid schemes of each donor have been combined effectively while recognizing their weaknesses, in order to achieve desired program objectives and outcomes that could not have been achieved without such cooperation.

It is therefore recommended that the government continue the current policy of accepting diverse assistance schemes and aid modalities. It is also recommended that donors actively seek out complementarities among themselves through dialogue and information sharing within the respective sectors or cross-cutting issues.

One of the critical issues that has emerged from our study is the difficulty facing the government in collecting information about external assistance. This has been a major obstacle for the government to manage and coordinate external assistance, and therefore should be addressed as a matter of first priority. The information on past and present donor assistance will help ministries analyze trends in the amount and area focus of assistance and coordinate ministerial activities in day-to-day management. The information on future assistance is required for the planning of future activities in each ministry and agency which receive donor assistance, in particular for the preparation of Public Investment Programme (PIP) and Medium Term Expenditure Framework (MTEF). As we saw in Chapter 3, MOEYS has made major progress in developing a fairly comprehensive database of external assistance (AID Management Information System) with assistance from the CDC. It has also collected comprehensive assistance data for annual joint sector reviews in 2003.

A challenge remains in the compilation of future assistance information because the actual amounts of donor assistance have not always been consistent with the amounts pledged, as some government officials noted. However, donors expressed the view that they have been making the best effort possible to provide accurate data on future assistance, and therefore pledged amounts should be interpreted as working figures which are subject to change.

It is therefore recommended that MOEYS’ good practice be replicated in the other sectors receiving external assistance, under the leadership of respective ministries and agencies. The initial step for the government is to compile a comprehensive record of past and current assistance. Subsequently, the government is recommended to explore ways to compile comprehensive data on projected future assistance, in consultation and collaboration with its development partners. External partners might want to consider supporting this government initiative and ensuring full cooperation in the provision of assistance information at the request of the government ministries and agencies.

Therefore, capacity development of public financial management should be carried out as a matter of urgency. Strengthening public financial management (including provinces and spending units) is a major undertaking and requires systematic, coordinated efforts between the government and development partners. Although MOEYS and MOH have been already undertaking some capacity development activities, the impact would likely be greater if it is coordinated with MEF and potentially with Seila (at provinces and communes). One way to develop new partnerships in this area might be to build on the achievements of TCAP as discussed below.

This recommendation also applies both to developing aid coordination in new areas and to further improving aid coordination in the four cases studied. It is a general consensus that low pay in the civil service has been a major obstacle to implementing the programs studied. Salary supplements have been provided under many donor-funded projects and programs as a temporary remedial measure for the last ten years, even though it has been recognized that salary supplements cause a number of problems in the civil service in terms of accountability, transparency and fairness (see Chapter 2 for more details). As Cambodia has already entered into an era of sustainable development after a long period of emergency relief and rehabilitation, now is the time to take action to address low pay problems through partnerships between the government and donors.

The government has already made important progress with the preparation of the Strategy to Reform Civil Service (SRCS) and a series of recent studies under the initiative of the Council for Administrative Reform (CAR). The average pay levels of civil servants have been raised gradually in the last few years, but further efforts are needed to bring about a fundamental change in the system of incentives facing civil servants. Any recommendations to address the low pay problem and seek an appropriate form of aid coordination require detailed studies, and are therefore beyond the scope of this report. However, the current study points to the urgent need for the government to seek feasible options to address low pay problems, and for donors and NGOs to cooperate with the government in pursuing this initiative and provide technical support where needed. In considering the options, it should be bourne in mind that timely disbursement of salaries to the provincial and district levels must be ensured. At the same time, the efforts of the government to broaden its revenue base need to be continued in order to be able to afford enhanced civil service pay in the future.

A number of interesting suggestions for improvement have been provided to the study team over the course of interviews of government officials. For instance, the officials involved in TCAP have generally appreciated the work of technical advisors, but they suggested that recipients’ views should be considered in the design of future technical assistance. The main points that emerged from interviews are as follows:

This recommendation is concerned with the ways in which cross-sector problems are addressed in partnerships. The delay in PAP disbursements has become a major issue between the government and its development partners. As a result, some donors have had difficulty implementing their programs without the government's matching funds. This is a cross-sectoral problem as MOEYS, MOH and MEF are involved. However, progress has been made in resolving the issue because officials at MOEYS, MOH, and MEF have taken a lead in addressing the problem and development partners have actively supported them. In order to expedite the process, the Fiscal Reform Working Group and Social Development Working Group established the PAP Taskforce and assisted all concerned ministries in identifying the causes of the problem and articulating effective measures to address them. The collaboration of the two Working Groups presents a good example of dynamic partnerships in which working groups under the CG mechanisms work collectively to address cross-sectoral problems.

The roles of existing working groups under the CG mechanisms are being reviewed by the government and its partners. In the review process, the government is recommended to consider developing effective working groups to address cross-sectoral issues such as PAP. It is important to ensure that the process of the review be open, transparent and broadly participatory. Donors might want to participate in the government initiative of the review and consider providing technical inputs when requested.

The key issues and recommendations specific to the respective cases are summarized in Table 4-2. In the following, a priority issue for each case is highlighted. Chapters 6-9 discuss all key issues in greater detail.

MOEYS and development partners have recognized that the number of technical advisors has increased to the extent that MOEYS can not fully grasp the overall picture of assistance and therefore the technical advisors need to be better coordinated. According to a recent survey, the volume of technical advisors assigned to the education sector was estimated as 958 person-months (581 for central departments and 377 for provincial departments; 506 international advisors and 452 national advisors) in 2003.²⁷ This implies that around 80 advisors are working full time in this sector.

Avoiding overlaps of assistance is one issue, but more serious is the role of technical advisors to the ministry. It was reported that advisors’ work has often focused on providing advice to high-level officials and drafting official documents of MOEYS, whereas limited emphasis has been put on developing the capacity of the officials who are actually supposed to prepare the official documents. Although the advisors may have needed to play this role in the past, many officials interviewed strongly felt that the advisors’ task should not be to do the work, but to build the capacity of government officials to do the work. MOEYS officials appreciated technical assistance in the past, but also stressed that MOEYS and donor partners should make concerted efforts to develop capacity of officials who engage in day-to-day management, in particular of their program/project management as well as report writing skills.

As the first step to providing clarity on the status of technical assistance for MOEYS, ESWG is conducting quantitative and qualitative surveys on technical assistance in the education sector.²⁸ It is recommended that, building on the findings of the surveys, MOEYS and development partners discuss and agree on the scope of further work to facilitate better coordination of the provision and use of technical assistance. One action which could be taken immediately is to develop an informal network among technical advisors. As was found in Section 1 in this Chapter, informal networks among advisors have proved useful in TCAP and Health TB sub-sector.

Although the MOH has made major progress in developing the HSP with support of its partners, the HSP’s implementation remains a major challenge. It is critical that the capacity of civil servants keep up with the progress of activities planned under HSP. For instance, the need for capacity development to implement the HSP, particularly to plan and monitor program activities, is increasing rapidly at the sub-national levels, as MOH has devolved these functions to provincial and operational district offices, and they are set to devolve further down to hospitals and health centers in 2004. There are also indications that accounting and financial management capacity at local offices needs to be strengthened. Technical staffs such as midwives are in short supply. Some external partners are already providing technical assistance, and yet the unmet need for capacity building seems to be large.

It is therefore recommended that the MOH identify the capacity development required for the implementation of HSP for all provinces, and that interested donors then consider providing technical support to fill the gaps identified by MOH. It is important to make sure that new projects are aligned under the capacity development needs for HSP, and are complementary to ongoing projects.

Partnerships and aid coordination at the sub-national level are increasingly important as the government implements decentralization and deconcentration policies, and as external assistance will be increasingly directed to the poor in rural Cambodia under the NPRS.

Seila demonstrates a good practice in which provincial level coordination on local development is carried out through District Integration Workshops (DIWs) in all provinces in Cambodia. The DIWs coordinate development activities of line departments, commune councils, and projects funded by donors and NGOs to ensure consistency among them. However, it was reported that the activities of line departments often do not meet Communes’ requests, and that NGOs’ participation in the DIWs is limited in some areas. Therefore, further effort is needed to enhance the DIWs function of aid coordination and to involve a wider range of stakeholders in the development at the provincial level.

While Seila’s DIWs present a good practice of aid coordination at the provincial level, there appears to be a number of other mechanisms of coordination (meetings, forums, etc.) which are independently organized by line departments and other organizations, such as Provincial Coordination Committee (ProCoCom) in Health. There may be a need to review those existing arrangements and consider options to further enhance aid coordination capacity at the provincial level.

Before the existing arrangements are reviewed, however, the roles and functions of provincial and district authorities need to be clearly defined. This would require the adoption of an Organic Law defining the roles and responsibilities of the provincial and district administration, as part of de-concentration policies.

Building on the achievements of Seila DIWs and others, the government might want to review the existing arrangements for aid coordination and consider ways to enhance aid coordination capacity at the sub-national level. Donors and NGOs working in province are recommended to participate in the government’s coordination activities and provide technical support for the preparation of the Organic Law.

In addition, commune councils need further support for capacity building, especially to enhance capacity for accounting and financial management. According to the LAU staff interviewed at Prey Veng Province, the capacity of commune councils varies depending on the communes, but their basic knowledge of local administration and management is generally very low. Commune councilors need more time and support for gaining experience, skills, and knowledge to manage commune council activities by themselves. Also, the FU staff interviewed reported that the capacity of commune clerks was extremely low and they needed more training.

TCAP has brought a new program approach to economic and public finance reform in Cambodia. Through TCAP, the MEF has gained experience and capacity in coordinating a wide range of technical assistance activities in public finance reform. At the beginning of the program, institutional arrangements for fund management turned out to be inefficient and caused some delays in disbursement and procurement. However, the problem has been addressed by making the position of Program Manager into full-time to work for project management. TCAP has produced a number of substantial outputs which have had a significant impact on institutional reforms and capacity development in this area, and laid a solid foundation for future activities.

However, it is well recognized that a lot more need to be done. The IFAPER by the ADB and World Bank has recently developed a comprehensive action plan in which strengthening public financial management is a key component. As pointed out in the previous chapters, the need for strengthening public financial management is urgent in Education, Health and other line ministries, and also in the provinces. Enhancing the performance of new initiatives such as PAPs requires continuing efforts to match capacity development with the changes brought by these new initiatives.

It is therefore recommended that the government pursue the development of a comprehensive program for capacity development of public financial management, building on the achievements of TCAP. This program could include the implementation of capacity building at the sub-national levels, in close collaboration with Education, Health and other priority line ministries, and potentially with Seila. Active participation of all development partners providing assistance in this area will be critical.